RU2326495C2 - Method of communication of messages of any physical nature, for example, method of communication of audio messages and system for its implementation - Google Patents

Abstract

FIELD: physics, radiotechnics.

SUBSTANCE: invention relates to radiotechnics. The technical effect is the determination of a method of communication of messages in channels containing interferences. For that purpose, various options of the method are proposed, for active noise reduction, at an example of the communication of and reduction of noise in audio signals. The invention describes a process of amplification of electric signals, for example, in power amplifiers, as well as a multi-band threshold noise suppressor. In electric signal amplification, the traditional feedback is designed as a multi-function assembly, a signal processing unit. This unit can be designed, for example, using a computer and allows to ensure the required electric parameters of the feedback circuit and to generate a number of new control signals for the amplifier operation mode optimisation.

EFFECT: optimisation of signal transmission in channels with interferences.

36 cl, 30 dwg

Description

Technical field

The invention relates to cybernetics and can be used, for example, in radio engineering.

State of the art

In science and technology there are a large number of different methods and methods for transmitting messages of various physical nature. They are based on filtering and processing signals in the frequency, time, and spatial domains through the use of the polarization properties of signals, their coding, etc.

As an example of a scientific approach to solving the problem of increasing the accuracy of transmitted messages, one can note the work of V.P. Bakalov "optimal spatial-temporal filtering" published in the journal "Radio Engineering" (1984, No. 12, pp. 3-8). This article discusses the linear optimal spatio-temporal filtering of random stationary and non-stationary processes and fields. The optimal transfer functions of the compensation channels are determined that minimize the standard deviation (RMS) of the signals and maximize the signal-to-noise ratio.

The article shows that for an information system in which there is one or more sources of message signals with coordinates that do not vary in space, as well as one or more sources of interference, the coordinates of which are also unchanged, it is possible to find structural schemes of optimal filters within the framework of the problem statement model, giving an additional gain in noise immunity compared to conventional Wiener filters. The most indicative in this sense is the case when the spectral density of the interference is high compared with the spectral density of the message signal. The found structural schemes are an example of a scientific formal-symbolic solution to this problem within the framework of the model of its formulation. At the same time, the question of the practical synthesis of an optimal filter with a complex transfer characteristic, which is a complex frequency function obtained as a result of various mathematical operations on a number of other complex frequency functions: transfer functions by signal and noise, estimates of the channel transfer functions from the emission points of the signals, and interference to measuring points (reception), spectral density of the signal and interference. Obtaining these functions requires a special, preliminary study of the information system.

The main disadvantage of this method is the narrowness of the practical areas of its use and the high complexity in the implementation of such filters. Indeed, a filter synthesized for one time interval will no longer be optimal for another time interval if, for example, a relative movement of the location points of message sources and interference sources occurs in the information system, which takes place in the vast majority of real applied problems of information theory. In addition, the currently known methods for the synthesis of filters with the desired transfer characteristic for a given type of amplitude-frequency characteristic (AFC) and phase or time-frequency characteristics (PFC or VFC) are reduced to experimental methods of iterative selection of various combinations of low-pass filters (LPF) ), high-pass filters (HPF) of a certain order, or a combination of band-pass filters (PF) of different quality factors. Since this method does not provide an unambiguous technical solution to the problem described through a set of well-known nodes, blocks and connections between them, and the problem is solved in the form of formal-symbolic, scientific categories (formulas), it is not a patentable technical solution.

Given these circumstances, the model of the information system should be complicated. Unlike the system model described above, it is necessary to complicate it and solve the problem of optimal filtering of non-stationary, non-ergodic processes, fields, signals, or, in another way, the task of transmitting messages of any physical nature in channels with changing parameters and random noise, when the location of signal and interference transmitters , as well as message receiving points, may randomly change. This task can be further complicated if we do not consider the actual filtering of electrical signals in the optimal filter, but consider all the actions and transformations on the signals (of any physical nature) by their transmission from one point in space and time to another point - the point of receipt or reception of a message of the same physical nature. It is clear that a similarly formulated problem can be considered as the central task of the theory of information, since its solution can give an answer to the question: how can one minimize information loss or, in another way, increase the entropy of any real communication system.

In the framework of the above statement, this problem was solved in the invention, “A method for optimal transmission of messages of any physical nature, for example, a method for optimal sound reproduction and a system for its implementation. A method for optimal spatial active lowering of signals of any physical nature” (RU, A, 2145446). In the inventions RU, A, 2106075; RU, A, 2106074; RU, A, 2106075 describes the technical solutions present in the method of optimal messaging as part of it.

The optimal messaging method described in this invention is based on the basic principles of cybernetics. The system for its implementation is built according to the scheme with feedback and special processing of the transmitted and received messages. The processing of these signals is carried out by the multiband method, in which the energy and time parameters of the components of the signals of messages and interference falling into the corresponding signal analysis bands are analyzed in real time. As a result of this processing, it is possible directly in the process of transmitting messages to carry out optimal predistortions of the signals of the transmitted message and generate signals for actively reducing noise at the point of receipt of the message. Moreover, all actions are fully automated and do not require the intervention of a person who, for example, as in traditional sound reproduction, performs similar actions to compensate for the distorting properties of the communication channel using the equalizer and corrector of the frequency-time delays of the signals by ear or using the corresponding measuring complex.

Thus, this method of messaging allows you to automatically calculate the entropy of the messaging system all the time and compensate, as much as possible, the growth of entropy. The growth of entropy in the systems is “slowed down” due to an external energy source, the energy of which is used to generate noise reducing signals, energy pre-emphasis of message signals, signal processing and a fundamentally new structural diagram of a message transmission system containing new nodes: a sounding device (for example, a microphone) installed near the point of receipt of the message, the signal processing unit and a communication line for transmitting feedback signals to the signal processing unit. It is important to note that the location of the probing device near the point of receipt of the message is due, on the one hand, to the implementation of the above functionality of the processing unit, and, on the other hand, to the extremely broad statement of the problem, in which there is no information about the form, properties of the communication channel, a priori information about the characteristics interference and signals, their locations, information about their possible relative movements. Therefore, the use of a single sounding device logically follows due to the lack of a priori information, primarily about the communication channel. If the receiving element of the probing device is removed from the point of receipt of the message or using an additional receiving element at random placing it in space, then you can find yourself outside the communication channel or transmitting messages, for example, outside the passenger compartment of the vehicle in which audio messages are transmitted. Therefore, the method is optimal from a structurally functional point of view only within the framework of the formulation of the objectives of this invention. When changing the statement of the problem, this method can be improved by introducing new actions (essential features). For example, in the invention RU, A, 2211491, the problem was solved of increasing the accuracy of transmitting messages of any physical nature in channels with slightly changing parameters of remoteness of the emission points of message signals from their receiving points in the presence of interference. This additional information about the properties of the communication system made it possible to find a new technical solution that can improve the efficiency of noise suppression. But the presence of a delay time of signals for actively reducing noise and interference did not allow fully synchronizing the time of arrival of interference at the point of receipt of the message and signals for their active suppression. In addition, studies have shown that the effect of suppressing interference and noise in the methods described in patents RU, A, 2145446 and RU, A, 2211491 is limited not only by the above delays, but also by the level of isolation of the signals received by the probing device near the point of receipt of the message, and signals for active noise reduction emitted at the same point in space. With such a structural scheme, the system is prone to self-excitation. The turbulence spectrum of such a system depends on a number of parameters: the frequency of the suppressed interference, the transmission rate of the message in the channel, the delay of the signals for active noise reduction and resembles the beating between noise signals and signals for their active suppression. As a result of this, it is necessary to limit the dynamic range of the signals for active noise reduction, which leads to an insignificant (on the order of decibel units) effect of noise reduction. It will be shown below that if you change the above statement of the problem of increasing the accuracy of the transmitted message in the interference channels, then you can improve the known methods of message transfer and noise reduction methods known from the prior art.

Since the above-described methods for transmitting messages as part of them contain methods for reducing noise, it is advisable to make a brief overview of known from the prior art methods for reducing noise.

A known method of suppressing noise mainly aircraft (JP, A, 52-15721). The method consists in receiving and converting sound noise signals of an aircraft engine (aircraft) into electrical signals, transmitting the received signals to the place of their processing, processing these signals by means of a processing unit, to the input of which these signals are supplied, made with the possibility of energy and temporal predistortion of the received signals in the formation of its output electrical signals intended for active reduction of noise signals, their conversion into sound signals, the place p memory location signal reception point is closer to the noise source (the engine) than the area of the noise reduction space. This area is on the ground.

The energy and temporary predistortions of the received noise signals are carried out by selecting the levels and phases of the components of the received electric signal and storing (fixing) the energy and temporary predistortions that provide the effect of noise reduction in the above region of space. Structurally, the processing unit contains a transistor, on the emitter and collector of which antiphase signals are generated, as well as a node in which these signals form an output signal, the phase and level of which provides a noise reduction effect. In an embodiment, amplifiers and loudspeakers are used for multichannel noise reduction in all directions from the aircraft. The spatial separation of the receiving microphone and loudspeaker provides a sufficiently high level of isolation between the received and emitted signals. This allows you to generate signals for noise reduction, the level of which is comparable to the level of noise signals. As a result, the efficiency of reducing noise signals, for example, at the frequencies of their highest intensity, is quite high.

The disadvantage of this system is the narrow-band nature of the noise reduction effect. The reason for this is the difference in the spectra of the noise signal of the noise source — the engine at different points in the region of space on the ground — from the spectrum of the noise signal at the point of reception, since the noise source is a spatially distributed set of various noisy assemblies, parts and assemblies of the engine, which is, for example, a barrel an assembly with a diameter of about 1-2 m and a length of about 2-5 m, suspended from the wing of an aircraft. The components of sound noise signals from various engine nodes come to the area of space on the ground and to the microphone location point for receiving noise signals on an airplane with different levels and phases. Therefore, using the processing unit described above, it is possible to select the levels and phases of the components of the noise signals only in a narrow frequency band. For other noise components, energy and temporal predistortions are not optimal. Instead of the effect of noise reduction at these frequencies, on the contrary, the effect of amplification of the noise level can be observed. For this reason, the option of reducing the noise level in all directions from the aircraft is not physically possible. As a result, the overall level of noise reduction in such systems is low. This, apparently, explains the lack of implementation of the method of reducing noise in air transport.

It is more effective to achieve a noise-reducing effect if one uses the methods of multi-band processing of noise signals, which are described in detail in the above inventions (BU, A, 2145446 and RU, A, 2211491). In these inventions it is shown that by multi-band filtering the signals (in this case, noise), it is possible by experimental selection of the levels and phases (delay time) of the components of these signals falling into these frequency bands to a certain degree independently, to generate the required frequency-energy and frequency -Temporal predistortions of these signals, for example, for generating signals of active noise reduction. In other words, the formation of the frequency response and phase response of the filter, providing the greatest effect of noise reduction in the desired region of space. It will be shown below that by appropriately finalizing these inventions, one can get rid of their shortcomings and increase the accuracy of message transmission due to more effective active suppression of interference and noise.

In a certain sense, a similar problem - improving the accuracy of messaging, has long been solved in the field of radio engineering such as the design of power amplifiers, for example, low frequencies.

There are a large number of technical solutions aimed at increasing the linearity of amplifiers of electrical signals, for example, by means of a general negative feedback, which Harold Black invented and patented in 1927.

The basis of this technical solution is the principle of summing the signal from the output of the amplification stage or the entire amplifier, if it is made in the form of several series-connected amplification stages, and an input signal in antiphase and applying the sum of these signals to the amplifier input. This technical solution allows to increase the linearity of the amplifier by reducing the products of nonlinear and intermodulation distortion occurring in the amplifier of electrical signals. This process is largely similar to the process of actively reducing noise, but not in space, for example, somewhere outside the nodes and blocks of the noise reduction system, but directly inside the amplifier. A disadvantage of the known technical solutions for increasing the linearity of amplifiers can be considered the low accuracy of the formation of the transfer characteristic of the feedback circuit using traditional solutions.

As a prototype of a method for transmitting messages of any physical nature in this invention, a device for reproducing audio signals in the presence of noise and noise San Francisco CD 70 of Blaupunkt firm (Avtozvuk magazine No. 12, December 2001, pp. 103-115) was selected. This is a CD player and a radio, structurally combined with a multifunctional signal processing unit. You can connect a measuring microphone and a number of peripheral devices (additional amplifiers, loudspeakers) to this device. The measuring microphone is designed to solve two problems.

“Firstly, this is an automatic volume control depending on background noise, actually - on the speed of movement. The system does not respond to short-term bursts - such time constants are already there. But it responds to a predictable smooth increase in noise ... promptly and accurately. Second San Francisco CD 70 has a seven-band equalizer, and there are three presets for it: EQ1, EQ2, EQ3. Using a microphone, you can automatically equalize the frequency response with sound pressure at different fillings ... Cor The section is set independently for the L and R channels (!), and if that ... manually ... In the sound path, in addition to the equalizer, there is also a usual bass and treble tone control ... Built-in crossover - low-pass filter for subwoofer output and high-pass filter for the rest ... Loudness here also has five gradations of processing depth ... we managed to get rid of low vibration resistance. " This player is equipped with a more advanced anti-vibration system for car CD players - anti-shock. This device implements the following method of transmitting audio messages.

Message signals are converted to electrical signals from a message source using a CD player or radio receiver. These signals are fed to one of the inputs of the signal processing unit, which is configured to carry out energy (by means of a volume or volume control or by delaying the message signals in the anti-vibration system) predistortions of the electrical signals of the message source when generating output signals. signals of the processing unit. These signals are amplified, converted into signals of the same physical nature (into sound signals) and radiated into the channel for transmitting messages to the area of the space for their receipt - in the car interior. Noises present in this car do not suppress, but increase the volume level in the car interior in proportion to the average noise level. As a result of these actions, the useful information signal of the message is modulated by an average noise level, i.e. acquires a new specific form of distortion. This is not only noticeable by ear and annoys the listener (driver) with a “floating” volume level, but also leads to increased fatigue of the listener, who, out of his desire, has to listen to message signals at an increased volume level together with a high noise level. The use of such a system dramatically increases the accident rate and is in no way consistent with the well-known patented provisions of the modern information theory. Such methods of increasing noise immunity can be applied in radio communications, where a message is transmitted by means of a high-frequency message carrier - carrier, increasing the level of which, it is possible to increase the signal-to-noise ratio (noise) at the point of reception of the radio signal. But at the same time, the growth of the radio signal is compensated in the radio receiver, for example, using the AGC system, as well as as a result of the specific features of the detection of radio signals with different types of modulation. In sound engineering, this approach is not applicable. There are no message carriers here. A process of a specific physical nature transmitted from one point in space and time to another point (a signal, a field, etc. - these are all synonyms), for example, an audio signal, is precisely the message itself. Modern scientific provisions on the principles of message transmission are well known and described in detail, for example, in patent RU, A, 2145446. This source of information is strictly scientifically proven that when implementing optimal methods of transmitting messages in the presence of interference, it is necessary to extremely accurately repeat the signal waveform at a point its receipt, and noise and interference must be suppressed. Only by implementing this approach, it is possible to create systems of the highest form of organization, quality class, complexity group and really improve, for example, consumer properties of sound-reproducing equipment. Below we will show how this approach can be implemented if it is possible to obtain a priori information about the message channel.

Disclosure of invention

The present invention is based on the task of creating such methods and systems for their implementation, which can improve the accuracy of the transmission of messages of any physical nature in the interference channels, if it is known in advance where the sources of interference are located in relation to the area of the message receiving space, what is their spectrum and what are the size and shape of the area of the space for receiving the message, and thus, it is more efficient to reduce the level of interference and noise, and to increase the signal-to-noise ratio (interference) when transmitting messages to any physical tion of nature. An additional technical result is to increase the signal-to-doom ratio of the electrical signals of the message source and to increase the linearity and noise immunity of the amplifiers of the electrical signals.

The problem is solved in that in the known method of transmitting messages of any physical nature, which consists in converting messages into electrical signals of the message source, transferring them to the processing site, processing the source signals through the processing unit, to the second input of which these signals are supplied, and made the implementation of energy and / or temporary pre-emphasis of the electrical signals of the message source during the formation of the output signals of the processing unit, their amplification, conversion into signals of the same physical nature and radiation into the message channel to the area of the space for their receipt, in which there are interference or noise of the same physical nature. According to the invention, noise signals of any physical nature are received, converted into received noise electrical signals, transmitted to a processing site, processed by these signals by a processing unit provided with a first input to which these signals are supplied, and configured to perform energy and time predistortions received noise signals in the formation of the output electrical signals of the processing unit, designed to actively reduce noise in the field of floor space The message is amplified, converted into signals of the same physical nature and emitted to the area of the message receiving space, and the place of receiving noise signals is chosen closer to the noise source than the area of the message receiving space, and the energy and time predistortions of the received noise electrical signals are carried out by selecting levels and phases of their components at a zero level of electrical signals of a message source or at a level of electrical signals of a message source below a predetermined Orogen, storing energy and temporal de-emphasis, providing the effect of reducing the noise level in the surrounding area receive the message.

Possible implementations of the method, such that:

electrical signals for active noise reduction are generated in a limited frequency band of the low-frequency part of the noise spectrum of the received noise signal to increase the size of the region of the noise reduction space;

energy pre-emphasis of the electrical signals of the message source is carried out by adjusting the signal level of the message source, energy and time by light-compensated control of the signal source, and temporary by delaying the signal of the message source, for example, in the anti-vibration system of an automobile CD player;

to adjust the spectrum of message signals taking into account possible distortions of transmitted messages in the channel, non-automatic frequency-energy and / or time-frequency predistortions of the electrical signals of the message source are carried out by means of a processing unit configured to multiband process the electrical signals of the message source when generating its output signals;

frequency-energy predistortions are carried out by means of an equalizer, frequency-time predistortions are performed using frequency-time or frequency-phase correctors, and frequency-energy and frequency-time predistortions are carried out by means of timbral blocks or fixed presets of amplitude-frequency and phase-frequency characteristics in predistortion filters;

to implement the mode of lowering the level of various noise components in the entire above region of the noise spectrum, taking into account possible almost constant differences in the spectrum of noise signals in the region of the message receiving space and the spectrum of the received noise signal, non-automatic frequency-energy and time-frequency predistortions of the received noise electrical signals are carried out by the processing unit made with the possibility of multi-band processing of received noise signals in the formation of its output signals;

in the area of the message receiving space, additional reception of signals of any physical nature is carried out, which is a message signal plus noise, the spectrum of which can change with time with respect to the spectrum of the received noise signal, not only due to current changes in the spectrum of the transmitted message, but also due to changes in the spectrum itself noise in the field of the space for receiving messages in relation to the spectrum of the received noise signal, conversion of an additionally received signal of any physical nature into electrical feedback signals, their transmission to the place of processing, their processing by means of a processing unit made with an additional input to which these signals are supplied, and configured to automatically generate energy and time or frequency-energy and / or frequency-time pre-emphasis received noise signals of any physical nature by introducing nodes into the processing unit that automate actions to select the levels and / or phases of the components of the generated signals for su- denoising and memorization and action for forming energy and / or time of the predistortion signals, providing the effect of lowering the noise level in the space receiving the message;

in the area of the message receiving space, additional reception of signals of any physical nature is carried out, which is a message signal plus noise, the spectrum of which can vary with respect to the spectrum of the source signals not only due to current changes in the spectrum of the transmitted message and random noise or noise, but also due to random changes in the transfer function of the message channel, the conversion of an additionally received signal of any physical nature into electrical feedback signals, their transmission to the place of processing, processing by means of a processing unit equipped with an additional input to which these signals are supplied, and configured to automatically generate frequency-energy and / or time-frequency predistortions of the electrical signals of the message source by introducing nodes into the processing unit that automate actions by selection of levels and / or phases of multiband components of the signal source of the message source;

to implement the mode of lowering the level of various noise components in the entire above region of the noise spectrum, taking into account the almost constant differences in the spectrum of noise signals in the region of the space for receiving messages in relation to the spectrum of the received noise signal, frequency-frequency and time-frequency predistortions of the received noise electrical signals are carried out by the processing unit made with the possibility of multi-band non-automatic processing of received noise electrical signals when forming any output signals of the processing unit;

to implement the mode of lowering the level of various noise components in the entire above region of the noise spectrum, taking into account possible random changes in time of the spectrum of noise signals in the region of the space for receiving messages in relation to the spectrum of the received noise signal, automatic frequency-energy and time-frequency predistortions of the received noise electrical signals are carried out by means of a processing unit configured to perform these predistortions only at those times when and the signal source signal level is lower than the predetermined one, for example, before message transmission or in pauses during message transmission, by introducing nodes into the processing unit that automate actions to select the levels and phases of multiband signals to actively reduce noise at these times and memorize them and the formation of the stored values of the pre-emphasis during the transmission of signals of the message source, when the signal level of the message source is higher than the predetermined;

the processing unit is configured to summarize the processed signals of the message source and the processed received noise signals of any physical nature by an adder to thereby generate its output signals and implement a single-channel method of transmitting messages of any physical nature;

the processing unit is configured to summarize the multi-band processed signals of the message source and the multi-band processed received noise signals of any physical nature by at least two adders to thereby generate its output signals and implement a multi-band method for transmitting messages of any physical nature;

the conversion of messages into electrical signals of the message source is carried out in a multichannel way to implement a multichannel method of transmitting messages to the corresponding number of channels in the message receiving space region, in each of these channels one of the described variants of the message transmission method is carried out, while the received noise signal is common to these channels and served on additional inputs of processing units, the number of which is equal to the number of message channels;

the conversion of messages into electrical signals of the message source is carried out in a multichannel way to implement a multichannel method of transmitting messages to the corresponding number of channels in the area of the message receiving space, in each of these channels one of the described variants of the message transmission method is carried out, while the number of received noise signals is equal to the number of message transmission channels , and they are fed to the additional inputs of the processing units, the number of which is equal to the number of message channels;

each output signal of the processing unit is amplified by an electric signal amplifier and radiated to a region or regions of a message receiving space of any physical nature;

the locations of the processing unit and the signal source of the messages do not spatially coincide for the possibility of their block-by-block implementation and configuration of the message transmission system with blocks with the above functionality, or their locations are nearby for their constructive implementation in one device;

sound messages are used as messages of any physical nature.

The object of the invention is solved in that in the known method of transmitting messages of any physical nature, which consists in converting messages into electrical signals of a message source, in amplifying these signals by means of an amplifier made in the form of one or more amplification stages, supplying amplified electrical signals of a message source to a load , the formation of electrical feedback signals, their supply to at least one of the amplification stages, according to the invention, at least one of the generated feedback channels are subjected to multiband processing by a signal processing unit that is supplied to its first input; in the processing unit, feedback signals are multiband filtered to form at least two channels of multiband processing of feedback signal components, each multiband generated channel is configured to perform energy and time predistortion of feedback signal components to improve the accuracy of the transmitted message by realizing the possibility of more its exact formation of the required transfer function of the feedback circuit, multi-band processed components of the feedback signals are used to generate the output signals of the signal processing unit, which are supplied to the aforementioned at least one of the amplification stages of the electrical message signals.

Possible implementations of the method, such that:

in the signal processing unit, the multi-band processed components of the feedback signals are added to form the output signals of the processing unit;

in the processing unit, the multi-band processed feedback signals are additionally filtered to reduce non-linear and intermodulation distortion of the components of the feedback signals in the processing bands, the multi-band processed and additionally filtered components of the feedback signals are summed, the signals thus obtained are output signals of the processing unit;

feedback signals are negative or positive feedback signals;

amplification of the electrical signals of the message source is carried out by multistage amplification with combined negative and positive feedback, and at least one of the feedbacks is subjected to multiband processing by the signal processing unit in accordance with the sequence of the above operations, if several generated feedback signals are subjected to multiband processing, use the appropriate number of signal processing units, the first inputs of which are fed accordingly generated feedback signals, and the corresponding output signals of the processing units are supplied to the corresponding amplification stages of the electrical signals of the message source;

a power amplifier is used as an amplifier, the amplifier load is a signal emitter designed to convert electrical amplified message signals to message signals of any physical nature, and electrical feedback signals are generated from the amplifier output, or from the emitter input, or by additional signal conversion of any physical nature in the immediate vicinity of the emitter into electrical feedback signals;

energy and temporary pre-emphasis of the components of the electrical feedback signals is carried out by preliminary non-automatic selection of adjustments of the levels and phases of the multi-band processed feedback signals from the tuning messages of a convenient form for this, for example, noise-like signals or sinusoidal signals of various frequencies according to a criterion, for example, the minimum standard deviation signals of the transmitted message;

energy and temporary predistortions of the components of feedback signals are carried out automatically according to the signals of the message source, for example, noise-like shape or actually according to the transmitted message signals by means of a signal processing unit, configured to automate the above actions and equipped with a second input to which the electrical signals of the message source from the amplifier input or the corresponding amplification stage of the electrical signals of the message source covered by the inverse communication, while in the processing unit, multiband filtering of the electrical signals of the message source is carried out by means of filters with characteristics similar to filters, designed for multiband filtering of feedback signals, the joint processing of feedback signals and messages filtered in the corresponding bands, generating control signals allowing automate the processes of energy and temporary predistortion of feedback signals;

the signal processing unit is performed with a second input, to which the electrical signals of the message source or amplified in the previous stages electric signals of the message source are fed, in the processing unit, multiband filtering of the electrical signals of the message source is carried out by means of filters with characteristics similar to filters designed for multiband filtering of the feedback signals at the same time, at least two channels of multiband processing of signal components of the sources are formed and messages, each multiband channel formed is capable of performing energy and temporal pre-emphasis of message source signal components to improve the accuracy of the transmitted message by compensating for amplitude-frequency and / or phase-frequency distortion of message signals in a portion of a message transmission circuit covered by feedback, for example, in a cascade amplifications, in an amplifier, or in an amplifier connected to a load, multiband-processed signal source signal components summarize with the multi-band processed components of the feedback signals, the signals obtained in this way are the output signals of the processing unit, which are fed to the input of the amplifier or the above stage, covered by feedback;

energy and temporary pre-emphasis of the components of the electrical signals of the message source is carried out automatically in the processing unit by jointly processing the multi-band formed components of the feedback signals and message signals, generating control signals that automate the processes of energy and / or temporary pre-emphasis of the components of the message signals;

in the process of transmitting messages in the processing unit, the analysis of the levels of the components of the message signals in the processing bands is carried out and control signals are generated by which form the previously adjusted adjustments of the levels of the components of the signal source of the messages and / or feedback signals from the time when the level of the components of the message signals in the corresponding strip processing above a predetermined threshold level, at other times, the levels of the generated components of the signal source messages and / or feedback signals in the processing unit are automatically changed to reduce the transmission coefficient of the components of the message signals in this frequency band and to reduce due to this level the components of the products of nonlinear and intermodulation distortion, to reduce noise, interference and interference;

To synchronize in time the processes of automatic multiband threshold adjustment of the levels of the components of the signal signals of the message source with the moment of exceeding the level of the components of the signal signals in advance of a predetermined threshold during the implementation of temporary pre-emphasis of the components of the signal signals of the message source, they are additionally delayed by the amount of time required to generate control signals in the processing unit providing threshold adjustment of signal component levels at output and signal processing;

in the process of transmitting messages of any physical nature, in the signal source signal processing bands, the current levels of message signal components are analyzed with respect to at least two thresholds and, depending on the current level of message signal components, various transmission coefficients and the form of the amplitude characteristic in these message signal processing bands are generated;

a two-threshold analysis of the levels of message signal components is used, and message transmission coefficients in the corresponding frequency bands are generated according to the following algorithm: if the message signal level is lower than the first threshold, which can be pre-set, for example, at a level of 20–70 dB relative to the maximum undistorted amplified message signal , then the transmission coefficient in the corresponding signal processing band of the message source and / or feedback signals is chosen so as to reduce the gain a cascade or amplifier covered by feedback, and implement a bandpass threshold lowering level of amplified message signals if the message signal level in this frequency band exceeds the first threshold level, but below the second threshold level that the coins must be set in advance, for example, at - 10 ÷ 2 dB relative to the maximum undistorted amplified message signal, the transmission coefficient in this signal processing band is set with the previously selected energy and time parameters predistortions, moreover, the energy and temporal predistortions are constant and do not depend on the level of message signal components for implementing a linear amplitude response characteristic of message signals, if the level of message signals is higher than the second threshold, then variable energy predistortions of signals are generated to implement the compression mode of the dynamic range of amplified signals and formation non-linear amplitude characteristics of the corresponding band-pass processing channel, for example, a logarithmic ith form and reduction due to this increase in non-linear distortions associated with a possible sharp limitation of the amplitude of the amplified message signal;

depending on the ratio of the current level of message signals and the set threshold or thresholds of these levels, additional control signals for the state of the amplifier are generated, by means of which the voltage level and / or the operation mode of the amplifier are changed, for example, it is transferred from AV mode to A.

The processing unit is performed with at least one additional input, to which electrical signals are generated that are formed near interference sources, for example, an alternating voltage from one of the windings of a transformer supplying an amplifier, these signals in the processing unit are subjected to single or multi-band processing to generate signals intended for actively lowering the level of these signals in the above amplification stages, at the output of an amplifier or system, the amplifier and emitter, the processed signals are used to form the input signals of the processing unit by summing them with other output signals of the processing unit.

The object of the invention is solved in that in the known method of threshold noise reduction of electrical signals of a message source, which consists in automatically reducing the transmission coefficient of electrical signals of a message source from input to output of a noise canceler at times when the signal level of a message source is lower than a threshold, according to the invention, the input signal of a noise suppressor is filtered to receive several band signals of the message source, the number of which is not less than two, and at least in two of these their bands carry out threshold noise suppression of the components of the electrical signals of the message source, the signals processed in this way are used to form the output signals of the noise suppressor.

Possible implementations of the method, such that:

each multi-band processed signal is additionally filtered during the formation of the noise canceling output signals to reduce the level of nonlinear and intermodulation distortion of the noise canceling output signals;

multiband processed signals are summed to obtain one or more noise suppressor output signals for subsequent processing, for example, by power amplification in signal amplifiers.

The technical solutions described above fill with concrete content a new fundamental law of nature - THE LAW OF MINIMIZING ENTROPY GROWTH (communication system) or the law of "conservation" of information. Quotation marks mean that this is not about absolute conservation of information like the well-known classical laws of conservation of energy and matter, formulated and working in the framework of a physically never realized model - a circular process for which, by definition, the growth of entropy is zero, but the maximum possible saving information (signal parameters) in any real physically feasible system. Moreover, the growth of entropy in the system is “slowed down” due to an external energy source (power source) and a new structural scheme of the system that provides a current analysis of the distorting properties of all elements of the communication system, including the communication channel itself. The use of feedback signals allows one to “calculate” the entropy of the communication system all the time and compensate, as far as possible, the growth of entropy.

Unlike traditional approaches to calculating entropy using thermodynamic concepts and categories, the invention describes a method and related technical solutions using basic categories and concepts of information theory. The essence of the new fundamental law of nature is disclosed in this invention using categories that are more understandable for radio specialists: signal spectrum, frequency band, signal components and their parameters.

Thus, the use of methods of multi-band processing of message source signals, feedback signals, and signals of various noise sources makes it possible to increase the accuracy of transmitted messages by practically analyzing and processing these signals component-wise. This approach clearly corresponds to the basic principles of information theory (spectral analysis of signals) and cybernetics (automation of signal processing presses using feedback and appropriate algorithms).

These advantages, as well as features of the present invention will become apparent during consideration of the following embodiments of the invention.

Brief Description of the Drawings

Figure 1-4 depict generalized structural diagrams of embodiments of messaging systems according to the scheme with feedback and special signal processing.

5, 6 are generalized block diagrams of signal processing units of the systems shown in FIGS. 1-4.

Fig. 7 explains the principle of automatic generation of temporary predistortions when generating signals for actively lowering signals of any physical nature, and in the embodiment, sound noise signals.

Figs 8a), 8b) depict circuits of processing units and their connection with other nodes of aircraft active noise reduction systems.

Figures 9-12 also explain the principle of operation of aircraft noise reduction systems in an analogous invention.

Figa), 15b), 14, 15 depict structural diagrams of signal processing units according to the invention.

Fig.16 depicts one possible variant of the circuit diagram of the processing unit with multi-band formation of noise reduction signals.

17, 18 depict generalized structural diagrams of signal processing units for transmitting signals of any physical nature in channels with interference and noise according to the invention.

Fig.19, 20 depict examples of circuit design of the main nodes of the signal processing units according to the structural diagrams shown in Fig.13-15, 17, 18, the analysis of which explains the essential features and their relationships according to the claims.

21, 22 explain the principles of multi-band signal processing when generating feedback signals in electric signal amplifiers, and also explain the methods for transmitting messages in the system of the amplifier and emitter, covered by feedback according to the invention.

The best embodiment of the invention

Various techniques are known in the art for solving the problem of increasing the accuracy of transmitting messages (information) in channels with interference and noise. This is due not so much to the difference in the possible physical nature of the transmitted messages, but to a different general methodological approach to solving this problem and the conditions for its formulation.

A group of inventions is known in which this problem is solved. The basis of these inventions are well-known provisions of the modern theory of information (cybernetics). For the first time, in the form of a generalized scientific approach, these provisions were formulated in the book of N. Wiener "Cybernetics". The essence of this concept lies in the fact that the optimal communication or control system should be built according to a feedback scheme and special processing of transmitted and received messages (signals, commands, actions, etc.). The legitimacy of this approach is justified by the fact that the most effective control of any process (in particular, the process of transmitting messages) in animate and inanimate nature obeys some general universal (fundamental) laws of nature. In the above group of inventions, the basic technical solutions have been found that reveal the essence of this new fundamental law of nature - the law of minimizing the growth of entropy (communication or messaging system) or, in other words, the law of "conservation" of information. Therefore, before proceeding to the consideration of technical solutions according to the invention, it seems appropriate to recall the various technical solutions disclosed in the aforementioned group of inventions. These technical solutions, supplemented by new significant features, will be used in the invention.

Figure 1-4 shows a generalized block diagram of various options for systems with feedback. These systems contain: a signal signal source 1, a signal processing unit 2, configured to perform various operations on the source 1 signals and feedback signals — received at the point (region) of receiving messages by electric signals, an amplifier 3, an additional amplifier 9, an emitter 4, an additional emitter 10, a probing device 5, a communication line 6, through which the received electrical feedback signal is transmitted to the place of its processing, an additional filter 11. In Figs. 1-4 7 shows the same channel and an additional channel 12 is a noisy channel 8. Parameters 7 and 12, 8 and noise characteristic spatial coordinates of receiving messages, irradiation points are random parameters and can vary randomly. It is believed that the shape of the current (voltage) of the transmitted signal of source 1 is a random function of time (process).

Feedback systems (Figs. 1-4) comprise a multifunctional unit — a signal processing unit 2. Its inputs receive electrical signals of source 1 and feedback signals. In this block, the following operations on these signals can be implemented:

a) automatic calculation and generation of levels and / or delays (phases) of the signals of source 1 in various signal processing bands. In other words, in these bands, actions on their energy and temporal predistortions are performed on the filtered components of the message signals. Such actions can be characterized as frequency-energy and / or frequency-time predistortion of the signals of the source 1 messages. They are equivalent to the work of the automatic equalizer and / or automatic corrector of delays (phases) of the signal components of the message source 1;

b) automatic multi-band signal generation for active noise reduction. This operation consists in multi-band filtering of the received electrical feedback signals by means of bandpass filters identical to the bandpass filters that are used to filter the source 1 signals. These feedback signals consist of components of the source 1 signals distorted in the communication channel (energy parameters are distorted - the levels of the components and time parameters - the location of the components relative to each other in time), as well as the components of interference and noise.

Using special processing, multiband filtered feedback signals are generated from multiband filtered feedback signals (components) to actively reduce noise and interference, the levels and phases of which are also optimized automatically and as quickly as possible to achieve the greatest effect of actively reducing these noise at the point or area of the message;

c) the output signals of the processing unit 2 are generated as a result of various combinatorial options for summing the multiband processed signals of the message source 1 and multiband generated signals for active noise reduction.

Various combinations of the functional capabilities of signal processing unit 2 and the options for constructing generalized block diagrams of systems for transmitting messages allow implementing various versions of systems with feedback and special signal processing. For example, to solve the problem of transmitting messages under conditions of complete uncertainty of various parameters of signals, interference and characteristics of the communication channel, you can use the structural diagram shown in figure 1. If in block 2 of the signal processing is carried out full-parameter signal processing, then the system (figure 1) can be considered optimal. In this system, a complex, composite broadband signal is formed at the output of the processing unit 2. It contains both signal components for active noise reduction and components of the pre-emphasized information signal of source 1. This signal is amplified and radiated into the channel to the point of receipt of the message, in the immediate vicinity of which the receiving element of the probing device 5 is installed, by means of which an electrical feedback signal is generated . Using an additional communication line 6, this signal is transmitted to the place of its processing - to block 2, in which the above signal processing is carried out and an output signal is generated.

Figure 2 shows a variant of constructing a structural diagram in which the number of output signals of the signal processing unit 2 is at least two. The additional outputs of block 2 allow the implementation of multi-band amplification and emission components of both the signals of the source 1 and the signal components for active noise reduction.

The processing unit 2 and the signal source can also be made with several outputs, but for another function — for multi-channel transmission of messages in the field of the receiving space, for example, in stereo systems or stereo multi-personal (collective) systems described in the invention RU, A, 2145446.

This invention shows that if you adhere to the well-known provisions of cybernetics, then for one person (listener) at the stage of recording and playing back audio messages, only two channels are needed, and not 3, 4, 5, or even 8, as it is proposed to do in a number of the latest Alpine, Sony, Kenwood, Pioneer, etc. high-quality sound reproduction systems due to the inclusion of a signal processing unit, for example, Dolby Surround / Prologic, Dolby Digital (AC-3), DTS, Dolby Digital Surround EX, Dolby Digital THX Sorround EX, Super Audio CD (SACD), PXA H900 (Alpine F # 1 Status) (see, for example, the magazine "Master 12 Vol t ", February-March 2002, str.62-66).

The multiband principle of signal transmission of source 1 can also be obtained by additional filtering of the output broadband signal of processing unit 2 - Fig. 4, or by additional filtering of the amplified output signal of unit 2 - Fig. 5, which, in essence, requires performing equivalent actions on the signals , but from a structurally practical point of view is not very convenient.

2-4 are various embodiments of multiband messaging systems.

Block diagrams of processing units with one or more outputs are shown in FIG. 5 and FIG. 6, respectively. They may contain: control device 13, band-pass filters 14, blocks 15 for calculating and generating levels and / or delays of signal components of source 1, signal generating blocks 16 for active noise reduction, adder (s) 17, phase shifters 29, phase shifter control devices 30 29, the second controlled amplifiers 31. Block 13 as an auxiliary unit is shown schematically. The main function of block 13 is to control the nodes of blocks 15 during the formation of one of the possible algorithms described in the analogous inventions, during cyclic correlation processing of signals in blocks 18, as well as the possible optimization of the operation of blocks 18 and 16.

Blocks 15 can be made in the form of blocks 18 and / or 19. In blocks 18 (due to the operation of the detectors 25, low-pass filters 26, resistors 27, acting as an adder of signals from the outputs of the low-pass filter 26, and the first controlled amplifiers 28), the formation ( calculation, if the corresponding nodes are implemented not in hardware, but in virtual - software way) of signal levels in each signal processing band of source 1. In blocks 18, by means of blocks 20, 21, 22, 23, 24, the predistortion delays of the components of source signals 1 are generated. operation of these units is performed cyclic correlation analysis components message signal, reverse allocation delays of these components, provides alignment phase signal components at the point of receiving the message according to their location in the interim signal transmitted message. Figure 5 shows an embodiment of a signal processing unit 2 with a single adder 17 for receiving one wideband signal at the output of this block. In Fig.6 introduced an additional adder (s) 17 to generate multiple output signals of the processing unit 2 and implement a multi-band method of transmitting messages.

The operation and interaction of all nodes of the systems shown in figures 1-4, and nodes of blocks 2 of the signal processing is well known. As block 2, a specialized microprocessor device or computer can be used, for example, a personal computer (PC) with appropriate software, as well as input devices (ADCs) and information output (DACs) (for example, audio signals). Structurally, these devices can be combined into a single unit (for example, a central processor unit for controlling automobile or aircraft systems, a multifunctional sound reproducing device for a car) or be distributed among individual functional modules or system blocks.

As already noted, depending on the functionality of the processing unit 2, various options for the formation of its output signals are possible as a result of general or group summation of multiband processed signals:

- summarized multiband processed only at the levels of the signals of source 1;

- summed the multi-band processed only by the delays of the signals of the source 1;

- summed the multi-band processed by levels and delays signals of source 1;

- summarized multiband signal-processed signals of source 1 and multiband generated signals for active noise reduction;

- summarized the multi-band processed by levels and delays signals of the source 1 and the multi-band generated signals for active noise reduction;

- summed the multi-band delay-processed signals of the source 1 and the multi-band generated signals for active noise reduction;

- summarized multiband-generated signals for active noise reduction.

The last embodiment of the processing unit 2 does not relate to a message transmission system in conditions of noise and interference, but to an active noise reduction system. Generalized structural diagrams of various options for noise reduction systems are obtained from the circuits shown in figures 1-4, by excluding from their composition the signal of the source 1 and a number of nodes from the structural diagrams of the processing units 2 shown in figure 5, 6. Indeed, if in the region of the noise reduction space, trying to “transmit” a zero-level message (“silence”) by applying a feedback signal generated in the noise reduction region to the 2 input of block 2, and a received noise signal with pre-selected parameters to the 1st input, it is possible from structural schemes of processing units 2 (FIGS. 5, 6) to exclude nodes 20, 21, 22, 23, 28 in the signal paths of source 1. In the noise reduction system, these blocks are simply not needed. Thus, the block diagram of the signal processing unit 2 may contain filters 14 (at the inputs of block 2), detectors 25, an low-pass filter 26, phase shifters 29, phase shifter control devices 30, second controlled amplifiers 31, and adder (s) 17. This option will be considered below in details. Concluding the general review of well-known technical solutions of feedback systems, we recall the principle of automatic signal generation for active noise reduction using Fig. 7. It shows a controlled phase shifter 29 containing a multi-position controlled switch 32 and first additional delay lines 33 ₁ , 33 ₂ , .., 33. The outputs of the first additional delay lines 35 are connected to the inputs of the controlled multi-position switch 32. The delay value can be selected with some step e.g. constant or variable. The phase shifter control device 30 includes: a second additional delay line 34, additional detectors 35, additional low-pass filters 36, an additional comparison circuit 37. The input of the device 30 is a common input of two branches, in one of which the first additional detector 35 and the first additional low-pass filter 36 are connected in series , in another, a second additional delay line 34, a second additional detector 35 and a second additional low-pass filter 36 are connected to each other in series. The outputs of the additional low-pass filters are connected to the inputs. effective comparison schemes 37.

The phase shifter operates (Fig. 7) as follows. A multi-position controlled switch 32 sequentially switches contacts in direct relation to delay line numbers or in reverse order. The switch 32 is controlled by the signal from the output of the additional comparison circuit 37 according to the rule: when the next signal arrives, the order of switching the delay lines 33 is reversed. In the device 30, a noise signal and a noise signal delayed for a time τ34 are detected and integrated in different circuits, forming input signals for an additional comparison circuit 37 (comparator). The output signal of the comparison circuit 37 is formed if the level of the delayed signal becomes greater than the level of the uncontrolled signal. This means that the delay introduced at the next step of the experimental selection of the phase of the noise-reducing signal has become non-optimal, and it is necessary to begin the selection of the phase in the opposite direction. These actions are equivalent to studying the phase parameter at an extremum, or rather, an extremum study of the noise function, which is a superposition (sum) of noise and noise reduction signals. This method allows you to automate the process of experimental selection of the phase of the signals for active noise reduction, providing the greatest noise reducing effect (minimum noise function).

In block 31, the optimal signal level for active noise reduction is formed. This level also leads to the greatest noise reduction effect. It is clear that by choosing various time constants of the processes of automatic selection of phase and level, you can untie the sequence of optimization of energy and time predistortion of signals for active noise reduction in time. For example, choosing the time constant for adjusting the signal levels by an order of magnitude greater than the time constant for optimizing the phase parameter, during relatively slow adjustments of the noise reduction signal level, it is possible to manage to select the phase of this signal all the time, which ensures the lowest possible noise level in the region of its decrease. If only interference is present in the region of the message receiving space (near the probing device 5), and there is no useful information signal, the signal controlling the operation of block 31 is generated only due to the current level of noise components in the corresponding frequency band of their analysis and processing. The higher the level of these noise components, the greater the gain of block 31 and vice versa. The dynamic range of regulation of the gain of the block 31 and the range of regulation of the phase or delays determines the system's ability to reduce noise in various applications. Using information on the parameters of the communication channel and noise, it is possible, for example, to limit the dynamic range of adjusting the levels and phases of the generated signals up to the complete cessation of these adjustments and thereby speed up the process of searching for optimal signal pre-emphasis. For example, if it is known that the interference is quasistationary in nature, the geometric parameters of the system remain practically unchanged, there is a priori information about the spectra of signals and interference, then it is possible and even advisable to replace automatic nodes for selecting levels and phases of generated signals with non-automatic ones. Having preliminarily tuned the message transmission system or noise reduction and fixing the predistorted adjustment signals, it is possible to solve the corresponding problem, but for most applied problems of information theory this simplified version is not suitable. Let us explain this by the example of solving the problem of reducing the noise level of aircraft motor vehicles (aircraft or helicopters).

On Fig shows a device for generating signals for active noise reduction, described in the invention of JP, A, 52-13721 author Chorida Akira. It consists of a unit 2 for processing received noise signals 8 from a noise source through a microphone 0. Processing unit 2 is made in the form of a transistor, the input of which receives a noise signal from microphone 0. Antiphase noise signals from the emitter and collector of the transistor are fed to the unit for generating output signals of the unit 2 treatments. In this node, signals are generated for active noise reduction. These signals are either delivered to speaker 4 in a single-channel system, or through amplifiers 5, 9 to speakers 4, 10 in a multi-channel noise reduction system of an airplane. Sound attenuating waves propagate in all directions from the aircraft, as shown in Fig.9-11. On Fig shows the spectra of the noise signal - curve a) and the noise reduction signal - curve b), as well as the spectrum of the noise signal on the ground - in the region of its reduction space - curve c). The energy spectra of the signals shown in FIG. 12 do not accurately reflect their actual appearance. The spectra of the received noise signal near the microphone 0 and in different areas of space on the ground can differ very much due to the difference in the phases of the noise components due to the spatial distribution of various noisy nodes and units of the aircraft. Spatially distributed noise sources create sound noise waves that arrive at the microphone’s location on the plane and in the region of space on the ground with different energy and phase (time) parameters, and the time delays of individual components can vary very much and can be units or even tens of wavelengths . Therefore, as a result of the signal processing described in the above application, it is possible to select the levels and phases of noise reducing signals for a very narrow frequency band. These pre-emphasis for other frequencies will no longer be optimal, and instead of the noise reduction effect, the opposite effect can be observed - the noise amplification effect. The noise reducing effect for a narrow frequency band will also be provided in some solid angle and in the corresponding area on the ground. For other points of space on the earth this effect may not be, but on the contrary, noise amplification due to the common-mode addition of noise and noise-reducing waves can take place. Due to this circumstance, it is impossible to actively reduce noise by methods of active noise reduction in all directions from the aircraft, as shown in figb).

The active noise reduction method described in this invention is of interest solely in terms of the possibility of providing high isolation between received noise signals and noise reduction signals emitted into the region of the noise reduction space. To eliminate the above disadvantages of this system, it is necessary to use a more advanced embodiment of a block 2 for processing received noise signals. In this block, it is advisable to use methods of band-signal processing in which the phase and energy parameters of the signals should be automatically or non-automatically selected, providing the effect of noise reduction and increasing the accuracy of the transmitted message in the corresponding region of the message receiving space. To ensure a guaranteed noise-reducing effect, it is necessary to know where the area of noise reduction and / or receiving a message is located, what shape it has, and in connection with this information, it is necessary to select the highest frequency of the generated signals for active noise reduction. This follows from the fact that if for higher frequencies the noise-reducing effect is observed in a certain local region of space, then it will also be lower than the higher frequency for all frequencies, since the wavelengths of these frequencies are longer, and for them the limiting sizes of the region of guaranteed noise reduction increase . The above limitation of the frequency band of the noise spectrum can be implemented by filter 14. As this filter, you can use the low-pass filter or PF. For various tasks, the size of the area of noise reduction (receiving a message) may differ, as well as the location of the points of reception and emission of signals. For some of these tasks, multiband analysis and selection of signal parameters within the above frequency band of the noise spectrum may be required. Therefore, the relationship between the points of reception, emission of signals, the highest frequency of the generated signals and the size, shape and position of the region of the noise reduction space can be considered essential features, indicating them in the claims.

For example, if you provide a noise-reducing effect in a limited frequency band from 20 to 50 Hz in a certain direction with respect to the plane (for example, "down") in the corresponding solid angle, then the projection of this solid angle onto the earth’s surface, and therefore the size, shape and the location of the area of the noise reduction space will vary with respect to both the point of reception of noise signals on the plane and the emitters (loudspeakers) of the noise reduction system.

If the higher frequency of noise reduction signals is changed, for example, by decreasing it from 50 Hz to 40 Hz, then the solid angle will change accordingly, in which noise and noise reduction waves give a total noise reduction effect. This solid angle and its projection on the earth will increase, and the effect of noise reduction at frequencies from 20 to 40 Hz will be in the area of space on the earth of large sizes and most likely of a different shape, since the earth's surface has a relief surface. In this area of space, the noise-reducing effect at frequencies from 40 to 50 Hz will no longer be. Thus, the limitation of the signal band for active noise reduction leads to an increase in the size of the area of the space of guaranteed noise reduction. It is clear that this effect can be used in various applications, regardless of the physical nature of noise signals, processes, fields, etc. Therefore, when formulating the appropriate method of reducing noise or transmitting messages, you can use the extremely generalized formulation of an essential feature in the form: "signals of any physical nature." For the convenience of presenting technical solutions, audio signals (messages and interference) will be used as such signals in the future. In the claims, these technical solutions will be indicated in the dependent paragraphs as one of the possible embodiments of the above generalized characteristic.

If the signal spectra in the region of the noise reduction space or the receipt of a message and the receiving point of the noise signals have almost constant differences, then these differences can be compensated by multiband processing of the noise signals, by performing non-automatic frequency-energy and time-frequency predistortions of the received noise signals.

13 a) shows an embodiment of a signal processing unit 2 for multi-band non-automatic processing of noise signal components during signal generation for active noise reduction. This option can be used, for example, in solving the problem of reducing the noise level in air transport, road transport or railway transport. By installing a microphone or microphones 0 for receiving noise signals near noise sources (engines, exhaust pipe, wheels, etc.), it is possible to experimentally select the energy and time parameters of the components of noise signals, as well as selecting the location of microphones and signal emitters noise reduction in places of the most likely location of people, for example, near the auditory organs of a person sitting in this vehicle. It is clear that since a person has two hearing organs, the area of the noise reduction space should be two localized areas of space near each of the hearing organs. Considering that the head of a passenger or a driver of a vehicle practically does not move in space over considerable distances during most of the travel time, then by limiting the size of local areas, for example, by spheres with a radius of the order of several centimeters, it is possible to increase the higher frequency of noise-reducing signals to about kilohertz and effectively lower the noise level at the lower frequencies by selecting the frequency-energy and time-frequency parameters of the components of noise signals. With a sufficiently large number of noise signal processing bands (of the order of 50-100), the maximum level of noise reduction can be about 55-40 dB. This level will be determined mainly by the products of nonlinear and intermodulation distortions that occur in loudspeakers of noise reduction systems. In HI-End class distortion systems, speakers are on the order of 1-3%. The location of the receiving microphone near the noise source, and the emitters of noise-reducing signals near the human hearing organs provide a high level of isolation of the received and emitted signals, as well as the ability to accurately synchronize in time the noise and noise-reducing signals in the area of the noise reduction space near the human hearing organs. In addition to two-channel noise reduction systems, options are possible in which a noise reduction signal for each local area of the noise reduction space is formed from one received noise signal, or an option in which several noise signals are received for each channel or group of channels, for example, passengers sitting in one row . With lower noise reduction efficiency, a single-channel noise reduction method at lower frequencies of the noise spectrum can also be used.

On figb) shows an embodiment of block 2 for multi-band non-automatic processing of components of noise signals and message signals. The selection of parameters of message signals compensating for distortions arising in the communication channel is carried out by means of nodes 28, 20, 24. As can be seen from a comparison of the circuits shown in FIGS. 5, 6 and 13a), b), non-automatic options for selecting frequency-energy and time-frequency predistortions of message signals and interference are obtained from the well-known by eliminating a number of nodes responsible for the automation of these processes. If the spectrum of the noise signal in the region of the space of its reduction can randomly and very strongly change with respect to the spectrum of the noise source, then it is necessary to install the receiving element of the sounding device 5 in the region of the space of noise reduction (or receiving a message) for additional reception of the noise signal. This signal is a feedback signal and allows you to automatically generate signals for active noise reduction. In this embodiment, the processing unit 2 contains additional nodes that automate actions to select the levels and / or phases of multiband signal components to actively reduce noise, as well as actions to memorize and generate energy and temporal predistortions that provide the effect of reducing noise and improving the accuracy of the transmitted message . On Fig, 15 shows embodiments of automatic processing units for received noise and additionally received signals. Variants are possible when only temporal or energy parameters of received noise signals are automatically corrected, or the range of adjustments is limited. For example, phase adjustment can be carried out in the range of ± 30 °, and the level of signal components in the range of ± 6 dB. It is advisable to use such options if there is information about changes in the signal spectra. Using possible combinatorial options for summing multiband processed received noise signals and message signals, it is possible to realize: a) single-channel - broadband or multi-band; b) multichannel broadband or multiband systems, allowing the formation of single or multi-localized areas of the space for receiving messages and reducing noise. In the practical implementation of such systems, you can use industrially produced devices that allow you to take actions on the selection of energy and time parameters of signals in several channels, as well as perform multi-band filtering of signals. For example, using a signal processing device — the Alpine RXA-900 processor, you can implement a number of the technical solutions described above or use simpler devices based on modern operational amplifiers (op amps). An example of such an embodiment of the noise signal processing unit 2 is shown in FIG. 16.

The input four-band filter is made on the OA D1 and allows you to form four half-octave band processing of noise signals, tuned, for example, at 30, 45, 60, 90 Hz. Temporary correctors are made in the form of phase shifters at op-amp D 2. The level controllers of the components of the noise signals in the processing bands are made in the form of tuning resistors R 7, 16, 24, 32. The adder 17 is made at the op-amp R 3. Tuning resistors R 2, 14, 19 , 27 are used to tune filters 14 to the above frequencies. The calculation of the parameters of the band-pass filters 14 is carried out according to well-known methods. The parameters of capacities and filter resistors are selected with an accuracy of 1%. Resistors R 6, 15, 23, 31 are used to adjust the phase of the components of the received noise signals. For the specific task of reducing noise, experimentally find the location of the receiving element of the probing device 5 (microphone 0) and emitters 4, 10, providing the maximum possible noise reducing effect. For example, the receiving microphone 0 is installed in the engine compartment of the car, and the speakers in the rear shelf or side doors of the car. The noise reduction frequency band is also selected based on specific conditions. In this example, the frequency band of effective noise reduction is about 20 ÷ 120 Hz. It is clear that the highest frequency can be adjusted depending on the task 4 of a particular type of vehicle. Using such signal processing units 2, it is possible to create many localized areas of reduced noise, for example, near the heads of passengers of airplanes, cars, trains - in the places of their most likely location. In this case, the signal processing unit 2 is executed in the form, for example, of the same processing units 2 shown in FIG. 16. If the task of reducing noise outside a vehicle, such as an aircraft, is solved, then multi-channel or multi-band emitting sound noise-reducing signals with pre-selected pre-distorting adjustments of their parameters, it is possible to form a kind of noise-reducing diagram in a certain direction, for example, "forward" and "down" when approaching the airport. It is clear that the laws by which this noise reduction diagram of an airplane is formed are in many respects similar to the laws of radiation formation in the radio range. For both radio signals and sound signals (waves), the appearance of this diagram can be “needle-shaped” or be multi-petal, for example, “forward” and “down” or “forward” and “sideways”, etc. For the formation of such diagrams, it is necessary to experimentally select the optimal predistortion of noise signals that provide the desired form of this diagram. For this, you can use, for example, the following technique. On the ground with a certain step (100-500 m), measuring microphones are installed in a line or squares over which an airplane flies with the noise reduction system turned on. Moreover, the aforementioned energy and time predistortions in the signal processing bands are automatically switched at a certain frequency and according to a known algorithm. As a result of these switchings, a total enumeration of all possible combinations of energy and temporal predistortions is carried out. As a result, a certain data array is formed in the form of an ensemble for the realization of sound signals recorded by microphones on the ground. Using any of the known methods for synchronizing the recordings of these realizations with the flight path and speed of the aircraft, as well as with the sequence of the above-mentioned switching of signal predistortions during the formation of noise reducing signals, it is possible to process this data and establish at which noise distortion pre-emphasis the greatest noise reducing effect is required , or vice versa, which pre-emphasis is necessary to create, in order to provide min mal noise. At the same time, in order to reduce the volume of research, it is possible to carry out rough, preliminary research on the ground, around a stationary plane. These studies can also be used in the formation of noise reduction diagrams during taxiing of the aircraft to the runway, takeoff or landing of the aircraft to reduce noise in the direction of the airport. Other techniques are also possible, for example, with a spatial flight of an airplane on another airplane and recording the noise level in the desired direction and the corresponding selection of pre-emphasis adjustments with the coordination of these actions by radio.

The technical implementation of the nodes shown in FIGS. 14, 15 may be different — hardware, using, for example, an op-amp, or virtual, using a computer (PC). For example, the nodes 33, 34, 37 shown in Fig. 7 can be implemented on the basis of the op-amp, and the switch 32 can be implemented in the form of a simple electromechanical (step-by-step) switch with an electric motor, to which voltage of different polarity is applied from the output of circuit 37, which determines the direction of switching of the switch 32 or in equivalent electronic form using a multiplexer. But it is much more practical to execute these nodes virtually - in the form of corresponding subprograms of the type: “delay line with constant or variable delay time”, “integrator”, “detector”, “comparator”, “switch”, etc. The main advantages of the virtual signal processing method are well-known: high repeatability of a large number of modules of the same type (for example, filters 14), size reduction, power consumption of processing units 2, low cost of production of a software product compared to its hardware counterpart. The second controlled amplifier 31 can also be performed in any known manner: in the form of an analog controlled amplifier, attenuator or virtual node - the corresponding operator or function.

The above options for the implementation of various methods of noise reduction (signal level of any physical nature) allow you to go on to consider methods of transmitting messages and reduce noise according to the invention.

It is advisable to set forth these methods on the example of the transmission of sound messages, which is due to the fact that sound engineering, like radio communications, can be considered leaders in the number of known methods of transmitting messages, various options for setting these tasks and solving them. If technical solutions are being introduced into industrial production in radio communications, as a rule, strictly corresponding to well-known provisions of the theory of information and cybernetics, then in sound engineering, and in particular in household, this does not always happen. This trend has already become noticeable even to non-specialists. For example, demanding consumers of high-quality sound reproduction equipment cannot understand why Sony and Philips manufacturers increased the frequency range of their latest SACD system to 100 kHz, and the signal-to-noise ratio to 120 dB. It is well known that a healthy, young person with good hearing simply cannot physically hear sound vibrations at frequencies above 20 ÷ 24 kHz, older people, as a rule, do not hear sounds with a frequency above 10 ÷ 12 kHz, and in real acoustic conditions they are living (residential) In principle, a listener will never be able to hear quiet sounds reproduced by this system almost at the threshold of a person’s auditory abilities, due to a significantly higher level of interference or noise. The level of these noise in the living room is about 30 ÷ 80 dB, and in the car about 50 ÷ 100 dB. Without special measures to combat interference and noise, using a system with a dynamic range of 120 dB is not logically justified. In addition, most of the recorded phonograms, records or CDs have a signal to noise ratio (noise) of not more than 50 ÷ 90 dB. Recordings with a real signal to noise ratio (noise) of 120 dB are almost impossible to obtain even in studio conditions. Such a signal-to-noise ratio is achievable only under conditions of artificially synthesized sounds, for example, using electronic synthesizers in conjunction with noise reduction systems of electrical signals of message sources 1. Wealthier buyers have even more questions when they are offered to purchase sound-producing equipment of the so-called HI-End class at a price of, for example, $ 38,000 for a Wavac HE-805 single-cycle tube amplifier assembled on lamps developed in the 1950s century, or the volume control of the same company ATT-Q at a price of 6900 US dollars, the function of which is only to adjust the signal level. No less surprising is the endless debate and talk about which amplifier is better, tube or transistor, debate about the advisability of using negative feedback in power amplifiers or connecting speakers to an amplifier with conductors specially designed for this purpose, the cost of which is comparable to the cost of an amplifier, and which without heating, they can withstand currents of tens or even hundreds of amperes. The latest trends in audio and video technology for the use of large-format images, in which the consumer sees the standard format in a very flattened, distorted form, supplemented by the so-called "HI-End class home theaters" with an incredible number of sound channels and sound pseudo-effects, are also surprising. From which provisions of the modern theory of information these improvements follow and are presented as useful latest technical solutions that the consumer must additionally pay for, it is completely not clear. Modern methods of processing audio signals to obtain pseudo-acoustic effects and some methods to combat noise in no way fit into the well-known provisions of modern information theory. One such approach is, for example, a method of automatically adjusting the volume depending on background noise in a Blaupunkt (BOSCH) San Francisco CD 70 car CD receiver. As a result of automation of volume control, the listener receives sound information in a very distorted form. The sound message signals are modulated in this device by an average noise level, which is noticeable by ear, annoys the listener with the instability of the sound volume and does not allow listening to messages at the desired volume level. As a result of the operation of this system, the total, total level of sound signals of messages and noise increases sharply, which leads to driver fatigue. Accident rate increases in vehicles during the operation of such systems. This system does not essentially solve the problem of reducing the noise level, which must be done in accordance with modern scientific principles of information theory. At the same time, this system implements a function that allows you to automate the regulation of frequency response by sound pressure at various acoustic properties of the vehicle interior. This function corresponds to the modern principles of information theory, patented in the above-described group of inventions with feedback, and will actually increase the accuracy of the transmitted message in the space of its receipt. Given the rather high level of technical and commercial development of this technical solution, it seems appropriate to choose Blaupunkt's San Francisco system as a prototype of new methods and systems for sending messages of any physical nature. Supplementing it with technical solutions for active noise reduction, it is possible to implement a more advanced and safe device for high-quality reproduction of sound signals, designed for installation on motor vehicles or other vehicles (airplanes, trains, ships, etc.).

In the framework of the objectives of the present invention, it is also advisable to consider a number of particular tasks aimed at improving the accuracy of the transmitted message by improving a number of actions and the corresponding nodes of the messaging system that are part of the invention. The combination of all these technical solutions will be the subject of the invention.

The general idea of the present invention is that better methods for transmitting messages in the presence of interference can be obtained by a variety of combinatorial combinations of the known noise reduction methods described above and the known message transmission methods.

A new, fundamentally important point, necessary and sufficient for combining these technical solutions, is that the preliminary selection of signal levels and phases for active noise reduction must be carried out at a zero level of electric signals of source 1 or at a level of signals of source 1 lower than a predetermined one - below a certain the threshold. It will be shown below that this condition is not trivial and obvious for specialists, clearly following the prior art, if only for the operation of such systems, it is necessary to implement the above condition at the level of the system setup algorithm and perform appropriate actions and include new ones in the system nodes allowing, for example, to automate the implementation of this condition. Otherwise, the message signals will interfere with the noise reduction system, and any of the options for tuning the noise reduction system will be impossible.

The above condition for a non-automatic version of the implementation of energy and time or frequency-energy and frequency-time predistortion of signals is achieved as a result of the appropriate technique (sequence of actions) - the selection of adjustments of the levels and phases of message signals and signals for active noise reduction. First, the noise reduction system is set up, for example, when the signal source is turned off, and then pre-emphasis signals of the message source 1 are already selected, if the noise sources cannot be turned off. When implementing automatic versions of noise reduction and message transmission systems, corresponding nodes will be required, connected via certain new connections with other nodes of the signal processing unit 2. The functional purpose of these nodes is to determine the moments of time when the signal level of the message source 1 becomes lower than the predetermined one, and starting from this point in time, turn on the nodes to automate the processes of selecting the energy and time parameters of the signals for active noise reduction. At other points in time, previously selected signal parameters (corresponding adjustments to the levels and phases of signal components) that provide optimal predistortion of noise signals are stored in the memory of processing unit 2 and maintained until their next auto-tuning at a low signal level of the source or its components in the corresponding signal processing band. This approach can be implemented when analyzing the signal level of source 1 as in the entire band, by automatically selecting signal parameters for active noise reduction, for example, before sending messages or in pauses between blocks of information messages (songs, words, etc.), or directly in the process of transmitting messages by analyzing the multi-band filtered components of message signals in the time and frequency domains. When the level of message signal components in a given analysis band is lower than a predetermined threshold in the corresponding frequency band, automatic adjustment of signal parameters for active noise reduction can be automatically performed. In other analysis bands, where the level of the signal source 1 signal components at this time is higher than the threshold, noise reducing components are formed due to previously selected pre-emphasis. The levels and phases (delay times) at this time remain unchanged. It is clear that the threshold relative to which there is a transition from a non-automatic option to reduce noise to automatic can be varied and thereby further optimize the system in unpredictable messaging conditions. A criterion for such optimization can also be selected as a criterion for the minimum standard deviation of the signal of the transmitted message at the output of the signal source 1 and the signal at the point of receipt of messages (for example, at the output of the probing device 5). This additional, optimization function of the signal processing unit 2 can be schematically attributed to the functions of the control unit 15 shown in FIGS. 5, 6.

From a methodological point of view, the same approach can be used to increase the accuracy of message transmission when they pass through a number of elements of the message transmission system, for example, when they are amplified in amplifiers (3, 9) or when they pass through a subsystem, the amplifier (3,9) associated with emitter (4, 10). Message transmission in these subsystems can also be improved using, for example, multi-band processing of message signals and multi-band processing of feedback signals generated from the output of the amplifier (3, 9) or at least one of its stages, or feedback signals generated near the radiator (4, 10). The principle of multi-band signal processing and the use of processing unit 2 for a new purpose allows us to consider these new technical solutions as independent inventions that allow us to solve the problem of the invention. Moreover, a number of particular embodiments of the signal processing unit 2 allows to improve well-known circuits of noise suppressors of electrical signals, for example, to implement a fundamentally new multiband threshold principle (method) of noise reduction and a corresponding device for its implementation.

To clarify the foregoing, we use Fig.14, 17, 18, 19.

On Fig shows one of the possible embodiments of the block 2 processing the signal source 1 messages and received noise signals, the levels and phases of the components of which are set automatically. The signals of the source 1 messages are fed to 2 input (level or volume control R 1). Signals received near a particular noise source are fed to 1 input of a signal processing unit 2. One of the possible embodiments of the system can be obtained by eliminating the nodes 25 and 26 shown in dashed in Fig. 14 (15).

So that the signals of source 1 do not interfere with the noise components, in the process of automatic generation of signals for active noise reduction due to the operation of nodes 29, 30, 31, it is possible to decouple them from the source 1 signal in the frequency and time domains. Additional selection of the signal source 1 messages, for example, using a high-pass filter (14 ') with a cutoff frequency of, for example, 100 Hz and filtering the received noise signals with a PF (14 _R ) at frequencies below 100 Hz ensures the fulfillment of the above condition (at the source signal level ) at any given time. The signal of the message source at any time will not interfere with the automatic noise reduction system operating at frequencies below 100 Hz and providing a noise reduction effect in the corresponding space region. As a high-pass filter (14 '), you can use a high-pass filter, for example, of the sixth order or replace it with a PF with a passband of 300 ÷ 3000 Hz in voice information systems for transmitting messages in transport, at airports, railway stations, etc.

The embodiment shown in FIG. 14 has a significant drawback. There is a loss of information messages at noise reduction frequencies. Such systems are not suitable for high-quality sound reproduction. The implementation options of the signal processing units 2 shown in FIGS. 17, 18 are deprived of this drawback.

On Fig shows an embodiment of a processing unit 2 with the functions of automatic selection of energy and time predistortion of the components of the transmitted message and non-automatic predistortion of signal components to actively reduce thoughts. The system using the processing unit 2 according to FIG. 17 requires preliminary tuning according to the following algorithm. When the signal source 1 is turned off or low, messages are experimentally selected by the nodes 31, 29 for adjusting the levels and phases (delay time, if this delay is longer than the wavelength of the signal) of the multi-band filtered noise signals received in order to achieve the maximum noise reducing effect in the area of the noise reduction space. The node 31 may be implemented as a controlled amplifier or attenuator. The node 29 can be made in the form of a phase shifter or switched delay lines, shown in Fig. 7 in the form of nodes 33 ₁ , 33 ₂ , ... 33 _S and switch 52. After setting up the noise reduction nodes, the system is ready for operation. The selection and optimization of energy and temporary predistortion of message signals is carried out automatically according to well-known algorithms described in the inventions-analogues (Fig.17).

There are other options for implementing systems, for example, with automatic optimization of energy and non-automatic (manual) optimization of time (phase) parameters of message signals. It is also possible that the energy and temporary predistortions of message signals are carried out by means of non-automatic, preliminary, experimental selection of adjustments of the levels and phases of the components of message signals. It is desirable to carry out this setting by disconnecting the received noise signals from 1 input, as well as by disconnecting all kinds of sources of noise and interference. If this is not possible, then the selection of the parameters of the message signals should be carried out with the noise reduction system turned on. The above settings must be made for specific communication channels, equipment and areas of the space for receiving messages and reducing interference and noise. Selected adjustments provide optimal signal predistortion taking into account all sorts of individual properties of the communication channel, the location of the receiving microphones, emitters and parameters of the equipment nodes. For these embodiments of message transmission systems (noise reduction), any significant change in the properties of the communication channel and other elements of the system is not allowed. Otherwise, a new system setup is required. For example, to solve the problem of reducing noise inside an aircraft or to improve the quality of sound messages for air passengers, you can use non-automatic options for implementing systems, since the noise spectra at the point of reception, for example, near the engine and in the passenger compartment, near the passenger’s head, change relative to each other during flight slightly due to possible destabilizing factors determining the stability of the transfer function of the communication channel or signal transmission for active noise reduction. In a car, this transfer function of the communication channel is significantly more unstable, for example, due to the possible opening of windows, as well as the emergence of strong external sources of noise and interference — nearby cars of various designs and with a different spectrum of noise signal. For systems with unstable conditions of relative position of all kinds of elements and unpredictable changes in the transfer function of the communication channel, as well as unpredictable changes in the spectra of signals in the region of the space for receiving messages (noise reduction) with respect to the spectra of the transmitted message (at the output of signal message source 1) and received noise signal near the source or sources of noise, it is necessary to use an embodiment of the processing unit 2, shown, for example, in Fig. 18.

In this embodiment, the selection and optimization of the energy and time parameters of the signal source signal components and signals for active noise reduction is fully automated. To ensure the above conditions for the automation of the selection of signal predistortions for active noise reduction during the low level of message signal components in the corresponding frequency band, new functional relationships have been introduced between blocks 26 and 28, 31, 30, which also contain additional nodes that make it possible to fulfill this condition (on 18 these nodes are not shown. These nodes will be described in detail below).

For a more detailed explanation of the foregoing, we will use Fig. 19, which shows a schematic diagram of one of the possible embodiments of block 2. This diagram will allow a comparative analysis of known and new technical solutions, as well as to explain the essence of the technical solutions of this invention in a convenient, visual way for specialists. and their advantages over analogues.

In the circuit shown in FIG. 19, most nodes are based on an op-amp. The filters 14 at the inputs of the signal processing unit 2 are made in the form of the same type of FS, implemented in the form of three resonant FS with a different tuning frequency and a different schematic and constructive location in the band-pass channels for processing the signal components. On Fig shows the main nodes of the signal processing unit 2 for only one of the many processing bands, as well as one of the possible adders 17, by which the output signals of the processing unit 2 are generated by summing the multi-band processed signals of the message source 1 and received noise signals. In all processing bands, similar circuit units and elements are used.

A feature of the implementation of filters 14, with the help of which the formation of frequency bands of signal processing is carried out, is the distribution of resonant FS along the path of the signal components in the band of their processing. Two of the PFs are structurally located at the input of each of the multi-band signal processing channels, and one is located at the output of this channel. The input resonant filters can be tuned, for example, to the lower and upper frequencies of a given signal processing band, then the output filter must be tuned to the middle frequency of this band. It serves to filter out the products of nonlinear and intermodulation distortions arising in this multiband signal processing channel. The resulting end-to-end frequency response of the band-pass signal processing channel is formed by these three resonant filters and can have, for example, the most flat form. Other options are possible, for example, with more than three, the number of detuned resonance filters or, conversely, with a smaller number, for example, when using one input and one output resonant filters.

When processing the received noise signals arriving at 1 input of the processing unit 2 (Fig. 19), their multi-band filtering is carried out using the chip D 6.

Resonance filters are made on the op amp D 6.2 and D 6.4. At the op amp D 6.1 to D 6.5, signal repeaters are made necessary for decoupling the resonant filters and realizing the possibility of more accurate formation of the required frequency response of the filter 14. Similar circuitry solutions are used to filter the signals at the other inputs of processing unit 2: for multi-band filtering of the source signals supplied to 2 input of processing unit 2 (microcircuits D 1.1 - D 1.4); with multi-band filtering of feedback signals (D 4.1 - D 4.4), which are fed to the additional input of signal processing unit 2.

On the op-amp D 7.3 - D 7.4 and switch SA4, a non-automatic phase (delay time) regulator of the components of the received noise signals is made. Functionally, these elements are equivalent to nodes 32 and 29, for example, shown in Fig.17. By means of the switch SA4, it is possible to temporarily pre-emphasize the received noise signals while generating signals for actively reducing the noise. The signal from the output of switch SA4 is fed to the input of the second controlled amplifier 31. In the proposed embodiment (Fig. 19), this node performs a number of new functions. It is implemented on the VT4 field-effect transistor, which is a controlled key that ensures the transition of the controlled attenuator made on the VT5 transistor from the mode of automatic control of signal component levels for active noise reduction to the mode of non-automatic level formation depending on the current level of message signal components. The control transistor VT4 signal is generated at the output of the inverter D 5.3 comparator, performed on the chip D 5.2. The comparator compares the current signal level of the message with the reference signal level that determines the threshold for switching the comparator. The reference signal is set using the tuning resistor R 38 and is fed to the inverting input of the op-amp D 5.2. The non-inverting input of this microcircuit receives rectified by means of diodes VD 11, VD 12 and smoothed by an integrator made on the elements R 37, C17, a message signal, the level of which is compared with the level of the reference signal of the comparator. When in this analysis band the signal signal level is higher than the predetermined threshold level, the transistor VT4 is closed and the gate voltage of the transistor VT5 does not change and is determined by the residual voltage of the previously charged capacitor C 44. It acts as a memory node. The discharge time constant of the capacitor C 44 is selected based on the available a priori data on the parameters of the transmitted message signals and a specific task. For example, in systems for transmitting sound musical signals of messages, the time constant can be selected on the order of 20 ÷ 60 sec. When the level of the components of the message signals in this analysis band falls below the threshold, the comparator switches, and a signal is generated at the output of the D 5.3 chip, which opens the key transistor VT4. The speed at which the VT4 transistor opens is determined by the time constant R 78, C 42. As a result, the drain-source channel resistance of this transistor decreases from about 10 mOhm to 100 ÷ 300 Ohm over a certain period of time, which excludes the possibility of transients. Capacitor C4 charges quickly enough to the voltage generated at the output of the op amp D 9.4. This voltage corresponds to the current noise level in the region of the message receiving space and in this frequency band, since the level of message signals in this frequency band and at a given time is lower than a predetermined threshold, for example, in a pause between songs or words of a speaker. This control voltage provides the formation of an optimal level of signal components for active noise reduction. The transfer coefficient of the controlled attenuator depends on the voltage at the gate of the VT5 transistor. The operating point of the controlled attenuator is set by the tuning resistor R 82. The dynamic range of the input signal level control is determined by the ratio of the resistors R 79, R 81, as well as the drain-source resistance of the VT5 field-effect transistor. When used as part of a controlled attenuator, for example, transistor KP 103, the dynamic range of level adjustment (at a level of nonlinear distortion of not more than 3%) is about 40 ÷ 50 dB. The maximum dynamic range is achieved at zero resistance of the resistor R 81. As the value of the resistor R 81 increases, the dynamic range of regulation of the level of the input signal of the attenuator decreases. With the asymptotic tendency of the value of the resistor R 81 to infinity, the dynamic range of the signal level control tends to zero. In this case, the automatic implementation of the second controlled amplifier 31 degenerates into a non-automatic one. The controlled attenuator, made according to the circuit shown in Fig. 19, has a significant drawback. At levels of input signals above about 0.3 ÷ 0.5 V, a sharp increase in non-linear distortions of this signal processing unit begins. In this case, the resulting level of nonlinear and intermodulation distortions arising in each band-pass signal processing channel is mainly determined by the nonlinear distortions of the controlled attenuator. To reduce non-linear distortion, the signal from the output of the controlled attenuator (from the drain of the VT5 transistor) is fed to the output filter of the band-pass signal processing channel. This filter is made on the shelter D 8.4, D 9.1, D 9.2. Such a technical solution allows you to effectively filter the products of nonlinear and intermodulation distortion and simultaneously carry out additional amplification of the signal attenuated in the attenuator. This function of the second controlled amplifier 31 is a new additional function, and is also a new independent technical solution in devices with multi-band signal processing. The distribution of filters 14 at the input and output of the band-pass signal processing channels allows to solve the problem of the invention to improve the accuracy of the transmitted message, and this technical solution can be described in the independent claim. The novelty of this technical solution, for example, with respect to the solutions used in the design of radio receivers containing at the input and output stages of radio amplification or intermediate frequencies oscillatory circuits tuned in resonance or detuned relative to each other, lies in the difference not only in the types of filtered signals, but also in the number of filtering bands in relation to the signal spectrum. A radio signal is a carrier modulated by a message signal (for example, an audio or video signal), and this signal is filtered in the entire frequency band corresponding to the spectrum of the radio signal, and in the technical solution described above, the components of the noise signal in several different frequency bands of its spectrum or part of the spectrum are filtered , for example in the low frequency region.

It is clear that a similar technical solution to reduce the level of nonlinear and intermodulation distortion of the components of multiband processed signals can be applied to multiband processing of the signal source 1 messages at input 2 of the signal processing unit 2 (Fig. 19). These signals are first filtered (D 1.1 - D 1.4), then the filtered signals are corrected in phase by means of a non-automatic phase shifter made on the OA D 2.1. Next, automatic correction of the energy parameters of the transmitted message signals is carried out by means of a controlled key (VT1) and a controlled attenuator (VT2). The output filtering of the signal is carried out by means of a filter made on the op amp D 2.3. The phases of the components of the message signals and the received noise signals are adjusted using the tuning resistors R 9 and through the switch SA4, and the levels of these signals are experimentally selected using the resistors R 19, R 87 at the stage of the system pre-tuning. The multi-band processed messages and noise signals arrive through the switches SA5, SA3 to the adders 17 (op-amp D3.4, D 5.4). Multiband processed signals in other processing bands also go to these adders (to resistors R 43, 44, 47, 48). As already noted, various combinatorial combinations of the signals supplied to the adders and the number of adders 17 are possible, for example, when implementing multiband systems. From the outputs of the adders 17, the signals are fed to the amplifier 3 or amplifiers 3, 9 according to the specific structural diagram of the system, for example, shown in figure 2.

It should be noted that the control signals supplied to the transistors VT2 and VT5, functionally depend on the current ratio of the energy spectra of the message signals and signals, respectively, at the point (region) of the message and signals in the field of message reception (noise reduction) and in the noise receiving space signals near a noise source. This allows you to implement an automatic control mode of signal levels of messages or signals for active noise reduction, depending on the relative change in the current levels of signal components, both in the direction of increasing and decreasing these levels relative to each other. To do this, in the circuit shown in Fig. 19, bipolar detection of the compared signals, filtering of the detected signals, necessary for smoothing the rectified pulsed unipolar signals, and summing of bipolar, smoothed signals are performed. The detectors VD 1, VD 3, VD 6, VD 7, and also VD 4, VD 5, VD 8, VD 9 were used as detectors of bipolar signals. To filter the signals, the low-pass filter (26), made on the elements R 36, C 25, was used , C 27, R 58, C 33. The adder is made on resistors R 39 and R 59, which is connected through an additional low-pass filter (R40, C26) to the gate of the transistor VT2. Another group of low-pass filters (26) is made on the elements R 53, C 30, C 29, R 92, C 55, and the adder is made on resistors R 93, R 52, R 94 connected to the drain of the transistor VT4 through a repeater D 9.4.

The nodes of the automatic regulation of the energy predistortion of the signal components (Fig.19) work as follows.

Suppose that a pre-adjustment of this bandpass signal conditioning channel for active noise reduction has already been carried out. As a result of preliminary experimental selection of the phases of the received noise signals, the state of switch SA4 is found, which provides for the formation of antiphase noise-reducing waves for a given area of space and microphone locations for receiving noise signals and emitters, the level of which, for example, is slightly lower than the level of noise waves. This level of noise reduction waves is formed by the regulator R87. For definiteness, we assume that as a result of such a preliminary adjustment, it is possible to lower the noise level in this frequency band of signal processing, for example, by 20 dB. As a result of this, a certain residual noise level remains in the region of the noise reduction space, which is received by the receiving element (microphone) of the probing device 5, is converted into an additionally received noise signal (feedback signal), and is fed to the additional input of the signal processing unit 2 (Fig. 19) . This signal is filtered by filters 14 (op-amp D 4.1-4), amplified by an amplifier made on op-amp D 7.1, and fed through detectors R 57 and isolation capacitors C51, C32 to detectors VD 4, VD 5, VD 6, VD 7. Level The signal rectified on these diodes is proportional to the total level of all sound vibrations at a given time, in a given region of space and in a given frequency band. Suppose that at some point in time in the analysis frequency band there are no message signal components, and the constant voltage level at the diode outputs functionally depends only on the current noise level. If the operating conditions of this system allow it to be simulated as stationary or quasi-stationary, then for any moment the levels of detected signals received in the area of the noise reduction space and detected detected noise signals (at the output VD 8, VD 9) will always be in the same volume the same ratio as far as possible changing the spectrum of received noise signals - proportionally increasing or proportionally decreasing. In this situation, signals detected in different polarity will give, for example, zero. At the output of the adder and, accordingly, the chip D 9.4 will be the same voltage. This voltage can be set equal, for example, to zero by appropriate selection of the amplification factors of amplifiers D 7.1 and D 9.3. Therefore, for such stationary systems, there is no need to automate the process of selecting the energy and time (phase) parameters of noise-reducing signals. In such systems, there is no need to use transistors VT 4, VT 5. They can be excluded from the circuit of Fig. 19, which, in fact, turns into the circuit shown in Fig. 16.

If there are destabilizing factors in the system that somehow change the noise spectrum in the region of its reduction space with respect to the spectrum of the received noise signal, then this system is unsteady and works as follows. Suppose that as a result of changing the properties of the channel in this analysis band, the level of noise components increased, for example, due to a change in the acoustic properties of the noise reduction space region as a result of the appearance of an additional re-reflector of sound waves arriving in phase with the direct wave. An increase in the level of noise components will lead to the fact that the output level of the detectors VD 5, VD 4 in proportion to the level of the additional common-mode wave increases the level of constant voltage. At the output of the adder and repeater (D 9.4), the voltage from zero will increase to some positive value. This voltage, through the open drain-source channel of the transistor VT 4, will go to the gate of the transistor VT 5. The drain-source resistance of this transistor will begin to increase, and the signal level at its drain will begin to increase. The noise reduction signal at the output of the processing unit 2 (at the output of the adder 17 - OA D 5.4) and in the area of the noise reduction space also starts to grow and compensate for the increase in the level of the noise signal, which is a superposition of noise and noise reduction signals (waves). As a result of this compensatory automatic action, the level of the detected signal begins to decrease and takes on a value that provides a balance of noise and noise reduction signals in the region of the noise reduction space at a residual noise level of 20 dB. Obviously, if as a result of destabilizing factors there was a decrease in the noise level below a predetermined level of 20 dB, then at the output of the op amp D 9.4 a negative signal will be generated that will open the transistor VT 5. The noise reduction signal will decrease while in the region of the noise reduction space for a given frequency band again, the noise level will not reach the level of 20 dB, etc. If the phase of the additional interference and, accordingly, the phase of the resulting signal somehow changes randomly, then, in addition to selecting the energetic predistortions, it will be necessary to select temporary (phase) predistortions of noise-reducing signals, for example, in accordance with the algorithm shown in the form of an automatic phase shifter circuit (Fig. .7). On Fig nodes, allowing to automate these actions are not shown, because in analog form the implementation of the circuit shown in Fig. 7 is very cumbersome, primarily because of the need to implement a large number of delay lines, performed, for example, in in the form of series-connected phase shifters (similar to phase shifters D 7.5-4 of Fig. 19), the number of which can be hundreds or even thousands for a number of particular tasks. For these purposes, it is more practical to use digital methods for generating delays and nodes (29, 30, 31).

The main disadvantage of the above option for automating the adjustment of signal component levels for active noise reduction is the need to generate noise reduction signals with a lower (or dual - higher) level than the current level of noise components in each of the signal processing bands. This residual level depends on the range of possible changes in the noise level. This option is advisable to use in quasistationary systems with insignificant relative changes in the components of noise signals at the point of their reception near the noise source and in the region of its reduction space. Otherwise, with large, sharp changes in the spectra of the signals, the system may not have time to track which side it is necessary to carry out energy predistortion of the signals to actively reduce noise.

There are other options for constructing nodes and systems for implementing the function of automating energy predistortions of received noise signals. On Fig shows an additional node that can be included in the break of the circuit (Fig.19) between the diode VD 4 and the repeater D 9.4. This node allows you to exclude from the circuit (Fig.19) the elements: R 91, 88, 89, 90, 92, 93, D 9.3, C 54, 55, and also provide a lower residual noise level, which practically does not depend on the possible range changes in levels, spectra of noise signals at the point of their reception and lowering. In this embodiment, the principle of operation of nodes automating the selection of energy parameters of received noise signals differs significantly from the above. Since the signal level at the input of the circuit shown in Fig. 20 can only vary from zero to a certain positive level proportional to the total total level of all sound signals for a given frequency band, the main new functional purpose of this circuit is to determine the sign or direction, in which should be carried out predistortion adjustment - increase or decrease the level of signals to actively reduce noise.

The structure of this scheme includes the following main nodes. The comparator made on OS D 10.1. It allows you to switch the polarity of the output control voltage of this circuit from minus to plus when the input signal exceeds a certain threshold level. This threshold level is set using the resistor R 104 and is set, for example, at a level of - 20 ÷ 40 dB from the average level of noise signals at the time of system setup. The structure of the circuit (Fig. 20) also includes controlled keys made on field-effect transistors of different polarity of their junctions VT 6, VT 7, the outputs of which are combined by an adder (R 101, 102, 103), which also acts as an integrator due to parallel connection to the resistor R 105 of the capacitor C 59. The signal from the adder is fed to the input of the repeater D 9.4 and then to the transistor VT 4 (Fig.19). The circuit shown in Fig. 20 also contains a key made on the transistor VT 8, which allows you to reset the input of the comparator when the signal level of the messages is higher than the predetermined threshold. To the drains of transistors VT 6, VT 7, which are the inputs of the above keys, the signals are fed from the outputs of the op amp D 10.2 and D 10.5, which have the same level, but different polarity. The polarity sign just determines the direction of the predistortion of the energy parameters of the signals for active noise reduction - increasing or decreasing the level of the generated signal in this processing band depending on the specific, random situation of changes in the signal spectrum. The circuits shown in FIGS. 19, 20 operate as follows.

Suppose that at a given moment of time at the analysis frequencies there are no components of message signals or interference components in the region of the message receiving space — near the location of the receiving element of the probing device 5. At this time, a signal level of zero is formed at the output of the VD4 diode. Accordingly, at the input of the comparator (D 10.1) there is also a zero level. If this level is lower than some positive voltage generated at the inverting input of the op-amp D 10.1, then a negative signal level will be generated at the output of the comparator. This level determines the state of the key transistors. Transistor VT6 is open, and transistor VT7 is closed. Zero input signal level (from VD4 output), through integrating chains (R 55, 52, С 50, 29), repeater (D 10.2), key (VT6), adder-integrator (R 101, 105, С 59), repeater (D 9.4), through the public key - transistor VT4, goes to the control input of the second controlled amplifier 31 - the gate of the transistor VT5. This zero control level corresponds to a certain drain-source resistance of the VT5 transistor and the corresponding transfer coefficient of the controlled attenuator (for example, the transfer coefficient of the attenuator is set at -20 dB relative to the input signal using the resistor R 82, 81, 79).

If the zero level of interference in the area of their reduction space is due to the absence of the sources of these interference themselves, for example, due to the fact that the engine is turned off or the interference is cyclical, then appearing, then disappearing, then at the first input of signal processing unit 2 ( Fig. 19) the level of the received noise electric signal will be equal to zero.

If the zero, or more precisely, very low noise level in the region of the message receiving space is a consequence of the above process of active noise reduction, this situation corresponds to optimally selected adjustments of the phases and levels of signal components for active noise reduction. This state in the system will persist indefinitely while it is stationary.

If changes occur in the system, for example, additional noise signals appear, the components of which fall into this signal analysis and processing band, or the transmission coefficient of the noise signal channel for the given frequencies of their analysis changes, the noise and noise reduction signals will be unbalanced by levels and, in the general case phases. For simplicity, we will assume that, for example, an in-phase with respect to the noise waves proper interference of the same frequency has appeared. As a result of the superposition of this interference, the imbalance of noise and noise reduction signals is determined only by the mismatch of the levels of these signals. The level of the resulting interference has become greater than the level of the noise reduction signal. As a result of this change, a positive signal will appear at the output of the VD4 diode, the level of which will increase in proportion to the level of the additional common-mode noise. This overgrowing signal, the slew rate of which is determined by the integrator time constant (R 55, 52, C 50, 29), will be supplied through the above elements (D 10.2, VT 6, D 9.4, VT 4) to the gate of the transistor VT5 of the controlled attenuator of the second controlled amplifier 31. This positive voltage changes the state of the transistor VT5 - it closes. The gain of the controlled attenuator increases from - 20 dB, for example, to - 15 dB. In the area of noise reduction space, the level of noise reduction waves also increases to the state of noise (with a 5 dB higher noise level due to additional common-mode noise) and noise reduction waves.

An important point in the operation of the circuit shown in Fig. 20 is that positive signals are generated at the non-inverting input of the comparator (D 10.1) with an additional delay determined by the time constant of the integrator R 108, C 60. This action is in many respects similar to generating signal delays (via delay line 54) in the circuit shown in Fig.7. Therefore, with common-mode additional noise, the system manages to self-adjust and suppress the increase in noise level before a positive signal level at the input of the comparator exceeds a predetermined threshold. If the integrator time constant R 108, C 60 is chosen to be greater than the system self-regulation time constant (taking into account the delay time of the signals in the communication channel), then the comparator will not have time to switch. It should be noted that the drain-source resistance of the transistor VT8 is great in the simulated situation, since it is closed by the control voltage from the output of the op-amp D 5.3, similar to the transistor VT1, which performs the function of a bandpass threshold noise suppressor of the electrical signals of the message source. This is also a new technical solution. If the current level of message signals exceeds the threshold set by the tuning resistor R 58, then the signals at the output of the comparator (D 5.2) and inverter (D 5.3) will be reversed, and the transistor VT8 will open and the drain-source resistance will bypass the input of the comparator D 10.1, until the signal level of the messages in this analysis and processing band again falls below a predetermined level.

We will simulate a different situation. Let the interference level decrease as a result of the appearance of additional interference of the same frequency, but the opposite phase with respect to the main interference. Then the level of the positive signal at the output of the VD4 diode at the input of the circuit (Fig. 20) will also begin to increase due to a higher level of signals for active reduction of the thought relative to the level of total interference. Suppose that at this time the signal level of the messages at the analysis frequencies is lower than the predetermined one, and the transistor VT8 is closed by the control signal from the output of the op-amp D 5.3. In this situation, the growing positive level at the gate of the transistor VT5 of the controlled attenuator will close it, increasing the gain of the attenuator, and in the area of the noise reduction space, the level of noise reduction signals will begin to increase, which, in turn, will lead to a further increase in the level of residual noise. This process will increase until the constant voltage level at the non-inverting input of the comparator exceeds a threshold value. When this happens, the comparator will switch, and its output voltage from the minus will change to plus (the op amp is bipolar). Accordingly, the state of key transistors will also change. The VT6 transistor "softly" closes, and the VT7 transistor "softly" opens (due to the operation of the integrators R 99, C 57, R 100, C 58), and on the adder-integrator R 102, 103, C 59 also begins to form softly reverse polarity voltage. Thus, it recognizes the direction in which direction it is necessary to carry out the correction of the transfer coefficient of the second controlled amplifier 31 (in this circuit design, the attenuator). Negative control voltage transistor VT5 begins to open, and the gain of the attenuator decreases until a certain balance of noise reduction and noise signals occurs. This will happen until the voltage at the non-inverting input of the comparator again falls below the threshold. When this happens, the comparator will switch again, and this process will be repeated again. It will have the character of oscillations, in the general case with a different frequency for each of the signal analysis bands. In this case, the level of noise reduction signals will vary between some boundary noise levels. As a result of this situation model, there will also be some average residual noise level, which is undesirable. In order to get rid of this effect, it is possible to install a decoupling diode in front of or after the integrator resistor R 108, which will allow to maintain a high potential at the non-inverting input of the comparator until it is reset by means of the VT8 key, or self-discharge of the capacitor C60 due to voltage leakage through the input resistance of the comparator, the drain-source resistance of the VT8 transistor, the inverse resistance of the decoupling diode, or through the resistance of the capacitor C 60. Condenser discharge time constant S 60 can be formed by connecting a resistor in parallel to this capacitor. This resistor and isolation diode are not shown in FIG.

All the above time constants are selected based on the specific conditions of the problem (based on the range of the message channels, noise and noise reduction signals, the propagation speed of the signals in the channels, the frequency of the processed signals).

If in this frequency band of signal analysis and processing signal components of signal source 1 of the message appear, the level of which is higher than the predetermined one, then during the time when their level exceeds this threshold, the formation of signals for active noise reduction is carried out from memory due to the previously charged capacitor C 44 At this time, the transistor VT4 is closed, and whatever signals are generated at this time in the circuit shown in Fig. 20, they do not enter the gate of the transistor VT5 of the controlled attenuator. During this time, the transistor VT8 is open, and at the non-inverting input of the comparator D 10.1 there is a signal whose level is close to zero.

When the level of the components of the message signals becomes lower than the predetermined level, the transistor VT8 is closed, and the process of automatically correcting the level of noise reduction signals is repeated according to the above algorithm. The system automatically reduces the noise level to a predetermined level close to zero. Moreover, the residual noise level will be determined not only by the algorithm for automatically generating noise-reducing signals, but also by distortions that occur mainly in emitters (4, 10). The above process also occurs when not only the level of the resulting noise signal changes, but also the phases of its components. Tracking and optimization of the phase (time) parameters of the signals for active noise reduction can be carried out in accordance with the well-known algorithm (Fig.7). Choosing the system self-adaptation time constants accordingly during optimization (selection and "capture") of the phase and level in a given signal processing band, these processes can be carried out simultaneously in time. A variant of the sequential implementation of these actions in the form of an iterative process of successive approximations to the optimal predistortions of the received noise signals, which provide the greatest noise reduction effect, is also possible. To exclude conditions under which the system can self-excite, it is necessary to appropriately choose the time constants of regulation of signal parameters both within each of the processing bands, and in a certain ratio between the processing bands. Due to the partial mutual overlap of these bands (due to the imperfectly rectangular frequency response of these processing bands), the resulting signal is determined not only by the level and phase of the component passing through the corresponding filter, but also by the frequency components passing also through other band-pass filters (naturally, if they open at this time). The contribution of these additional signals due to the “tails” of the frequency response of the filters 14 may be insignificant, but in a number of applied problems they must be taken into account when generating the output signals of the signal processing unit 2. Therefore, it is desirable to carry out all possible adjustments quite smoothly with respect to the slew rate of the signals in the processing bands, similar to the actions of sound engineers manually performing similar actions, for example, when mixing signals or when working with the equalizer. Otherwise, abrupt switching of the band signals will lead to transients that will be noticeable by ear and will be perceived, for example, as clicks or pops when transmitting audio messages. For this, the automatic control time constants must be at least 5-10 signal periods of the corresponding frequencies for the keys VT1, VT4, VT6, VT7, VT8. For controlled attenuators VT2, VT5, these time constants should be even longer (several times or even several orders of magnitude) to exclude spurious modulation of message signals and signals for active noise reduction by the pulsating, poorly filtered, control voltage supplied to the gates of these transistors. By the way, the output filters of the bandpass signal processing channels, in addition to reducing non-linear and intermodulation signal distortions, will partially reduce this negative modulation effect. Even higher should be the time constants of the integrating chains R 108, C 60 due to the above features of the operation of the circuit shown in Fig.20. It should be noted that in noise reduction systems a number of elements can be eliminated in comparison with message transmission systems under noise conditions. For example, you can exclude the transistor VT8 and its connection with the op-amp D 5.3. In this case, it is also possible to dispense with the diodes VD5, VD4 by removing the equivalent control signal from the diodes VD6, VD7, and in embodiments of automatic message systems and from the diodes VD1, VD2 in accordance with the structural diagrams shown in Figs. 14, 15, 18 Given the opposite sign of the polarity of the signals rectified by the VD4, VD5 diodes, it is necessary to put an inverter at the input of the circuit (Fig. 20), which changes the sign of the input signal, or otherwise modify this circuit, for example, by changing to the reverse voltage that determines the threshold of the comparator D 10.1 . There are other options for automating the processes of selecting optimal signal pre-emphasis in block 2 of their processing.

As already noted, the main advantage of the above-described noise reduction and message transmission systems with respect to analogs in which a noise signal is received in the region of its reduction space is the removal of the noise signal receiving point from the point (region) of the noise reduction space and obtaining due to this action higher decoupling of noise and noise reducing signals and significantly more effective reduction of noise level in the required area of space. For example, near the human auditory organs. On Fig shows the nodes of one of the possible variants of the circuit noise reduction analogue. In this embodiment, noise reduction signals are generated at times when there are no signal components of the message source 1 in the corresponding analysis band by means of a key made on the transistor VT5. For simplicity, FIG. 19 shows a non-automatic variant of generating energy (by means of resistor R 64) and temporal (by means of resistor R 95) predistortions of noise signals received in the region of the noise reduction space. It is clear that automatic adjustment of the levels and phases of these signals can be obtained using the above-described examples of nodes that automate these actions, for example, by switching on a controlled attenuator between the VT3 transistor and the D 8.1 op-amp, made like attenuators of a band-pass channel for processing message signals (VT 2) or received noise signals (VT 5), and the automatic phase shifter according to the scheme shown in Fig.7. For this technical solution, the noise reduction effect is very small and amounts to about several dB. In the above embodiment, it can be tens of dB. This is the main advantage of the proposed technical solutions over analogues. Therefore, the sign "the place of reception of noise signals is chosen closer to the noise source than the region of the message receiving space" is an essential sign necessary and sufficient to solve the problem of the invention to improve the accuracy of the transmitted message from one point in space and time to another point in the presence of interference and noise with A known location of the source or sources of noise. This feature, as well as the above-described condition for the implementation of energy and temporary predistortions of received noise signals at a zero level of electrical signals of a message source 1 or at a level of electrical signals of a message source 1 below a predetermined threshold, memorizing energy and temporary predistortions providing an effect of reducing the noise level in areas of the message receiving space should be included in the restrictive part of the first claim.

The process of processing the signal source 1 messages is in many ways similar to the processing of noise signals. It is, as it were, a priority inverse to the actions of the noise reduction process. Let us explain these processes using Fig. 19.

The signals of the message source 1 and the signals received by the probing device 5 in the region of the space for receiving messages are filtered by filters 14 with the same frequency response performed on the op-amps D 1.2, D 1.4 and D 4.2, D 4.4, amplified by amplifiers made on the op-amps D 5.1 and D 7.1 are bipolarly detected, smoothed and summed by the above elements of the circuit.

For any point in time, the control signal generated at point D (Fig. 19) carries information about the current ratio of component levels for each of the analysis bands of the transmitted and received signal of message source 1. For example, if the zero level of the control signal (at point D) corresponds to the situation when there are no noise components in the analysis band and the presence of signal source 1 components, the level of which in the region of the message receiving space exactly corresponds to the required level previously set by the tuning resistor R 25 at the above stage presetting this band-pass message channel in a particular message channel, then with any changes in the system, the corresponding In this way, the control signal (at point D) will also change. The preliminary adjustment of the transmission coefficient of a given band-pass signal transmission channel can be made according to the types of signals convenient for this, for example, noise-like or sinusoidal-like forms.

Suppose that in-phase components of an interference signal appear that are a consequence of, for example, multipath of a communication channel. Due to this, the signal level at the output of the detectors VD6, VD7 will be modulo higher than the level at the output of the detectors VD1, VD5 and at point D the signal level from zero will begin to decrease to a certain value proportional to the level of common mode noise. This negative voltage will go to the gate of the transistor VT2. It will begin to open so that the signal of the message source 1, passed through the input filter 14, the phase shifter D 2.1 and the key VT1, begins to decrease. In the area of the message receiving space, the level of sound vibrations of the message signal will also begin to decrease until the voltage at point D becomes zero again, and the total total energy of the message signals plus noise at the point of receipt of the message is equal to the energy of the message signal without this interference in this analysis strip. This is a fundamentally different version of the implementation of the method of adjusting the signal level of the message source 1 in comparison with the principle of operation of the analogue - the Blaupunkt San Francisco system, in which the signal signal level is always raised in the space of its receipt with an increase in the average noise level.

If the interference is out of phase with the components of the message signals, then the gate of the transistor VT2 receives a voltage of positive polarity, and the transistor VT2 closes, and the level of the sound signals of the messages in the area of the receiving space increases exactly as much as the level of the out-of-phase interference components. As a result, the listener in the area of the message receiving space perceives the components of the sound signals of the messages and the components of noise, the total level of which corresponds as closely as possible to the level of the components of the signals of the message source or the level of the original sources of sounds, if the recording of these sources and the formation of signals in source 1 were carried out in accordance with the method in the patent RU, A, 2145446. Moreover, if at the output of the signal signal source 1 or at the input of the processing unit 2 set the level control, then COROLLARY this regulator listener can choose for themselves comfortable listening level signal source 1, such as at 80 or 110 dB. The volume level chosen by him will be automatically monitored and maintained by the system taking into account, for example, some residual noise level. Such operation of the signal processing unit 2 is equivalent to the implementation of an automatic version of the well-known Wiener filter. For example, if at a given time at certain frequencies low-frequency music should be played, created by a bass guitar at, for example, 80 dB, and at the same time at this frequency there is a noise signal not suppressed by the system of the same frequency, same phase and with the same level, as a result of the above adjustment, the state of the transistor VT2 will be such that the signal of the message source 1 will be attenuated by 80 dB, that is, to the zero level. At this time, the listener will hear the noise components at these frequencies instead of the bass components of the message signal source 1, and not just the sum of the noise components and the signal signals amplified in proportion to the noise, as in the San Francisco CD 70 system of Blaupunkt. From the point of view of the provisions of the modern theory of information, the listener is absolutely indifferent that at a given moment of time, at specific frequencies, is an emitter of sounds. A loudspeaker speaker (4, 10) emitting bass sounds, or sounds from an exhaust pipe, for example, a car in which it listens to music. It is important for him to hear the same sequence of sounds of certain frequencies, of a certain level and phased in a certain way relative to each other, as the original source of these sounds. Therefore, the level of the components of the message signals should be adjusted taking into account the levels and phases of the components of interference and noise at a point (region) of the space for receiving the message. It is clear that if the area of the message receiving space has a predetermined size and shape, then, taking into account this a priori information, it is advisable to limit the frequency band of automatic regulation of the energy and time pre-emphasis of the signal source 1 messages similar to the above-described restrictions on the frequency band of the signals for active noise reduction. Otherwise, the optimally selected pre-emphasis for the point in space where the receiving element of the probing device 5 is located can be very different for the points of space where at the given moment the hearing organs (head) of a person are located.

Part of the nodes and elements of the circuit shown in Fig. 19 are a new technical solution for multiband threshold noise reduction of the electrical signals of the message source 1. The structure of this noise suppressor includes multiband channels of threshold noise reduction, performed on OS D 1.1-4; D 2.1; VT1; D 3.1-4; D 5.1; VD 2, VD 2; D 5.2-3, as well as a number of other elements of the circuit (Fig.19). The phase shifter (D 2.1) and the output filter (D 3.1-3) can be excluded from the noise suppressor circuit. This device allows you to effectively suppress the noise of the electrical signal signals in each (or part) of the signal processing bands when the level of the components of the input signal of the noise suppressor is lower than the predetermined value (threshold). This threshold is set by the tuning resistor R 38 and can be set, for example, at the level of - 20 ÷ 70 dB from the maximum level of the input signal. The threshold level may be the same or different for different signal processing bands of the message source 1. Such noise suppressors can be equipped with various radio systems and devices, tape recorders, players, mixing consoles, etc. Structurally, they can be implemented in the form of sound cards (PC software) or consoles for block acquisition of equipment. Such noise suppressors can be very useful for processing and restoration of old records, records, or used for processing sound signals by special services, for example, when decrypting telemetry information or negotiating airplane pilots that have crashed. The effectiveness of the noise reduction effect is mainly determined by the number of bandpass channels for reducing noise (number of filters 14) and the time constant for turning on and off the band channels (C 56, R 37, C 17, R 14, C 7). The noise-reducing effect of this squelch also depends on the current spectrum of the signal source 1 of the message, and is also determined by the noise characteristics of the output operational amplifiers (D 3.1-4), the ratio of the resistor R 21 and the drain-source resistance of the transistor VT 1 and can make noise suppression for each band - 100 ÷ 150 dB. For example, if the number of signal processing bands in the squelch is one hundred, and the signal spectrum represents a sinusoidal-like signal with a frequency of 10 kHz, then all processing bands, except the one into which the signal has entered, will be closed, both above and below this frequency, and the resulting residual noise level will be reduced one hundred times (40 dB). For comparison, the well-known dynamic noise suppressors, in which the cut-off frequency of the filter is set, as a rule, at frequencies of the order of 3-6 kHz and broadband threshold noise suppressors in this situation will not produce a noise-reducing effect, and the effect of various Dolby noise canceling options with signal processing during phonogram recording and its reproduction will not exceed 15 ÷ 25 dB. Thus, the proposed option of noise reduction of the electrical signals of the message source 1 is devoid of the main disadvantages of well-known noise reduction devices in the form of partial suppression of the useful signal and the inevitably high nonlinear and intermodulation distortions of the signals during their nonlinear processing during compression and expansion of the dynamic range during the double processing of these signals in compander noise reducing devices. For example, this squelch is practically devoid of a noticeable “puffing” effect characteristic of the Panasonic NFD threshold squelch, the DNL squelch proposed by Philips, or the well-known domestic Beacon squelch, because the process of switching on the band channels does not occur simultaneously, at different frequencies where the signal components of message source 1 appear. The energy of the noise “tails” that create the above unpleasant effect is determined by the energy of the noise that has fallen into the frequency band of the signal component, which is, for example, one hundred times smaller than the entire band of sound signals. Therefore, the residual “tails” of the noise become almost invisible to the ear. These noise suppressors can be used in HIFI systems. Since this technical solution is part of the patented invention and solves the problem of the invention, it can be described in the independent claim. The requirement for unity of invention is not violated.

The above explanations of the principle of operation of the band-pass channels for processing message signals and received noise signals allow us to understand in detail and in dynamics the principle of operation of the entire system and the corresponding methods for transmitting messages of any physical nature in interference channels. In methods and systems with non-automatic adjustment of energy and temporary signal pre-emphasis, the process of active noise reduction is carried out all the time in accordance with the previously set (remembered) adjustments of these predistortions. If the channel parameters are unstable, and constant adjustment of the pre-emphasis of the received noise signals is required, then it is carried out only at those times when in this analysis band the level of the components of the message signals is lower than the predetermined level, and these signals do not interfere with the automatic tuning of the noise reduction system. When the level of the components of the message signals is higher than the predetermined one, the previously found pre-emphasis adjustments of the received noise signals are stored, providing a noise reduction mode in memory. The energy pre-emphasis in each of the signal processing bands can be stored by means of the capacitor C44, as shown in Fig. 19, and the temporary pre-emphasis can be stored (fixed), for example, by means of a relay that cuts off the power supply to the electromotor of the electromechanical embodiment of the controlled switch 32 (Fig. 7) or its analogue in the circuit shown in Fig.19 - switch SA4. There are other technical solutions for the implementation of controlled phase shifters using multiplexers that switch signals, like switch SA4, with different delays to their output. In this case, the transfer of the state of the multiplexer to the memory mode can be carried out by fixing the control pulses. In this case, message signals are generated in the area of the space of their receipt according to well-known algorithms in accordance with the structural diagram of the system.

When implementing such systems and methods for transmitting messages, an important circumstance is the need to take into account the delays of a number of the above-described processes for regulating predistortions regarding message signals. In these systems, it is desirable to synchronize the times when the signals exceed the threshold (or thresholds) with the times of switching keys (VT1, VT4). If you do not take appropriate measures, then as a result of delays in the development of signal switching process controllers, a part of information messages may be lost. Indeed, in order to generate control signals for switching, for example, transistor VT1, a certain time is required, determined by the time constant of the elements R 96, C 56, R 56, C 17, R 14, C7. Their use is caused, on the one hand, by the need for stable operation of the comparator (D 5.2), especially in low-threshold analysis modes of the signal level, and the exclusion of frequent switching on of the corresponding band channel due to noise, on the other hand, the integrator R 14, C 17 serves to form the so-called "soft" or smooth switching process of the field effect transistor VT1. If this is not done, then during sudden switching of this transistor, transients of the “click” type may occur, which are typical, for example, for integrated analog switches running on K176KT1 type microcircuits.

The above condition for the synchronization of processes can be realized if we additionally delay the message signal in each band channel in the interval between the input filter 14 and the above-mentioned key transistor VT1. A delay can be generated by several phase shifters in series. The delay time of these phase shifters must correspond to the time constant of the formation of signals that control the switching of transistors VT1 in the corresponding signal processing bands. To exclude temporary distortions of the components of message signals, the delay of these components must be made the same or different, taking into account temporary pre-emphasis of signal components, when temporary distortions in the communication channel are possible. Implementation of this condition will ensure that all components arrive in phase in the area of the message receiving space and minimizes information loss. This condition can significantly improve the quality of signal processing in a multi-band threshold noise suppressor. When implementing analog versions of processing units 2 or multiband threshold noise suppressors, it is advisable to replace the controlled attenuators made on the VT2 transistor with specialized microcircuits - controlled amplifiers that have significantly lower non-linear distortions and higher other parameters. For these purposes, you can use, for example, the SSM2164 chip, consisting of four voltage-controlled amplifiers. The gain control range of this chip (relative to 0.775 V) is from -94 to +20 dB with a slope of control of the signal level of 33 mV / dB. The output stage can operate in class A mode (non-linear distortion coefficient K _G = 0.15%) or AB (K _G = 0.3%). Instead of discrete diodes, you can use the K157DA1 chip - a two-channel, half-wave amplitude detector combined with a signal amplifier. In HI-End class systems, it is desirable to use the SSM2018T chip with even higher technical parameters as a controlled amplifier. For example, the coefficient of nonlinear distortion of this microcircuit in modes A and AB does not exceed 0.04%. As the operational amplifiers can be used TL chip 074 (four OS in a single housing, R _T = 0.003%, f _W = 15 nV / Hz) or NE5554 chip A having still higher settings. With a sufficiently large number of filters 14, made on high-Q PF, upset relative to each other with additional filtering of the processed signal components at the outputs of the multiband channels for their processing, the level of nonlinear and intermodulation distortion of the signals in these processing bands when using these microcircuits will not exceed 0.003 ÷ 0.005% and will mainly be determined by distortion of the filter (D 2.5), repeater (D 2.4) and adder (D 3.4). Filter D 2.3 will effectively filter all possible products of nonlinear and intermodulation distortion appearing in previous signal processing stages. These microcircuits can be used in HI-End class multi-band threshold noise suppressors. When implementing processing units 2 in the form of multiband threshold noise suppressors or noise reduction systems using modern digital signal processing systems, it is possible to implement super-HI-End class systems, which with great reason can be attributed to the class of Formula 1 systems, if you adhere to the classification, proposed by Alpina. The number of signal processing bands in these systems can be practically any due to the possibility of building up processors or a PC for signal processing. For example, one of the processors or a PC may be responsible for synchronizing the above-described processes of multi-band signal processing and for data storage. Other processors or PCs can carry out the above steps to process the signals in the desired number of sub-frequency bands of audio signals. In such systems, it is possible to carry out an almost component-wise analysis of signals, correct them, as well as filter out noise components and effectively actively reduce them. Systems of this level of complexity relate to machines of the highest form of organization. They were invented and synthesized not as a result of a haphazard search for “successful” technical solutions, but as a result of a systematic approach to solving the corresponding problems in strict accordance with the provisions of information theory. For example, the basic idea of using multiple bands for signal processing directly follows from well-known formulas - Fourier series and integrals, as well as from well-known properties of integrals that the integral of the sum of functions (signal components) is equal to the sum of the integrals. These basic principles of information theory can be represented from formal mathematical symbols in equivalent circuitry form - in the form of a set of signal processing channels combined by an adder or adders 17. It is clear that by increasing the number of these strip channels, in the limit, we can switch to component analysis of signals, in shape close to sinusoids. Therefore, the optimization of energy and temporal predistortions of these sinusoidal signal components gives an exhaustive - full-parameter analysis and processing of these signals. There is simply nothing more to optimize and something logically justified to do with these signals. Therefore, the above basic principles of operation of such systems clearly correspond to the new fundamental law of nature - the law of minimizing the growth of entropy (communication system). These are machines with feedback and special processing of transmitted and received messages (or noise signals). The work of these systems or machines is targeted, occurs according to the above algorithms and allows directly in the process of transmitting message signals to study the properties of these signals, the properties of interference (their spectra) and the properties of the channel of the communication system, generate current control signals and achieve the main goal - to transmit from one point in space and time to another point (space) message with the least possible distortion and with minimal loss of information. Only from these positions can we create new, promising technical solutions, for example, in the field of household radio equipment.

As an illustration to the foregoing, we will consider a number of particular, urgent problems directly related to the task of the invention, and show how they should be solved using the provisions of the new fundamental law of nature.

As already noted, the above methods and systems for transmitting messages of any physical nature include actions to enhance the electrical signals of messages and, accordingly, devices designed to implement these actions are amplifiers 3, 9. At present, a large number of technical solutions for amplifying electrical signals are known. . Such devices, for example, can include all kinds of amplifiers-correctors, microphone amplifiers, normalizing amplifiers, power amplifiers of various classes. At the same time, disputes still arise periodically, both among specialists and experts evaluating new developments and industrially manufactured products, and among a fairly large army of music lovers, about the advantages and disadvantages of certain technical solutions or the used elemental base when amplifying electric signals, and first of all, when amplifying signals in power amplifiers. In the vast majority of cases, these discussions are conducted at the level of emotions from listening to sound signals on various equipment. As a result, conclusions are often made that are in no way based on the provisions of modern science. For example, the feasibility of using negative feedback in amplifiers is questioned. Thus, the basic, fundamental principles of cybernetics, electronics, and radio engineering are called into question. In defense of these provisions are experts competent in these matters. For example, in the journal Radio, No. 4, 2003, pp. 16-19, S. Ageev's article "Designing Amplifiers with a General OOS" was published. This article outlines the main points that should be given attention when designing high-linear amplifiers with environmental protection. We will quote a number of these main points and then comment on them.

"Recall that the main reason for the invention of amplifiers with OOS by Harold Black in 1927 was precisely the need to increase the linearity of amplifiers ...

H. Black studied the problem of increasing the linearity of amplifiers since 1921. It was he who developed almost all the known methods for distortion compensation, in particular, distortion correction by the so-called direct coupling, as well as distortion compensation by summing the distorted output signal with an out-of-phase signal distortions. These measures, of course, yielded results, but they were not enough.

The cardinal solution to the linearity problem was precisely the invention of amplifiers with OOC and, most importantly, their correct practical implementation, which was impossible without the creation of an appropriate theory ("there is nothing more practical than a good theory!"). The first step in the construction of the theory was made by Harry Nyquist, who found a method used today to determine stability even before closing the OOS loop based on the type of frequency response and phase response of an open system (Nyquist hodograph).

However, not all so simple. Despite the simplicity and apparent obviousness of the operating principle of the OOS, in order to actually obtain the advantages that can be achieved with its use, it was necessary to create a very extensive feedback theory, which was not at all reduced to ensuring stability (lack of generation). Its construction was almost completed by an outstanding American mathematician of Dutch origin, Henrik Wade Bode, only by 1945. To understand the real complexity of the problems, we note that even Black’s first patent for an amplifier with OOS, which describes far from any problems, has the volume of a small book - it has 87 pages. By the way, all H. Black received 347 patents, a significant part of which is associated with the implementation of amplifiers with environmental protection. Compared to such a volume of work, all the claims of modern "subversive fundamentals", who did not create anything close in level and often even never read (or did not understand) the works of Black, Nyquist and Bode, look at least overly self-confident. Therefore, the question is not in the use of environmental protection (in reality it is always there, just not always explicitly), but in the fact that this use is competent and brings the desired result.

So, what of the "not described in textbooks" you need to pay attention to when designing and evaluating circuitry of amplifiers with environmental protection?

First, recall that in the formula for the transmission coefficient (transfer function) of a feedback system

complex numbers and functions appear, namely:

- комплексный коэффициент передачи (передаточная функция) цепи с ОС;

- complex transfer coefficient (transfer function) of the circuit with the OS;

- комплексный коэффициент передачи (передаточная функция) исходного усилителя.

- complex transfer coefficient (transfer function) of the original amplifier.

To get the correct calculation results, it is necessary to conduct according to the rules of arithmetic of complex numbers, which even the authors of textbooks often forget.

The next moment. It is important to understand that the OOS cannot reduce the absolute value of distortions and noise brought to the input of the products compared to the situation when the OOS loop is open and the signal levels at the output are the same in both cases ... Note also that with a closed OOS loop, the distortion products especially high order, such as “teeth” of switching the shoulders of the output stage, are similar to high-frequency input signals, and the input low-pass filter cannot help here. That is why, in order to prevent a catastrophic expansion of the spectrum of intermodulation distortions upon the introduction of OOS, it is extremely desirable to ensure a faster decay of the envelope of the spectrum of distortion products without OOS than the decay rate of the loop gain. This condition, unfortunately, is not only little known (Bode only hints at it, considering it obvious), but it is also extremely rarely fulfilled. For the same reason, the frequency correction introduced for stability should not lead to a deterioration in the linearity of the amplifier in the entire frequency range, up to the frequency of unity gain, or even slightly higher. The most obvious way to achieve this is to perform the correction so as to directly reduce the input signal ...

The aforementioned explains the desirability of a large margin of linearity in the cascades preceding those where the main decay of the frequency response is formed - in amplifiers with OOS this is necessary, first of all, in order to prevent a significant expansion of the spectrum of distortion products.

To increase the linearity of input stages, it is most efficient to use local OOCs, which simultaneously provide the necessary decrease in frequency response and increase linearity (for example, with inductors in emitter circuits of input stages). The frequency-dependent local OOS can reduce the loss of depth of the overall OOS in the working frequency band.

Thus, when developing an amplifier with OOC, it is necessary to ensure an acceptable (at least not worse than a few percent) linearity and better stability of the characteristics without OOS precisely in the frequency region where the loop gain is large. A number of measures to improve the linearity at low and medium frequencies (for example, the introduction of the so-called servo coupling in a cascade amplifier) simultaneously leads to a deterioration in the stability of characteristics and (or) a decrease in linearity at the high frequencies. Therefore, their introduction into amplifiers with environmental protection is impractical.

In the case of using local OOS, to obtain good results, it is necessary to optimize their frequency characteristics, since each of them not only increases the linearity of this cascade, but also reduces the loop gain in the overall OOS chain. This task is not trivial, one cannot do without very accurate computer simulation. As a rule of the first approximation, we can assume that close to the optimal option is one in which the contribution of all stages to the resulting distortion of the amplifier with OOS (with a closed OOS loop!) Is approximately the same.

Further, for amplifiers with common feedback, the absence of dynamic tracking interruptions in the OOC circuit is critical. This means that dynamic non-linearities are unacceptable, leading to abrupt changes in the characteristics, for example, due to the appearance of grid currents in the lamps when the signal is supplied through an isolation capacitor. If such phenomena cannot be ruled out for any reason, it is necessary to take measures to level their influence in the frequency regions where the loop gain is small (especially in the frequency region of a single gain), using, for example, local OOS ...

The splitting of the general OOS loop into several local loops, advocated by a number of authors, despite the simplification of design, is not practical. The coverage of more than one cascade in the amplifier as a local feedback, as shown by Bode, leads to the loss of potentially achievable linearity. For example, two cascades connected in series with a 30-dB local OOS will have obviously worse linearity than these two cascades covered by a general OOS 60 dB deep in the same frequency band ...

Another frequency that is missed is a very important realizability factor - the design of cascades covered by feedback. It should ensure the absence of spurious peaks in the frequency response and bandwidth, forcing stability to artificially lower the performance of the amplifier as a whole. The presence of spurious peaks in the frequency response dramatically reduces the level of environmental protection attainable without self-excitation ...

Summing up the brief review, we note that any design is a set of compromises, so it is very important that the solutions used are mutually interconnected, and the design is a single whole. "This article describes other points that are advisable, according to author, pay attention when designing amplifiers with OOC.

The following can be noted as comments on the above recommendations when designing and evaluating the circuitry of amplifiers with environmental protection.

Of course, the amount of work that was done in this area of technology by H. Black, and the high level of scientific work by G. Nyquist and H.V. Bode, are very impressive. One cannot but agree that "there is nothing more practical than a good theory!" and that the solutions used should be interconnected, and the design of the amplifier should be a single whole. At the same time, it should be noted that other outstanding scientists subsequently dealt with similar problems. The great American mathematician and philosopher Norbert Wiener made a great contribution to the comprehension and systematic solution of such problems. In his famous book "Cybernetics" he was the first to set forth a fundamentally new look at solving such problems, and first of all, tasks related to the automation of control processes and the synthesis of the most advanced machines or systems. The main feature of this approach was a new model that describes the basic principles of operation of these machines or systems.

цепи ОС, и более того, максимально возможно быстро, стараться его подстраивать под текущие изменения передаточной функции

исходного усилителя или наоборот (что даже лучше) осуществлять предыскажения передаточной функции

исходного усилителя с тем, чтобы комплексный коэффициент обратной связи

оставался неизменным. Реализация подобного подхода и выполнение этих действий может дать наилучшие результаты по "взаимной увязке между собой применяемых решений" с точки зрения современной научной методологии - кибернетики. Бессистемно используя общеизвестные рекомендации по "грамотному" применению ООС, невозможно синтезировать усилитель высшей формы организации, класса качества, группы сложности и т.д. Для этого в состав усилителя должен входить новый по функции и весьма сложный узел - блок 2 обработки сигналов. С его помощью можно (вместо предварительного, малоэффективного компьютерного моделирования корректирующих цепей усилителя) провести в реальном масштабе времени, на реальном усилителе, полнопараметрический анализ входных и выходных сигналов усилителя, сформировать оптимальные предыскажения усиливаемых сигналов (например, в каждом каскаде усиления), сформировать компенсационные сигналы для активного подавления внутри усилителя продуктов нелинейных и инермодуляционных искажений. Понятно, что во времена Н.Винера идея использования ЭВМ для повышения линейности усилителя выглядела бы слишком смелой. Во-первых, тогда не было ЭВМ с требуемыми параметрами для реально-временной обработки звуковых сигналов. Во-вторых, стоили эти машины в тысячи и даже десятки тысяч раз дороже, чем усилители. Они имели размеры целого здания и потребляли огромное количество электроэнергия. По всей видимости, по этим причинам отсутствовал интерес инвесторов и стимул для специалистов по разработке алгоритмов обработки в подобных устройствах. В настоящее время ситуация коренным образом изменилась. Современные ПК уже стоят на порядок дешевле усилителей так называемого HI-Ehd класса, а их характеристики позволяют реализацию весьма сложных алгоритмов обработки сигналов в реальном масштабе времени. Структурные схемы и алгоритмы работы подобных устройств в настоящее время также уже изобретены и запатентованы. Ряд из них подробно рассмотрен выше. Эти варианты реализации блоков 2 можно использовать по новому назначению для решения задачи повышения точности усиления электрических сигналов в усилителях различного назначения, например в усилителях мощности. Понятно, что новые функциональные связи между блоком 2 обработки и усилителем или подсистемой усилитель, связанный с нагрузкой, позволяют рассматривать эти технические решения как самостоятельные изобретения.In this model, the machines or systems of the valley work purposefully, and for this they must have a certain intelligence, which is used to adjust their work depending on internal or external factors. In this model, any controllable (or synthesized) object was considered within the framework of an extremely complicated model - as a “black box” with a very complex, most often unknown, unstable internal structure and which may also be externally affected, which somehow changes its properties and ability perform some action or solve a problem, for example, amplify electrical signals like a power amplifier. In this model, the properties of this object (amplifier) should be described by some more complex function than the traditional transfer function of the electric signal amplifier, presented in formula (1). This function can be not only a complex frequency function, but also depend on other parameters and factors. For example, from the current spectrum of the signal source 1 messages, temperature, pressure, humidity, mechanical stress (vibration, acceleration) and, of course, time, because these parameters can change over time. In the models used by H. Black, G. Nyquist and H. B. Bode, amplifiers were considered in a simpler model - as stationary systems. The influence of the above parameters was considered insignificant. This, in fact, was the main innovation in the construction of a new, more advanced theory of analysis and synthesis of machines. N. Wiener generalized this model of description of material objects to living and non-living objects, common to which was the presence of feedback signals and some organ or unit for processing signals and forming signals controlling the state of the object. Moreover, these auxiliary control signals should promptly have some "positive" effect on this "black box" and compensate for the effects of destabilizing factors. For example, to provide at the amplifier output a form of amplified signals that is extremely achievable that accurately repeats the shape of its input signals. Thus, the system proposed by N. Wiener, the machine, or in this particular case, the amplifier should be built according to some extremely enlarged, unified functional block diagram. The structure of this circuit should include some new node (an analogue of the brain in an animal or human), which should provide a comparative analysis of the input ("reference") and output ("distorted") signals of the amplifier, and generate control signals that correct its operation. Both in the works (patents) of H. Black and in the modern circuitry of amplifiers, as such a unit - processing unit 2, the simplest devices were used, most often consisting of resistors of the adder 17, frequency-correcting capacitors and inductances. It is clear that with the help of these simplest elements (resistors R ₀ , R1 and the input resistance of the inverting input of amplifier A1), providing, for example, OOC in the amplifier shown in Fig. 21, in principle, it is impossible to solve such a difficult task, since in actual use the amplifier can change over time the properties and parameters of all the main elements of the circuit (Fig. 21). For example, the complex transfer coefficients of the amplifier A1, transistors VT9, VT10, the capacitance of the correction capacitor C1 can change, and in the most general case, the properties of the communication line connecting the amplifier with the load R _H can change (due, for example, to a change in its length, wave impedance ) and the complex resistance of the load itself. Secondly, to obtain an output signal that follows the input signal as accurately as possible, it is necessary to form an extremely accurate complex coefficient

OS chains, and moreover, as quickly as possible, try to adjust it to the current changes in the transfer function

the original amplifier or vice versa (which is even better) to carry out the predistortion of the transfer function

source amplifier so that the integrated feedback coefficient

remained unchanged. The implementation of such an approach and the implementation of these actions can give the best results on the "mutual coordination of the applied solutions" from the point of view of modern scientific methodology - cybernetics. Haphazardly using well-known recommendations for the "competent" use of environmental protection, it is impossible to synthesize an amplifier of the highest form of organization, quality class, complexity group, etc. To do this, the amplifier must include a new function and a very complex unit - signal processing unit 2. With its help, it is possible (instead of preliminary, ineffective computer simulation of the amplifier correction circuits) to carry out in real time, on a real amplifier, a full-parameter analysis of the input and output signals of the amplifier, generate optimal pre-emphasis of the amplified signals (for example, in each amplification stage), and form compensation signals for active suppression of non-linear and in-modulation distortion products inside the amplifier. It is clear that in the days of N. Wiener the idea of using a computer to increase the linearity of the amplifier would have seemed too daring. First, then there was no computer with the required parameters for real-time processing of audio signals. Secondly, these machines cost thousands and even tens of thousands of times more expensive than amplifiers. They were the size of the whole building and consumed a huge amount of electricity. Apparently, for these reasons there was no investor interest and incentive for specialists to develop processing algorithms in such devices. Currently, the situation has radically changed. Modern PCs are already much cheaper than amplifiers of the so-called HI-Ehd class, and their characteristics allow the implementation of very complex algorithms for processing signals in real time. Structural schemes and algorithms for the operation of such devices are now also invented and patented. A number of them are discussed in detail above. These options for the implementation of blocks 2 can be used for a new purpose to solve the problem of increasing the accuracy of amplification of electrical signals in amplifiers for various purposes, for example in power amplifiers. It is clear that the new functional relationships between the processing unit 2 and the amplifier or subsystem, the amplifier associated with the load, allow us to consider these technical solutions as independent inventions.

On Fig, schematically, shows a structural diagram of possible options for connecting the unit 2 to the amplifier 5. There are various options for generating feedback signals, for example a common OS, as well as various combinations of both negative and positive feedback. As feedback signals, you can use: the signal from the output of the amplifier (the signal fed to 1 input B of the signal processing unit 2); a signal from the input of the emitter 4 (a signal supplied to 1 input C of the processing unit 2); a signal generated in the vicinity of the emitter 4. This can be done using a microphone installed, for example, at a distance of about 1 ÷ 25 mm from the surface of the speaker cone 4 (a signal supplied to 1 input D), or this signal can be generated using an electromechanical transducer installed on the surface of the speaker cone 4 (the signal supplied to 1 input E of the signal processing unit 2),

At 1 input A of processing unit 2, signals can also come from the outputs of stages of amplifier 3, 9, for example, from the output of the first amplification stage A1. In this case, the local OS is also formed by the signal processing unit 2, for example, composed of two processing units 2 operating in parallel, but with different control time constants. Replacing the capacitor C1 with a multifunctional unit-signal processing unit 2 allows for more accurate correction of the operation of the first stage of the amplifier A1, for example, taking into account all the recommendations contained in the article by S. Ageev or in the patents of H. Black. It is clear that by means of the processing unit 2, it is possible to generate not only negative feedback signals, but also positive feedback signals, which functionally can serve as signals (compensating current i ₂ ) generated by the resistor R2 (Fig. 21). On Fig this option is schematically shown in the form of a dashed line connecting the block 2 with the bases of transistors VT9, VT10.

To improve the accuracy of signal amplification, the following structural options are possible for constructing a signal processing unit 2.

цепи ОС. Таких полос для выполнения требования к патентоспособности изобретения по новизне должно быть не менее двух. В крайних низкочастотных и высокочастотных полосах можно использовать в качестве фильтров 14 ФНЧ и ФВЧ, порядок которых подбирается (неавтоматически или автоматически) с использованием, например, общеизвестных методов синтеза фильтров по виду АЧХ и ФЧХ усилителя или его каскадов.The first embodiment of the processing unit 2 can be obtained if the resistor R1 shown in Fig. 21 is implemented as a node providing multi-band processing of feedback signals using, for example, input filters 14, phase regulators 29, and levels 31 included in the composition block 16 components of feedback signals, allowing through preliminary experimental fitting and selection of these adjustments to form the desired type of transfer function

OS chains. There must be at least two such bands to fulfill the patentability requirement of an invention. In the extreme low-frequency and high-frequency bands, 14 low-pass filters and high-pass filters can be used as filters, the order of which is selected (either automatically or automatically) using, for example, well-known methods of filter synthesis according to the form of the frequency response and phase response of an amplifier or its stages.

исходного усилителя), так и предыскажения энергетических и временных параметров сигналов обратной связи (для предыскажений комплексного коэффициента

) цепи ОС. Причем, как уже было показано выше, в этом варианте возможна самонастройка параметров системы обратной связи с учетом дестабилизирующих факторов (температуры, влажности, давления, старения элементов конструкций, например высыхание электролитических конденсаторов, параметров текущего спектра сигналов источника 1 сообщений и т.д.). В этом варианте также осуществляется автоматический поиск и компенсация всевозможных продуктов искажений. Эти действия являются следствием процесса фильтрации этих продуктов искажений - помех и их активного понижения (противофазного сложения) в усилителе. Помимо этого процесса, одновременно с ним, может осуществляться процесс многополосного порогового понижения шума электрических сигналов источника 1 сообщений перед их суммированием с сигналами для активного понижения продуктов искажений усилителя. Сформированный таким образом сигнал может подаваться как на вход усилителя 3, 9 или на его хотя бы один из каскадов. Этот метод двойной фильтрации сигналов позволяет эффективно фильтровать продукты искажений самого усилителя и фильтровать, например, стационарные помехи или наводки, например, частотой 50 или 100 Гц.The second embodiment of the processing unit 2 can be obtained if the resistor R _{0 of the} adder, as well as the resistor R1, is implemented as multiband signal processing channels of the message source 1. This embodiment of the processing unit 2 allows the automation mode of various adjustments of the energy and time parameters as signals of the message source 1 (for predistortion of the complex transfer coefficient

the initial amplifier), as well as the predistortion of the energy and time parameters of the feedback signals (for the predistortion of the complex coefficient

) OS chains. Moreover, as already shown above, in this embodiment, self-tuning of the feedback system parameters is possible taking into account destabilizing factors (temperature, humidity, pressure, aging of structural elements, for example, drying of electrolytic capacitors, parameters of the current spectrum of the signal source of 1 message, etc.) . This option also automatically searches for and compensates for all kinds of distortion products. These actions are a consequence of the process of filtering these distortion products - interference and their active reduction (antiphase addition) in the amplifier. In addition to this process, simultaneously with it, a process of multi-band threshold noise reduction of electrical signals of the message source 1 can be carried out before summing them with signals to actively reduce the distortion products of the amplifier. The signal thus formed can be supplied either to the input of amplifier 3, 9 or to at least one of the stages. This method of double filtering signals allows you to effectively filter the distortion products of the amplifier itself and filter, for example, stationary interference or interference, for example, at a frequency of 50 or 100 Hz.

It is also possible that the overall OOS in the amplifier 3, 9 remains unchanged in the form of traditional elements R ₀ , R1, structurally located inside the amplifier 3, 9, and the processing unit 2 is connected to the input and output of the standard amplifier 3 in the form of an additional prefix. This option is convenient in that it does not require any structural modifications of the amplifier 3, 9 or more resistant to the consequences of possible adaptive changes in the OOS circuit. The disadvantage of this option can be considered a limited gain, which is determined by the ratio of resistors R ₀ and R1, as well as the need to limit the range of frequency-energy and frequency-time pre-emphasis in the OOS circuit. The above adaptive changes in the OS circuit can only reduce the gain of the amplifier in the entire frequency band or in individual frequency sections.

As one example of possible circuitry solutions of the described options, you can use, for example, the nodes of the circuit shown in Fig.19, together with almost any standard amplifier 3, 9.

The signal source 1 of the message is fed to the 2 input of the circuit (Fig. 19) for each of the processing bands, which can be, for example, 50 or 60. The feedback signal, for example, from the output of amplifier 3, 9 is fed to the additional input of block 2 so that the same number of feedback signal processing bands, and, for example, a signal weakened to the operating level, for example, from one of the windings of the transformer of the power supply unit of the amplifier 5, is fed to the input 1 of the processing unit 2 to generate signals of active reduction of interference and interference caused by fields from these signals transformer pit amplifier 3, 9. It is clear that there can be several such inputs, for example, for actively reducing the impulse noise from the line transformer of a television receiver, for actively reducing the impulse noise in the power supply circuits of the amplifier 3, 9, etc. For each of these signals, the parameters of the feedback circuit are pre-selected (automatically or non-automatically) and remain unchanged or adjusted during operation of the amplifier 3, 9. Adaptive changes can also be made to the parameters of message signals. For example, in the circuit (Fig. 19), an amplifier can operate a multi-band threshold noise suppressor of the signal source 1 messages (with the switch SA5 lowered down) or a noise suppressor together with an automatic equalizer that compensates for the possible amplitude-frequency distortions of the original amplifier 3, 9, for example, -for heating and changing the capacitance of the capacitors of the frequency correction in the stages of the amplifier 3, 9. It is clear that, adding to the circuit shown in Fig. 19, nodes that automate the selection of phase parameters of the signals, m You can significantly expand the functionality of the amplifier 3, 9 and implement the option of a full-parameter analysis of the signals of the message source 1, feedback signals, received noise signals of interference and interference, process them and generate them using a single adder 17 (for example, executed on element D 3.4 by applying to it inputs signals from the output of switch SA5, resistors R 64, R 87) output signals of processing unit 2. These signals are fed to the input of amplifier 3, 9. Moreover, amplifier 3, 9 can have its own local and / or common feedbacks, both positive and negative. If the feedback signal is formed near the emitter (loudspeaker 4, 10), then it is possible to correct the operation of the amplifier 3, 9 in conjunction with other nodes of the system — the line connecting the amplifier 3, 9 with the loudspeaker 3, 10. In this embodiment, for example, negative feedback will effectively monitor and compensate not only the distortion products in the amplifier 3, 9 itself, but also all kinds of distortions of the signal source 1 messages that occur in the communication line and loudspeaker 4.10. Moreover, the implementation of this option, for example, in sound engineering is more preferable, since the main distortions occur precisely in the loudspeakers, and their level is orders of magnitude higher than in amplifiers 3, 9. The nonlinear distortions of the 4, 10 HI-FI class speakers amount to about 1 at the rated power -3% At maximum power, they increase to 10% and higher. The effectiveness of the suppression of the products of nonlinear and intermodulation distortion of the speakers will mainly be determined by the delay time of the feedback signal (the distance of the microphone that converts the sound vibrations of the diffuser into electrical feedback signals) and the signal frequency. The most effective suppression of nonlinear products will be in the low-frequency part of the sound spectrum, which, for example, is very important in active noise reduction systems. The coverage of such nodes of the entire messaging system with such "local" multiband feedback can give (due to the small delay of the feedback signal) a significant increase in the accuracy of the transmitted signals in the extended communication channel. In such a subsystem, the level of active suppression of the products of nonlinear distortion can already amount to not a few dB, as with analogs, but much more. This effect will be greatest for systems with electromechanical feedback. Moreover, in such systems, you can significantly reduce the requirements for the accuracy of the calculation of acoustic systems. The mismatch in the design of the dynamic head (the size and design of the loudspeaker box 4, 10) with its characteristics will be compensated by the system, which will form the frequency response and phase response of the loudspeaker associated with amplifier 3, 9, close to ideal. The main thing is to provide a sufficiently large box volume to increase the efficiency of low frequency reproduction, as well as provide the necessary power reserve for the amplifier 3, 9 and loudspeaker 4, 10 to compensate for the frequency response unevenness.

It should be noted that changes in the energy and time parameters of the signals of the source 1 messages and feedback signals should be consistent and interconnected. These adjustments must occur over time so that the amplifier 3, 9 is not excited in the corresponding frequency bands. For example, if the variation of the feedback signals by energy and time predistortions is completely eliminated by fixing them (working from memory), then the range of possible adjustments of the energy and time predistortions of signal parameters of the message source 1 (input) can be any. It is determined by the overload capacity of the amplifier. In this situation, with a linear amplitude characteristic of the OS circuit, the feedback signals will also change proportionally. Their levels and phases will synchronously repeat the predistortion of the input signals, of course, with the accuracy that this amplifier provides. If at the same time it is not consistent, at the same time to automatically correct the energy and time parameters of the feedback signals, for example, in several amplification stages, then for individual frequencies it is possible to fall into the region of unstable operation of the amplifier. Therefore, it is advisable to pre-configure the amplifier using signals convenient for this form (for example, noise-like signals) and, for example, configure individual stages of the amplifier 3, 9 sequentially in time. To identify the dynamic characteristics of the amplifier 3, 9, you can use the signal as test tuning signals such as "pink noise", the slew rate of which varies. Thus, it is possible to combine the research processes of amplifier 3, 9 in the frequency and time domains, generating a complex shape of the input test signals. If the capabilities of the processor or computer do not allow such fairly complex algorithms for full-parameter analysis of signals and self-tuning of the amplifier 3, 9, it is desirable to limit the possible range of adjustments of both the energy and time parameters of the feedback signals, based on the fact that as the amplifier 3, 9 a high class amplifier will be used. For example, you can introduce restrictions on the regulation of the levels of the components of the feedback signals of not more than 10%, and the phase of not more than 20 °. Moreover, the analysis of signals in the processing band can be carried out on amplified signals without using special test signals generated by the processing unit 2, and the corresponding predistortions can be formed in the pauses between the signals of the message source 1 (to exclude dynamic failures in the OS circuit).

When generating feedback signals, multi-threshold methods for analyzing the signals of the message source 1 can be used and, in accordance with their current level, the transfer properties of the amplifier 3, 9 can be formed in accordance with some algorithm that optimizes its operation. For example, in accordance with a certain, current level of message signal components in the corresponding analysis band, it is possible to generate some current value of the gain of the amplifier 3, 9 or to form a different type of its amplitude characteristic. These actions allow you to optimize the operation of the amplifier 3, 9 in accordance with the current energy spectrum of the message signal. For example, most experts in the comparative analysis of tube and transistor amplifiers operating at maximum power, prefer tube because of their soft, natural sound. There are several reasons for this. One of them is concluded in a significantly different form of their amplitude characteristics. In transistor amplifiers, the amplitude characteristic has a sharp kink. As a result of this, at large short-term levels of signals, their limitation occurs, and an extensive spectrum of various distortion products appears. For tube amplifiers, signal limitation occurs smoothly, and the levels and number of nonlinear signal components are significantly lower than for transistor amplifiers. These distortions are less noticeable by ear. In addition, due to reactive components of the load, for example, in multiband columns with separation filters on R, L, C elements, at separate frequencies, the load resistance can differ sharply from one typical (calculated) value, for example, decrease from 8 Ohms to 2 ÷ 3 ohms, and the power of the transistor amplifier to operate at this low load may simply not be enough. As a result, distortions also appear. Therefore, when designing transistor amplifiers, they try to provide a significant margin in power (so that there are no restrictions on the amplitude and current of the amplified signal). For example, in transistor amplifiers of the HI-End class, in the output stages, about 10 ÷ 20 powerful transistors are installed in each arm, operating in mode A or more often AB. The total current flowing through these transistors can be up to several amperes at a supply voltage of the order of ± 30 ÷ 50 V. Therefore, for amplifiers of this design, one of the first places is the problem of heat removal (from several hundred watts to about kilowatts of power), thermal stabilization - and the reliability of the amplifier. Overheating of the amplifier leads to a decrease in the level of amplification of the signals of the source 1 messages without significant distortion. Therefore, using the results of the analysis of the current energy spectrum of the signal of the messages arriving at the amplifier input, it is possible to generate control signals and generate the required type of amplitude characteristic of the transistor amplifier, which minimizes distortions associated with a sharp restriction of signals by level, and form the current value of the supply voltage, on the one hand , eliminating the limitation of peak signal values, on the other hand, minimizing the useless heat consumption for heating the environment, to optimize the optimal energy mode of the amplifier (for example, by transferring the amplifier from mode A to AB or B mode), to form the configuration of the amplifier, for example, by connecting additional cascades of current amplifiers to work - to minimize the overload of the amplifier by current when operating at a complex load and, finally, to form the optimal cooling mode of the amplifier 3, 9, for example, by turning on fans or a water cooling system. It is clear that all these functions can be implemented in any known manner through the above control signals and in accordance with the specific algorithm of these actions. To synchronize the processes of transferring the amplifier 3, 9 to the corresponding state in time, technical solutions similar to the above described can be used to additionally delay the signals received at the input of the amplifier 3, 9 according to the time required to generate these control signals, for example, turn off and turn off the fan and cooling amplifier during the signal time is high enough to mask the sound noise interference from its operation, if the amplifier does not include an active suppressed system I the interference.

Thus, the above-described options for step-by-step processing of signals: in amplification stages of signal amplifiers, in loudspeakers, a communication channel, as well as in signal processing units, they can quickly detect distortions of various nature and generate signals, on the one hand, compensating for these distortions, and on the other hand, adjust the parameters of the transmitted message signals. At the same time, methods of active antiphase signal addition or the method of interrupting the transmitted message and interference at frequencies where the signal level is small and the noise level is high are used to compensate for noise. Due to this, it is possible to increase the accuracy of the transmitted message.

In conclusion, it can be noted that the new technical solutions described in this invention are optimal not only for solving technical problems. They can be used in the development of new socio-economic models for the development of society and the organization of industrial production. For example, replacing technical categories and concepts (filters, amplifiers, comparison schemes, noise suppressors, correctors, etc.) with their analogues - economic and socio-social categories (ministries, departments, planning, management, supply departments, executive authorities - army, police, special services, the judiciary, etc.), it is possible from the traditional methods of development of civilization, based on unsystematic adoption, often erroneous decisions, to switch to new, scientific, progressive methods of managing public and industrial oh activity of people. This approach will bring technical and economic cybernetics into conformity, since the laws by which complex systems can operate most efficiently regardless of their physical nature, number of components, frequencies, etc., obey the same laws of nature - the law of MINIMIZATION OF GROWTH ENTROPIES. It is clear that in the economic, social and production systems, the corresponding “sources of 1 message signals” should be used. These "sources" include, for example, the basic provisions of the Bible, the basic and auxiliary laws contained in the constitutions of countries, all kinds of Codes, various international agreements, treaties, and in production - production plans and tasks, in business - business plans, etc. .d. The implementation and compliance with these norms, rules, work plans is in many ways similar to the transmission of messages in space and in time in the presence of interference. For example, highly efficient industrial production using conveyors requires the precise execution of strictly defined operations. The "levels" and "phases" of these operations (actions) should clearly correspond to the previously planned. Otherwise, the conveyor will stop, and highly efficient industrial production will become impossible. Highly effective management of investment resources also requires the rapid collection of both external economic, scientific and technical information, as well as information within the investment structure, in order to ensure the conditions for the fastest, most qualified processing of this information and the timely adoption of correct management decisions on investing in promising enterprises, countries or corporations . It is clear that the structures involved in information processing can be nested into each other in accordance with the hierarchy of the system, like local local and general feedbacks in highly organized messaging systems. They can have a different level of complexity, functionality and internal architecture. But any highly organized social or production system must function according to the same unified principles as a high-quality system for transmitting audio messages. The system should collect information in the entire "spectrum" - in all areas of the functioning of this system and information on the "interference". As a result of this processing, signals should be formed that correct its operation, for example, decisions on timely investment of funds in competitive firms. Failure to comply with these actions reduces the profits of investors and hinders the development of science, technology and civilization as a whole. This can be easily verified by analyzing the history of the development of civilization on Earth. There are plenty of examples.

Industrial applicability

The proposed technical solutions can be used in almost all areas of science, technology and production, since they reveal the content of a new fundamental law of nature - THE LAW OF MINIMIZATION OF ENTROPY GROWTH.

Claims (36)

1. A method of transmitting messages of any physical nature, which consists in converting messages into electrical signals of the message source, transferring them to the processing location, processing the source signals through the processing unit, to the second input of which these signals are supplied and configured to perform energy and / or temporary pre-emphasis electrical signals of the message source during the formation of the output signals of the processing unit, their amplification, conversion into signals of the same physical nature and radiation in the transmission of messages to the area of the space of their receipt, in which there are interference or noise of the same physical nature, characterized in that they receive noise signals of any physical nature, convert them into received noise electrical signals, transmit them to the place of processing, process these signals by a processing unit provided with a first input to which these signals are supplied and configured to perform energy and temporal predistortions of the received noise signals when formed and output electrical signals of the processing unit, designed to actively reduce noise in the region of the space for receiving messages, amplifying them, converting them into signals of the same physical nature and radiation to the region of the space for receiving messages, moreover, the place for receiving noise signals is chosen closer to the noise source than the space region receiving messages, and energy and temporary predistortion of received noise electrical signals is carried out by selecting the levels and phases of their components at a zero level of e electrical signals of the message source or at a level of electrical signals of the message source below a predetermined threshold, storing energy and temporary predistortions, providing the effect of lowering the noise level in the region of the space for receiving the message. 2. The method of transmitting messages of any physical nature according to claim 1, characterized in that the electrical signals for active noise reduction are formed in a limited frequency band of the low-frequency part of the noise spectrum of the received noise signal to increase the size of the area of the noise reduction space. 3. The method of transmitting messages of any physical nature according to claim 1, characterized in that the energy pre-emphasis of the electrical signals of the message source is carried out by adjusting the signal level of the message source, energy and time by means of loudly adjusting the signal level of the source, and temporary by delaying the signal of the message source , for example, in the anti-vibration system of an automobile CD player. 4. The method of transmitting messages of any physical nature according to claim 1, characterized in that for adjusting the spectrum of message signals taking into account possible distortions of transmitted messages in the channel, non-automatic frequency-energy and / or time-frequency predistortions of the electrical signals of the message source are carried out by the processing unit, made with the possibility of multi-band processing of the electrical signals of the message source in the formation of its output signals. 5. The method of transmitting messages of any physical nature according to claim 4, characterized in that the frequency-energy predistortions are carried out by means of an equalizer, the frequency-time predistortions are performed using frequency-time or frequency-phase correctors, and the frequency-energy and frequency-time predistortions carried out by means of tone blocks or fixed presets of amplitude-frequency and phase-frequency characteristics in predistortion filters. 6. A method for transmitting messages of any physical nature according to any one of claims 1 to 5, characterized in that for the implementation of the mode of lowering the level of various noise components in the entire above region of the noise spectrum, taking into account possible almost constant differences in the spectrum of noise signals in the region of the space for receiving messages the spectrum of the received noise signal is carried out by non-automatic frequency-energy and time-frequency predistortions of the received noise electrical signals by means of a processing unit made with zmozhnostyu multiband processing the received noise signal at the formation of its output signals. 7. A method for transmitting messages of any physical nature according to any one of claims 1 to 5, characterized in that in the region of the space for receiving messages, additional signals are received of any physical nature, which are a message signal plus noise, the spectrum of which can change over time the spectrum of the received noise signal not only due to the current changes in the spectrum of the transmitted message, but also due to changes in the noise spectrum itself in the region of the space for receiving messages in relation to the spectrum of the received about a noise signal, converting an additionally received signal of any physical nature into electrical feedback signals, transmitting them to the processing site, processing them by a processing unit made with an additional input to which these signals are fed, and configured to automatically generate energy and time or frequency-energy and / or frequency-time predistortions of received noise signals of any physical nature by introducing nodes into the processing unit that allow vtomatizirovat steps in selection of the levels and / or phase components of the signals generated by the active noise reduction as well as the action of memorization and forming energy and / or time of the predistortion signals, providing the effect of lowering the noise level in the surrounding area receive the message. 8. The method of transmitting messages of any physical nature according to claim 1, characterized in that in the region of the space for receiving messages, additional signals are received of any physical nature, which are a message signal plus noise, the spectrum of which can vary with respect to the spectrum of the source signals not only due to current changes in the spectrum of the transmitted message and random interference or noise, but also due to random changes in the transfer function of the message channel, the conversion of the additionally received signal Ala of any physical nature into electrical feedback signals, their transmission to the place of processing, processing by means of a processing unit equipped with an additional input to which these signals are supplied, and configured to automatically generate frequency-energy and / or time-frequency predistortions of the electrical source signals messages by introducing nodes into the processing unit that automate actions to select the levels and / or phases of multiband-generated components of the source signals Message Type. 9. The method of transmitting messages of any physical nature according to claim 8, characterized in that for the implementation of the mode of lowering the level of various noise components in the entire above region of the noise spectrum, taking into account possible almost constant differences in the spectrum of noise signals in the region of the space for receiving messages in relation to the spectrum of the received noise signal carry out frequency-energy and time-frequency predistortion of received noise electrical signals by means of a processing unit configured to lane non-automatic processing of noise received electrical signals when forming the output signal processing unit. 10. The method of transmitting messages of any physical nature according to claim 8, characterized in that for the implementation of the mode of lowering the level of various noise components in the entire above region of the noise spectrum, taking into account possible random changes in time of the spectrum of noise signals in the region of the space for receiving messages in relation to the spectrum of the received noise signal, automatic frequency-energy and time-frequency predistortions of the received noise electrical signals are carried out by the processing unit, performed о with the possibility of performing these pre-emphasis only at those times when the signal level of the message source is lower than the predetermined one, for example, before message transmission or in pauses during message transmission, by introducing nodes into the processing unit that automate actions to select multiband levels and phases generated signals for active noise reduction at these time points and their storage, and the formation of stored values of pre-emphasis during the transmission of signals of the message source, when source signals ram messages preassigned above. 11. A method for transmitting messages of any physical nature according to any one of claims 1 to 5, 8-10, characterized in that the processing unit is configured to summarize the processed signals of the message source and the processed received noise signals of any physical nature by means of an adder to thereby form it output signals and the implementation of a single-channel method of transmitting messages of any physical nature. 12. The method of transmitting messages of any physical nature according to claim 9 or 10, characterized in that the processing unit is configured to sum the multiband-processed signals of the message source and multiband-processed received noise signals of any physical nature by at least two adders to form in this way its output signals and the implementation of a multiband method of transmitting messages of any physical nature. 13. A method for transmitting messages of any physical nature according to any one of claims 1-5, 8-10, characterized in that the conversion of messages into electrical signals of the message source is carried out in a multi-channel way to implement a multi-channel method for transmitting messages to the corresponding number of channels areas of the message receiving space, in each of these channels, the method of message transmission described above is carried out, while the received noise signal is common to these channels and is fed to additional inputs of the processing units, whose layer is equal to the number of message channels. 14. A method for transmitting messages of any physical nature according to any one of claims 1-5, 8-10, characterized in that the conversion of messages into electrical signals of the message source is carried out in a multi-channel way to implement a multi-channel method for transmitting messages to the corresponding number of channels areas of the message receiving space, in each of these channels the method of message transmission described above is carried out, while the number of received noise signals is equal to the number of message transmission channels and they are fed to additional inputs odes of processing units, the number of which is equal to the number of message transmission channels. 15. The method of transmitting messages of any physical nature according to claim 12, characterized in that each output signal of the processing unit is amplified by an amplifier of electrical signals and radiated to a region or regions of the space for receiving messages of any physical nature. 16. A method for transmitting messages of any physical nature according to any one of claims 1-5, 3-10, characterized in that the locations of the processing unit and the signal source of the messages do not spatially coincide to enable them to be implemented in blocks and to complete the message transfer system with the blocks above functionality or their location are nearby for their constructive implementation in one device. 17. The method of transmitting messages of any physical nature according to any one of claims 1-5, 8-10, characterized in that sound messages are used as messages of any physical nature. 18. A method of transmitting messages of any physical nature, which consists in converting messages into electrical signals of the message source, in amplifying these signals through an amplifier. made in the form of one or more amplification stages, supplying the amplified electrical signals of the message source to the load, generating electrical feedback signals, supplying them to at least one of the amplification stages, characterized in that at least one of the generated feedback signals is subjected to multiband processing the processing unit of the signals that are fed to its first input, in the processing unit, the feedback signals are multi-band filtered to form at least two channels of the multi-band the work of the feedback signal components, each multiband channel formed is capable of performing energy and time predistortions of the feedback signal components to increase the accuracy of the transmitted message by realizing the possibility of more accurate formation of the required transfer function of the feedback circuit, multiband processed feedback signal components are used to generate the output signals of the signal processing unit, which are supplied to the above although b one of the amplification stages of electrical communications signals. 19. The method of transmitting messages of any physical nature according to claim 18, characterized in that in the signal processing unit, the multi-band processed components of the feedback signals are added to form the output signals of the processing unit. 20. The method of transmitting messages of any physical nature according to claim 18, characterized in that in the processing unit the multiband processed feedback signals are additionally filtered to reduce non-linear and intermodulation distortion of the components of the feedback signals in their processing bands, the multiband processed and additionally filtered components of the feedback signals communications summarize, the signals obtained in this way are the output signals of the processing unit. 21. The method of transmitting messages of any physical nature according to claim 18, characterized in that the feedback signals are negative or positive feedback signals. 22. The method of transmitting messages of any physical nature according to claim 18, characterized in that the amplification of the electrical signals of the message source is carried out by multistage amplification with combined feedback, negative and positive, with at least one of the feedbacks being subjected to multiband processing by means of a signal processing unit in accordance with the sequence of the above actions, if several generated feedback signals are subjected to multiband processing, then the existing number of signal processing units, at the first inputs of which the corresponding generated feedback signals are supplied, and the corresponding output signals of the processing units are fed to the corresponding amplification stages of the electrical signals of the message source. 23. A method for transmitting messages of any physical nature, p. 18, characterized in that a power amplifier is used as an amplifier, a signal emitter is used as an amplifier for converting electrical amplified message signals of any physical nature into message signals of any physical nature, and electrical reverse signals communications form from the output of the amplifier or from the input of the emitter, or by additional conversion of signals of any physical nature in the immediate vicinity of the radiation ator into electrical feedback signals. 24. The method of transmitting messages of any physical nature according to any one of claims 18 to 23, characterized in that the energy and temporary pre-emphasis of the components of the electrical feedback signals is carried out by preliminary non-automatic selection of adjustments of the levels and phases of the multi-band processed feedback signals from the training message signals convenient for this form, for example, noise-like signals or sinusoidal signals of different frequencies according to a criterion, for example, the minimum standard deviation signals transmitted message. 25. The method of transmitting messages of any physical nature according to any one of claims 18 to 23, characterized in that the energy and temporary pre-emphasis of the components of the electrical feedback signals is carried out automatically according to the signals of the message source, for example, noise-like form or actually according to the transmitted message signals through the block signal processing, configured to automate the above actions and equipped with a second input, to which electrical signals of the message source from the input an amplifier or a corresponding amplification stage covered by feedback, while the processing unit carries out multi-band filtering of the electrical signals of the message source using filters with characteristics similar to filters designed for multi-band filtering of feedback signals, the joint processing of the feedback signal components filtered in the corresponding bands and messages, the development of control signals that automate the processes of energy and temporary predistortion feedback signals. 26. The method of transmitting messages of any physical nature according to claim 18, characterized in that the signal processing unit is provided with a second input, to which electric signals of the message source or amplified in the previous stages electric signals of the message source are fed, multiband filtering of electric signals is carried out in the processing unit a message source by means of filters with characteristics similar to filters designed for multi-band filtering of feedback signals, while at least two channels of multiband processing of signal source signal components are processed, each multiband channel is configured to perform energy and time predistortion of message source signal components to improve the accuracy of the transmitted message by compensating for amplitude-frequency and / or phase-frequency distortions of message signals on a portion of the transmission chain messages covered by feedback, for example in an amplification stage, in an amplifier, or in an amplifier connected to a load Second, the treated components multiband feedback thus obtained signals are the output signals of the processing unit is added to the components of the multi-band message source signals processed signals are input to the amplifier or a cascade of said covered feedback. 27. The method of transmitting messages of any physical nature according to p. 26, characterized in that the energy and temporary pre-emphasis of the components of the electrical signals of the message source is carried out automatically in the processing unit by jointly processing the multi-band formed components of the feedback signals and message signals, generating control signals that automate processes of energy and / or temporary predistortion of message signal components. 28. The method of transmitting messages of any physical nature according to p. 26 or 27, characterized in that in the process of transmitting messages in the processing unit, the analysis of the levels of the components of the message signals in the processing bands is carried out and control signals are generated by which form the previously selected adjustments of the levels of the source signal components messages and / or feedback signals from the moment when the level of the components of the message signals in the corresponding processing band is higher than the predetermined threshold level, in others At these time points, the levels of the generated components of the signal signals of the message source and / or feedback signals in the processing unit are automatically changed to reduce the transmission coefficient of the components of the message signals in this frequency band and to reduce due to this level the components of the products of non-linear and intermodulation distortions, to reduce noise, interference and interference . 29. The method of transmitting messages of any physical nature according to claim 28, characterized in that for synchronizing in time the processes of automatic multiband threshold adjustment of signal source signal component levels with the moment of exceeding the level of message signal components at a predetermined threshold during temporary pre-emphasis of source signal components messages carry out their additional delay by the amount of time required to generate control signals in the processing unit, providing threshold adjustment of signal component levels at the output of the signal processing unit. 30. The method of transmitting messages of any physical nature according to claim 28, characterized in that in the process of transmitting messages of any physical nature in the signal processing bands of the message source, the current levels of the components of the message signals are analyzed with respect to at least two thresholds and depending on the current level of the components of the message signals form various gear ratios and the form of the amplitude characteristic in these signal processing bands of the messages. 31. The method of transmitting messages of any physical nature according to claim 30, characterized in that a two-threshold analysis of the levels of the components of the message signals is used, and the transmission coefficients of the messages in the corresponding frequency bands are formed according to the following algorithm: if the signal level of the messages is lower than the first threshold, which may be in advance set, for example, at a level of - 20 ÷ 70 dB relative to the maximum undistorted amplified message signal, then the transmission coefficient in the corresponding signal processing band of the source is notified th and / or feedback signals are chosen so as to reduce the gain of the cascade or amplifier covered by feedback, and implement a mode of strip lowering of the level of amplified message signals if the signal level of messages in this frequency band exceeds the first threshold level, but below the second threshold level , which can be pre-set, for example, at the level of - 10 ÷ 2 dB relative to the maximum undistorted amplified message signal, then the transmission coefficient in this signal processing band is set wound with previously selected parameters of energy and temporal predistortions, and energy and temporal predistortions are constant and independent of the level of components of message signals to implement a linear amplitude characteristic of amplification of message signals, if the level of message signals is higher than the second threshold, then variable energy predistortions of signals are generated to realize compression mode of the dynamic range of amplified signals and the formation of a nonlinear amplitude characteristic corresponding band-pass processing channel, for example, of a logarithmic form and due to this increase in non-linear distortions associated with a possible sharp limitation of the amplitude of the amplified message signal. 32. The method of transmitting messages of any physical nature according to any one of paragraphs 18-23, 26 and 27, characterized in that, depending on the ratio of the current level of message signals and the set threshold or thresholds of these levels, additional control signals for the state of the amplifier are generated, by which the voltage level of the power supply and / or the operation mode of the amplifier, for example, transfer it from AV mode to A. 33. The method of transmitting messages of any physical nature according to any one of paragraphs 18-23, 26 and 27, characterized in that the processing unit is performed with at least one additional input to which electrical signals are generated that are formed near interference sources, for example, an alternating voltage with one of the windings of the transformer supplying the amplifier, these signals in the processing unit are subjected to single or multi-band processing to generate signals designed to actively lower the level of these signals in the above amplification stages, to the output e of the amplifier or system, the amplifier and emitter the processed signals are used to generate the output signals of the processing unit by summing them with other output signals of the processing unit. 34. The method of threshold noise reduction of the electrical signals of the message source, which consists in automatically reducing the transmission coefficient of the electrical signals of the message source from the input to the output of the noise canceler at times when the signal level of the message source is lower than the threshold, characterized in that the input signal of the noise suppressor is filtered to obtain several band signals a source of messages, the number of which is at least two, and at least in two of these bands carry out threshold noise reduction of components The electrical communications source signals thus processed signals are used to form the squelch output signals. 35. The method of threshold noise reduction of electrical signals of a message source according to claim 34, characterized in that each multi-band processed signal is additionally filtered when generating noise reduction signals to reduce the level of nonlinear and intermodulation distortion of the noise reduction signals. 36. The method of threshold noise reduction of electrical signals of a message source according to claim 34 or 35, characterized in that the multi-band processed signals are summed to obtain one or more output noise reduction signals intended for their subsequent processing, for example, by amplifying them in power in signal amplifiers.

Images