KR100669566B1 - Method for transmitting information and suitable system therefor - Google Patents

Abstract

The present invention relates to a method for information transmission and a system particularly suited for digital transmission. At least one information signal consists of a reference frequency channel, at least one information frequency channel is generated, and thus the reference frequency channel and the information frequency channel each form a discrete state to generate a bit pattern. This allows the signal to be transmitted, for example, over a distance of several kilometers in water. Appropriate evaluation systems are also disclosed.

Description

Methods of transmitting information and systems performing the same {METHOD FOR TRANSMITTING INFORMATION AND SUITABLE SYSTEM THEREFOR}

The present invention relates to an information transmission method and a system suitable for information transmission.

In many technical fields, waves are used for transmission transmissions. It is, for example, electromagnetic waves or acoustic waves which propagate freely in a particular conductor or in a given transmission medium and pass from the transmitter or transmission unit to the receiver or receiver unit in this manner. If both units are tuned to a suitable frequency range or frequency provided for example for information transmission, the connection is established. Information can be transmitted in the connection in a number of ways.

To achieve this, the initial information, which may be provided in a suitable manner such as speech, text, a series of numbers, music, image data, or other forms of data, may be converted or radiated from the transmitter to the receiver medium in the form of a wave signal. Is encoded. The receiver receives the signal, converts it back to the original signal, i.e. decodes it, and generates information corresponding to the initial information.

Depending on the type of information encoded in the wave, there is a difference between analog and digital information transmission.

In the case of analog information transmission, the transmitted value is converted into a continuous spectrum without phases in the physical state. This typically occurs in the form of amplitude, frequency and / or phase modulation of the carrier. This makes it possible to transmit vast amounts of information at a given time.

In the case of digital information transmission, on the contrary, there is a restriction on a specific discrete state. However, in terms of transmission speed, if electromagnetic waves are used, there is no limit until now because the frequency of the carrier is very high, and different digital states can be realized in a very short time.

In transmission media such as water, the transmission of information by electromagnetic waves is possible in a limited range since it has a small range. Thus, the use of sound waves for information transmission in this environment has the potential to sometimes propagate over very long distances. Sound waves can be modulated in the manner described above. Sound waves, however, are efficient at information rates that are substantially independent of low frequencies and are mechanical pressure waves that are different from normal propagation. For example, this propagation speed is highly dependent on the specific ambient conditions.

The big problem that can arise in the transmission of sound information can be simply illustrated by the transmission of voice signals under water. In the case of propagation in sound space emanating from the transmitter, some waves may be reflected or bent from the surface of the water and / or the volume of the water, depending on depth, from various objects, particles, and even hierarchical heterogeneity in the water. have. Several other components of the sound wave will arrive at the receiver with different amplitude and phase relationships depending on the propagation length, the angular relationship and the acoustic properties of the confinement surface or medium involved. Depending on the result of the interference, the actual signal at the point of reception may suddenly be amplified, attenuated, distorted, or even completely eliminated, and the reception may also be distorted by what is referred to as reverberation.

To illustrate this problem in more detail, the simple situation in which a very short signal of a certain frequency, referred to as CWP (continuous wave pulse), is transmitted will be considered first. In this situation, the receiver can acquire the entire group of temporarily displaced individual pulses of different intensities in addition to the individual signals. This is referred to as the "channel response". In this case it is possible for the individual pulses to be distinguished on the receiving side, for example most suitable pulses are selected as "actual signals" (so that other pulses can be referred to as "interference signals"). The separation of this property in long wave package transmissions can generally no longer occur because the receiver only receives a sum or synthesized signal that may have the same frequency as the initial signal, but where different amplitudes and Interference signals and phase signals with phase positions overlap in such a way that unexpected variations in amplitude and phase position can occur. The above effects may make the evaluation of the signal different or even impossible in certain circumstances and are referred to as "intersymbol interaction". If the transmitter and receiver move relative to each other, additional problems may arise in the form of frequency shifts as a result of the Doppler effect.

Many problems can make remote control of underwater equipment very difficult, as well as underwater communication by ultrasonic waves between divers and / or underwater vehicles. In particular, analog information transmission can be used very limitedly. However, this is the frequency used for speech transmission, which uses the fact that humans can identify known words and semantic relationships even in the case of reception for very large noise interference. By proper implementation and agreement in a limited vocabulary, the identification rate may be somewhat improved. However, this process is not suitable for transmitting computer data or other information by physical means, for example. Therefore, there is a need for a suitable digital process in the field of acoustic information transmission.

Current digital systems, in particular digital systems for underwater use, are based on the continuous transmission of sound signals of constant height located in rather narrow frequency bands. In some applications, transmission is performed synchronously at high energy in the wide frequency band to achieve a greater likelihood range and also to exclude information loss due to the acoustic blind frequency range. Regardless of whether the transmission occurs in narrow or wide frequency bands, encoding by serial "clicks" only allows a limited data transfer rate, which allows different volumes of data, such as image transfers from underwater cameras, to differ. It makes transmission difficult and also makes it impossible. In addition to the relatively large energy consumption, this also means "acoustic environmental pollution", and relatively "rigid" systems have significant problems with the Doppler effect.

Apart from the distortion and loss caused by the transmission technology, various types of interference that can be included can be screened out or eliminated, and the signal parameters used to encode the information can be reconstructed at the receiving end in a complex manner. There is considerable difficulty in processing the information contained in. There is no process in the data transmission sector that can solve the above problems in a suitable and optimal manner.

It is an object of the present invention to provide a suitable system and method for the transmission of information that enables high data rates over a wide range.

Another object is to provide a method and system for the transmission of data that can prevent the causes of the aforementioned interference and adapt to different transmission conditions.

In particular, it is an object of the present invention to provide a suitable system for signal processing that can isolate and analyze signal components with the least possible loss of transmission with the highest possible exclusion and high selectivity of inter-symbol interactions from as many channel responses as possible. To provide.

Another object is to provide a method and system for signal processing that can guarantee the most complete compensation for the Doppler effect in the same environment.

Another object is to obtain the best quality of signal processing, thereby presupposing a substantial increase in the transmission range, and in some cases, transmission ranges, even in complex transmission conditions, such as when communicating with or in the presence of moving objects underwater. To create a condition.

This object is solved by a procedure by the features of claim 1 and by a device by the features of claim 31.

Depending on the application, at least two signal components, at least one reference component (BK) transmitted in the reference frequency channel and at least one information component (IK) transmitted in the information frequency channel or (I1; I2; ...; IN An information signal consisting of) is generated, so that several frequency channels and components may each be used. By using this simultaneously, more information units can be transmitted per hour. In addition, discrete states may be used for information components and information frequency channels that form a bit pattern in addition to the reference frequency channel and the reference component. As a substantial difference from previous radio technologies, in the case of the processing according to the present invention no high frequency carriers are used and are modulated to low frequencies. An information signal developed for information transmission is generated, which provides a wave comprising a duplication of at least one information frequency channel as well as a reference frequency channel.

In order to provide a bit pattern in the simplest case, the frequency or tone of the information frequency channel can be switched on and off, so that the presence and absence of the associated signal frequency component is binary information (on / off), i.e. 1 Or evaluates to zero. In this way, bits can be transmitted for each of the information channels. The signal component produces a bit pattern and the information can be encoded in any desired manner.

Although the simplest case above relates to all parameters of the information signal involved, it is possible to change so that different signal parameters in the on state can be identified in different digital states.

Another advantageous embodiment relates to the dependent claims.

According to claim 2, a temporary sequence of bit patterns is generated in a very simple manner.

Claim 3 includes an advantageous basic variant, wherein the frequency channel forms a harmonic row.

If according to claim 4 the reference frequency channel is formed as a fundamental tone or fundamental wave, at least one of the information frequency channels is formed as a harmonic tone or harmonic wave for the basic tone Or if all the information frequency channels are formed as harmonic tones for the fundamental tones, the individual frequencies and tones or signal components form a harmonic series and resonance system. A feature of the system according to the invention is that the basic tone together with the lowest frequency having a wide range can be transmitted permanently during the information transmission, thus forming a permanent bridge between the transmitter unit and the receiver unit in a speaking manner. The reference frequency channel, which is designed as the base tone, does not serve in this case for the actual transmission of information, but as a constant reference for the tuning of other frequency channels and, if appropriate, as described later, determines the determination of the relative phase position. Risk, and will also serve energy providers in the case of using non-linear effects to increase the range of the overall frequency system. Here, it should be noted that instead of low tones, the desired tone of any other predetermined frequency spectrum will be used as the reference tone or base tone if it is more advantageous in a given application with a particular environmental impact.

The information frequency channel is always determined by having limited separation from the reference frequency channel, so that a receiver unit whose corresponding separation and proportional factor is known identifies all other active information frequency channels on the basis of the fundamental tone and continues them in an operable manner. It can be assured that it is necessary to detect the reference frequency channel formed of the base tone so that it can be tuned with. The tuning process can be automated to the extent that the system can be adapted without additional major effort on the widest different transmission state. Self-adaptive tuning of information channels and automatic identification of the basic tones in some of the receiving units represent a significant advantage, in particular for communication between moving objects, which is a consequence of the usual processing by the Doppler effect. This is because the problem will go away if harmonic frequency channel systems are used.

If, in accordance with claim 5, the frequency of the reference frequency channel changes temporarily during transmission, it is subject to the receiver in compensating for a frequency displacement (such as the Doppler effect) in which a constant reconditioning occurs naturally in an adaptive system according to the above basis. Not only can it be done by means of, it also means that it is also possible for a constant time change of the frequency spectrum to be produced in a part of the transmitter unit without damaging the link to the receiver.

One or more frequency gradients may be provided if the time variation of the frequency of the reference frequency channel is carried out stepwise or continuously according to claim 6. This method is referred to below as the Frequency Gradient Method (FGM). In this way, for example, the result that the reflected or interfering signal is removed can be achieved. The modification of the reference or information signal based on the FGM is also referred to as VMT (variable multichannel transmission) hereinafter.

If the variation in the component always affects each other proportionally, pFGM or pVMT is taken as the starting point and vice versa in the case of the affected variation of the component, paFGM or paVMT is taken as the starting point.

The use of FGM is substantially easier to achieve than conventional techniques, such as where sharper and more reliable signals have a particularly fixed frequency channel. In this case, since the working frequency of the information frequency channel changes constantly, all signal components arriving at the receiver unit along different transmission paths at all given time points have different frequencies. Due to this frequency difference, the actual information frequency channel is separated from any interference components that may be present. In other words, the intersymbol interaction can be largely eliminated if not completely, and as a result a clearer image of the information signal emitted by the transmitter unit can be reconstructed by the receiver.

In the case of FGM, the method and system according to the present invention is very flexible because the frequency of the reference frequency channel, and the information frequency channel in a given proportionality synchronized with it, can also be changed almost as desired. Because of deliberately induced frequency shifts, the overlap of several transmission systems can be avoided and unwanted eavesdropping becomes more difficult.

If other signal parameters in addition to the frequency of the information frequency channel and the frequency of the reference frequency channel are derived to generate the bit pattern, the encoding becomes more complicated in a simple manner, and thus the information rate increases.
If the information signal is amplitude modulated according to claim 7, then the time at which the individual information frequency channels can be changed without interference interference known as "glitching" derived from the information signal is induced for modulation. Can be determined at the oscillation node of the specified amplitude.

If a bit pattern is generated at a given time pulse in accordance with claim 8, the pattern can be decoded by the receiver in a simple manner that increases the transmission precision.

If the bit pattern changes in a time pulse according to claim 9, the first part in the time pulse is used to identify which information frequency channel is basically used for information transmission and the rest of the rest of the generation of the bit pattern itself. Can be used to identify usage. Also in this case, the first part provides an additional reference in addition to the reference frequency channel, whereby the parameter of the signal component transmitted in the second pulse section can be determined with very high precision. In other words, the reliability of the transmission can be increased in this way.

Advantageous embodiments according to claim 10 support the possibility of adapting to different transmission environments and user requirements.

In the case of the measurement disclosed in claim 11, the transmission rate can be increased.

In combination with, or instead of, the switching on and off of the individual signal components described above, in particular, as a result of the high reception quality that can be achieved by the use of FGM, the information is encoded in a more detailed variation of a particular signal parameter or parameter combination. Can be. In addition to the frequency in the received signal, the amplitude and phase angle of the signal components have more strictly defined criteria for the originally generated signal, in particular all parameters can be integrated into the encoding. This can be done, for example, by gradual change.

A substantial advantage of the process in this situation includes the fact that internal signal interference can be used for encoding. By virtue of the above relationship, a situation where a bit pattern or symbol has already been identified based on one or more received pulses can be achieved without requiring additional reference to an external reference value.

Thus, for example, the phase angle can be determined in the form of the current relationship of a given time pulse between the individual information component and BK. The encoding method is designated as a relative phase-angle method, or simply RPWM (RPAM). In this way, previous history no longer works, and external time loses importance for signal evaluation. Instead a relative system-internal time is given, which time can be read out on the basis of the cycle time, for example at a specific moment in the BK, the time considered from outside depends in each case on the current frequency. The relative phase angle can be determined in a simple manner, for example, in the case where all signal components, namely the information frequency channel and the reference frequency channel, are first normalized to one constant period time in the evaluation process. But this is only to explain the principle. A wide range of projection and transformation procedures from signal processing are known, which can be derived to determine relative phase angles. The user thus has a wide range for practical implementation. In the result of the FGM in the process according to the invention, and in particular in the result of the pFGM, the range of interference effects can be eliminated, so that the relative phase angle can be determined with high precision, for fine discretization, ie It is important that it can be used for more digital status separation and thus increase in information rate.

For example, the other variant is the last relative calculated previously as referred to as the horizontal signal-interference criterion, rather than directly encoded in the phase angle of the individual components with respect to BK or GT where the information is referred to as the vertical signal-interference criterion. The difference is between the component of the phase angle and the phase angle. The method is designated as the relative phase difference method, which is simply RPDM. In the case of RPDM, the first pulse in each adjacent transmission sequence operates only as a horizontal reference. Under very complex transmission conditions, it may be advantageous for RPDM to be used with a process according to claim 9. Conversely, it is sufficient to use only horizontal signal internal interference for the determination of the relative phase angle in highly desirable transmission conditions. In this case, the reference frequency channel can be used for information encoding. In the case of RPDM besides RPWM, the absence of signal components or the undercutting of certain amplitude thresholds may include additional digital states.

If the number of information channels changes as a function of the transmission path according to claim 12, additional typical high frequencies are used, especially if the distance between the transmitter unit and the receiver unit is reduced, or the previous channel, for example another resonant frequency. The frequency located between them is used, on the contrary the situation where a very low frequency range is used in the case of very large spaces. In this measurement, optimal utilization of the propagation characteristics of the wave signal is achieved, which is particularly important for the use of speech signals. In this way, for example, in an underwater environment, in each case a maximum bit rate and / or transmission distance can be provided which is difficult to achieve so far. The above flexibility includes the principle that the adjustments made for a particular transmission state can be described in relation to the base standard in cases where the base standard allows the particular scope of work to be adequately covered.

In addition to the specific state or ratios presented so far, the information in the process in question can be encoded in instantaneous time changes, i.e., dynamic characteristics.

If the individual information frequency channels according to claim 13 are designed without overlapping into a wide band, the possibility of creating a continuous phase shift of the corresponding signal component and using it for information encoding is provided. The measurement is referred to as the phase gradient method, the phase velocity method (PGM). The distance from the reference tone is typically associated with a characteristic curve of the mean value of that channel. At each time pulse during information transmission, the frequency of the individual information frequency channel may be shifted or slightly changed continuously within a given channel, in each case up to 0.5% or less of the current reference value, thus relating to the reference tone or reference frequency channel. In each case, this results in a continuous, uniform or accelerated phase shift of the individual information frequency channels. The receiver unit not only knows whether a frequency is transmitted on that channel in a given time pulse, but also if a frequency exists, a relative describing its function as a function of the current cycle time, for example in the case of a reference tone or a reference frequency channel. Determining the phase angle and / or measurement parameters, and thus in addition to the actual state or proportional value, a time change for encoding can also be used. From the above inference, a wide range of variations and combination possibilities are derived that can be used for large adaptation of the transmission system under the use of different states and to increase the information transmission rate for the optimization of the device and its cost.

For the simple handling and processing of the information signal, the reference component is separated from the minimum value of one information component after reception according to claim 16.

According to claim 17, the processing in the pair of signal components each carrying information allows to achieve compensation of the Doppler effect with either the optimal component or the reference component, respectively. As a side effect, this processing step can also help prepare the frequency stabilization process. In the case of paFGM, this step can directly lead to the formation of a stable, ie constant intermediate frequency.

Another embodiment according to claim 18 guarantees the transmission of signal components at a constant intermediate frequency (Z1; Z2; ...; ZN + X), which may undergo other processes. One of the advantages is that the frequency window allows a constant intermediate frequency (Z1; Z2; ...; ZN + X) to be optimal for the next filter stage according to claim 20 and at the same time allow the use of certain sharp filters. Include the fact that it can be located at.

When using pFGM or pVMT, as an alternative to the procedure according to claims 16 or 18, the possibility of generating a constant intermediate frequency only by multiplication of the signal received in the current time pulse, for example by the received signal of the previous pulse. This is included without the use of heterodyne frequencies and without prior separation of signal components. Variations in signal processing according to claim 19 are preferably provided with the use of differential phase encoding.

The purpose of another embodiment according to claim 20 is to block the signal optimal signal portion for each signal component, or to filter the signal portion, from the frequency-stable spectrum of the multiple channel response, and in this state from the other signal portion. Minimize interference effects. The interference impact minimization procedure also includes the possibility that signal components in the sequence can be separated from each other if not implemented or fully implemented in accordance with claim 16.

For this purpose, in the simplest case, a specific filter can be used. In particular in each case components which do not require filtering, i.e. components which are not used for evaluation at the moment, are allowed. As a result, a clearly defined representation is required for each information-containing signal component, on which the signal parameters (eg amplitude and / or phase position) used for encoding the information can be reconstructed in the best way. . It also only represents the reference principle.

Of course, more complex methods can be used from known and diverse repertoires of signal processing, and also provide them with the included parameters in addition to the identification of signal components.

In another embodiment according to claim 21, the situation arrives at the fact that no error occurs during signal evaluation due to the processing procedure.

In another embodiment according to claim 22, for the current transmission environment, signal components and channel responses can be identified based on which signal parameters can be determined optimally, i.e. in the best way. In general, the component is the component with the highest energy, ie the energy with the highest possible quality of signal evaluation can be achieved. By channel tuning, the best possible filter settings can be determined by filtering the desired components as accurately as possible and by suppressing the interference effects of possible sidebands on the interference effects and optimal effects of other channel responses. The latter contributes in particular to the increased reception radius and / or increase in information rate. The better and reliably the received signal is evaluated, the more likely it is to use different combinations of parameter changes for finer gradation or even for encoding information.

With ongoing updating of the filter settings and ongoing identification of the most preferred reception components, the optimal reception results can be achieved even under varying transmission environments according to claim 23, so that one advantage of the process mentioned is that It is a fact that no interruption is required for channel tuning.

According to claim 27, the advantage that the Doppler compensation is optimal is achieved.

The method according to claim 28 is preferred and used for processing received signals with heavy Doppler burden, where each signal component is represented by only one channel value.

Other embodiments of the invention are described in the other dependent claims.

The invention is described in detail below with reference to the drawings.

1 illustrates the structure of an information signal that can be used in a method and system following an application consisting of a reference frequency signal and three information frequency channels.

2A shows the information signal of FIG. 1 subject to amplitude modulation.

2B shows a sequence of pulse information signals.

3 shows a schematic diagram for encoding of an item of information.

4 shows the encoding of FIG. 3 including only parallel FGM.

FIG. 5 shows signal analysis at the instant ti of the previous and next interference component by proportional FGM for three information frequency channels in a harmonic relation to each other.

FIG. 6 illustrates the basic principle of improving signal analysis for an interference signal according to FIG. 5 using a reference frequency signal and four information frequency channels.

FIG. 7 shows a schematic diagram using a stepped frequency shift that includes additional changes to the information frequency channel within a time pulse, so that half of the first pulse forms an additional horizontal reference for the RPDM.

8A shows a schematic diagram of encoding with only two frequency steps.

8B illustrates the principle of pensive encoding of an information frequency channel.

9A and 9B show two different phase slopes produced by pPGM.

10 shows the different phase slopes that can be produced by nPGM (top) and pPGM (bottom).

11 shows the basic structure of the transmitter unit of the system according to the present application.

12 shows another basic structure of a transmitter unit including amplitude modulation of a system according to the present application.

Fig. 13 shows the basic structure of the transmitter unit of the system according to the present application according to the first embodiment.

Fig. 14 shows another basic structure of the transmitter unit of the system according to the present application according to the second embodiment.

FIG. 15 shows signal analysis at the instant ti of before and after interference components by parallel FGM by reference to three information frequency channels that are harmonic to one another.

FIG. 16 shows several diagrams illustrating optimal frequency spacing in different applications.

17 illustrates various embodiments of a sequence of processes that follow the present application for signal processing.

18 shows an example of a temporary change in the frequency component of a pVMT received signal, which includes one reference component and three information components under a near ideal transmission state (minimum intersymbol interaction).

19 shows the received signal according to FIG. 18 after transmitting the first information for transmitting the signal component within the intermediate frequency.

20 shows by way of example that as a result of varying channel responses, the intensities of the various spectral components of a given receive component represent substantially transient variations.

FIG. 21 shows an example already shown in FIG. 20 after passing through a sharp filter step.

22 schematically illustrates a sequence of basic variations of a process following an application in which a channel tuning procedure has been executed.

Figure 23 schematically illustrates a very important processing step including different embodiments of the procedure for signal processing.

Fig. 24 shows the basic structure of the system according to the application for signal processing of the third embodiment.

25 shows the basic structure of a system following an application for channel tuning.

1 illustrates how the information signal IS is composed of the reference frequency channel BK formed as the reference component, which is formed of the three information frequency channels I1, I2, I3 formed as the information component and the reference component formed as the basic tone GT. The information frequency channels shown in Fig. 1 are harmonics HK1, HK2, HK3 for the basic tone GT which form the information signal by superposition. It can be seen from the figure that each information frequency channel can be provided depending on the presence or absence of a binary digital portion of information equal to 1 or 0 (see also FIG. 2B).

  The amplitude modulation in FIG. 2A shows the information signal IS of FIG. 1 to ensure a steady flow at the beginning and end of the pulse when the information signal is changed due to a temporary change in the information frequency channel.

The above change is shown, for example, in FIG. 2B, and the shape of the information signal changes from pulse to pulse, and thus in region I, with the overlap of the basic tone and the first and third harmonics (GT + HK2 + HK3). The configured information signal IS is provided. It is transmitted constantly with only the base tone (GT) because there are no second and third harmonics in the next pulse (section II), and then shows the changed information signal due to the superposition of the base tone and the first harmonic in the pulse, The information signal corresponds to another encoded bit pattern (see section III). In this way it is possible to transmit one bit pulse per hour in each said information channel. From this one bit pattern is derived for each time pulse, and the information can be encoded in any desired manner. In general, characters or other symbols may be encoded by the number of available information channels and as a function of the encoding system used.

Using 2, 4, 8, 12, 16 and more information channels, reference is made to the fact that direct compatibility is achieved by several different conventional procedures of electronic data processing.

In FIG. 3 a method is shown, for example in which the word "DolphinCom" is transmitted in a generally known ASCII code, using an information channel. The frequency system forming the information signal comprises four harmonic information frequency channels I1, I2, I3 and I4 as harmonics formed in the reference frequency channel used as the basic tone GT and the channel temporarily changed by the proportional FGM. . The encoding is only operated by switching on and off of the harmonics. The vertical lines in this case always show pulses with the same length. For each pulse, there is a specific bit pattern designated as a symbol. In each case, the two symbols produce one character that is ASCII code to each other. The word "dolphincomb" is shown. Basically, however, other desirable codes can also be used for the encoding of the transmitted information, which allows maximum user space for the manoevre of his programming and makes the system compatible with almost all EDP systems. do. As shown in Fig. 3, the reference frequency channels change continuously, so that the frequencies of the four information frequency channels I1, I2, I3 and I4 are replaced proportionally. In contrast, FIG. 4 shows how the ASCII code word “DolphinCom” can be transmitted using four information channels, so that FIG. 4 changes the reference frequency channel continuously. However, for example, the information frequency channel initially arranged in harmony with the reference frequency is always replaced in parallel with the change in the reference frequency channel.

5 illustrates how substantially sharp and more reliable signal analysis can be performed if the reference frequency channel changes continuously in terms of FGM. In the example shown in FIG. 5, three information frequency channels are selected by the embodiment based on FIG. 3, where in addition to adding the actual signal frequency, the leading and trailing frequencies respectively arrive at the receiver as an interference signal. Thus, each time shift is selected equal to all three information channels. In order to clarify the underlying principle, the pulse display is omitted. The vertical dashed line (starting from ti) clearly indicates that all information frequencies received at a given time ti are different. However, it is important that the above frequency change can separate the actual signal frequency from the interference frequency and that the intersymbol interaction can be eliminated very largely but not completely. In this sense, it is important that the amplitude and phase position of the received and "refined" signal component have a clear reference to the reference frequency channel. When using FGM, specific frequency filters can be used to separate the actual signal frequency from the individual interference frequencies. From Fig. 5, as the spacing of the signal frequencies from the interference frequency becomes larger, the slope df / dt of the frequency change becomes more inclined, i.e., the individual frequency ratio becomes larger. Since all the information frequency channels in the system shown in FIG. 5 are always changed proportionally with respect to each other, a more inclined slope is derived for the higher information frequency channel, thus providing better separation of the current signal frequency from the interference frequency. It is possible.

6 shows a schematic diagram of a system with four information frequency channels and one reference frequency channel, with the above functions and effects in each case having two adjacent interference frequencies. Dotted lines shown in Figure 6 represent the characteristics of commonly used filters. A constant window width of the filter for high information frequencies achieves better increased separation sharpness. In comparison with conventional processes, a noticeably better resolution is achieved overall. For processes following the application, it should be mentioned that it is a particularly high information frequency channel, which always gets strongly attenuated in the transmission stretch, resulting in a receiver with the lowest energy that can be released from noise. From the above process, in the case of interference frequencies located very densely at the actual signal frequency, the sloped frequency slope can be selected for better separation. That is, while the frequency shift is increased, a gentler slope may be sufficient as opposed to larger spacing. For this adaptation, the ready range of frequency displacement patterns can be used or an efficient adaptation process of the slope to the frequency change can be carried out. The latter is easily possible if the connection is bidirectional, i.e. used by the transmitter unit to receive and transmit by the receiver unit. In this way, for example, analysis of the channel response behavior can be carried out between the transmitter unit and the receiver unit and can be exchanged or the corresponding pattern training can be carried out, with the optimum slope being the frequency displacement in each case. Is set for However, if the interference value is very small, the slope of the limited case in a suitable fixed transmission state will be zero.

Here, in order to maximize the transmission rate, it is possible that the pulse frequency will change proportionally to the frequency level of the reference frequency channel, where only a certain number of oscillation cycle times are analyzed globally for the individual components containing the signal. Because it is necessary.

7 and 8a and 8b show another possibility of causing a stepped change in the reference frequency channel. This possibility is an interesting alternative to FGM if the temporal shift between signal frequency and interference frequency is large enough, for example in the range of several milliseconds. In this case, if all channels are simultaneously displaced by jumping from pulse to pulse, possibly higher or smaller frequency values, but remain constant within the step, good separation between signal frequencies and interference frequencies Can be achieved. In this case, it is advantageous that the internal signal ratio is carried out in such a way that the frequency shift is as uniform as possible in every step. This is easily accomplished by proportional or parallel staged changes. The above alternative is generally designated as a frequency jump method or a frequency step method (FSM). Based on FIGS. 7, 8A and 8B, it is clearly shown how the penetrating encoding of the individual information frequency channels is generated by further relative phase encoding. To achieve this, and to increase reliability, a reference signal is sent to all information frequency channels at the beginning of each pulse, which is performed at the second half of the pulse by means of a properly encoded signal. As shown in the side legends of FIGS. 7 and 8A, in each case a distinction may appear between four states, namely four digital stages with a No signal (0) and an RPDM. Thus, each pulse having an information signal composed of three possible information frequency channels HK and reference frequency channels BK, i.e. 5 ³ = 125, which can be used for encoding is derived.

8b illustrates the principle of penal encoding of an information frequency channel. The point indicated by reference characteristic D1 to D4 is the signal component, where its amplitude exceeds the threshold Ao and four different phase angles F or four angle ranges that can be digitally identified by RPWN or RPDM. State D5, which is the amplitude of the signal component less than the threshold Ao.

Another method for encoding information is for example a phase-slope method (PGM) or a phase-velocity method, which is shown in FIGS. 9A, 9B and 10. 9A and 9B show what is referred to as the proportional phase velocity method (pPGM), and FIG. 10 further shows the relationship with the non-proportional phase-slope method (nPGM).

The comprehensive principle of the PGM can be easily explained based on the following initial configuration.

A given starting point is a system in which information frequency channels, ie frequency changes from the harmonic sequence, are executed only by proportional FGM. Any desired information frequency channel may be selected, but the frequency in that channel is not exactly the harmonic of the GT (base tone), but is transmitted at a given pulse interval and the corresponding “reference frequency” (typically 0.5% or less of the reference value). Detune slightly upwards or downwards as compared to (see FIG. 10, top row). Basically, the frequency shift is carried out in this case, but the range is so small that it is difficult at the receiving end to identify it as a modulation on the basis of frequency analysis only, and consequently it is also not possible to interpret it as a digital state value. Frequency continues to exist in the sharp range of the corresponding analysis filter. Depending on whether the frequency is slightly higher or lower than the reference value, in superimposition with GT, the image shown in FIG. 9A or 9B appears, where the relative phase angle increases or decreases continuously. The phase of the frequency of the information frequency channel precedes or follows GT. At that pulse, a phase slope is generated, the direction of which is already identified by the eye and at the same time can be determined without difficulty. In evaluating the phase slope associated with the GT's cycle time at a particular moment, a constant rise in the present case is derived. The precondition for the linear characteristic is maintained even if the internal ratio can easily be changed at a given time in case of continuous frequency change of the entire frequency system. In other words, the relative detuning of the frequency of the information frequency channel does not change with respect to GT. The performance of these conditions is evident by the designated pPGMs, where the small 'p' stands for 'proportional'. Considering the period, for the pPGM, a constant right or left rotation is caused by the information frequency phase relative to the phase of the reference frequency channel, respectively.

The above results can be used very advantageously for information encoding, since in signal analysis, the side of rotation, ie the direction of the phase slope, can be determined more easily than the amount of phase shift, for example. In mathematical terms, this means that only the signal of the first selection of the relative phase shift between the information frequency signal and the frequency of the GT should be determined. That is, it must be determined whether the relative phase velocity is greater or less than zero (see top of FIG. 10).

The above process can be applied to each of the time pulses for each information channel. If used in combination with the simple switching on and off described above, four different discrete states are derived for each information channel at a given time pulse with respect to the GT as follows. The discrete states are 1. no signal, 2. a signal with a positive phase slope, 3. a signal with a negative phase slope, 4. a signal without a phase slope, and thus can be executed without a fourth state if appropriate, This is because quaternary encoding actually includes ternary phase gradient encoding, and since the value of one of the digital values is associated with a single speed value (zero), the trivariate phase gradient encoding is a particular environment. Can't run as reliably as binary encoding However, this problem depends on the specific reception quality since all slopes at the transmit end can generally be generated with high accuracy. In theory, in any case, it is possible in the first example to double the information rate in each channel with respect to simple switching on and off, and in the second it can be increased by the third.

As an alternative to this, very many information frequency channels can also be saved, in which case the frequency spectrum is narrowed overall, thus similarly bringing several advantages. In this case, the transducer does not have to be wide-band in design, especially when the transducers are used in series, saving individual or several components. This has the advantage of reducing the cost of the equipment, for example. On the other hand, for invariant device configurations, the large variability and adaptability of the derived system would be an advantage. For example, giving up high frequencies results in a large transmission range, but there is an option to increase the information rate by shortening the pulse time as a result of excluding low frequencies. There is ample reason to strive for the largest possible cardinality (digital step number) and thus increase the information density in the information channel.

The pPGM can use encoding of different increases in the linear phase slope in addition to the direction, which can be generated by detuning to different degrees of frequency of the information frequency channel, and thus of the discretization that can be achieved in certain cases. Depending on the degree, additional combinations and encoding possibilities are derived.

In combination with the FGM a positive or negative phase slope can be achieved by a frequency change of the information frequency channel, which change is not carried out exactly proportional to the change of the basic tone (see below in FIG. 10).

To identify it from the pPGM, the above modification is designated as nPGM, where 'n' stands for 'non-proportional'. The nPGM is implemented such that, for example, the frequency of a particular information frequency channel is changed at a given pulse interval such that it is slightly faster or slower than provided by the fundamental variant of the proportional FGM. The variants of the PGM are all used in the optimum harmonic frequency series with the proportional FGM.

In other words, the basic principle of an optional method for the generation of phase gradients involves modifying the FGM proportionately enough to produce a small additional frequency gradient that is typically linear for each signal component even within the pulse. Next, different characteristics are obtained differently from pPGM in superposition with GT instead of linear phase variation. That is, the direction and shape of a quadratic curve, which typically corresponds to acceleration angular motion, follows the position of the start and end values of individual frequency changes in relation to the reference value curve (see FIG. 10). In the case of nPGM, up to six configurations can be identified as a function of the cycle time of the GT, based on the symbols of the first and second derivatives of the relative phase angle. If two variants are adopted, a total of eight different symbol combinations are derived for the PGM.

However, other important properties of the nPGM can be used in addition to the symbol, a particular phase angle, for example the phase of the starting or ending value or the phase of the intersection with the reference curve at the mean of the RPWM.

11 shows the basic structure of a transmitter unit for encoding information. The basic principle characterizes the information unit, which allows the information to be encoded at the encoder 3. The encoder encodes the information conveyed by the information unit into an encoding required for a frequency channel consisting of a reference frequency channel and an information frequency channel, and at least one in the form of a medium and a generator 5 for generating the reference information channel. To the information frequency channel of the generator, which is activated by the control module (7). The wave components generated by the generator with a given amplitude, frequency and phase are transmitted to a mixer 9 which is operated by the control module 7.

According to this embodiment, the information signal generated in the mixer is directed to the power amplifier 11 if necessary, which leads the information signal to a converter or a converter cascade adapted to the transmission medium.

According to this embodiment a generator is provided for each frequency channel.

The embodiment shown in FIG. 12 of a preferred transmitter unit is shown in which amplitude modulation is performed on an information signal. To this end, before the information signal, which is divided into separate components for the reference frequency channel and the information channels, is directed to the mixer, the signal is led to a modulator for each frequency channel, for example under the control and operation of the control module. do.

13 shows an embodiment of a receiver unit of a system according to the present application. The converter or converter cascade adapted according to the transmission medium receives the input information signal shown as the acoustic signal in FIG. 13 and induces it into the amplifier 23. Behind the amplifier is a filter 25 for the separation and analysis of the individual frequency channels, in particular for the filtering of the reference frequency channel. In the case of a low pass filter, the signal of the filter is directed to a reference frequency detector 27 which determines the reference frequency and the received strength. The above data is transmitted to the control module 29 of the receiving unit. In parallel, the information signal is selected before entering the low pass filter 25 and directed to a filter 31 which is controllable for the individual information frequency channels. The output signal of the filter is analyzed at the controllable threshold switch 33 and directed to a decoder which decodes the original information.

14 shows another embodiment, for example, showing a phase detector for a phase difference method or a proportional or non-proportional phase-rate method. For the analysis of the phase components, the phase detector 32 is arranged between the controllable filter 31 and the controllable threshold switch 33 and is preferably arranged in accordance with the number of information frequency channels to be analyzed and thus the phase determination. For this purpose, the reference frequency is adopted as the reference in the basic tone detector.

Regarding the preferred signal processing apparatus, different embodiments for signal processing are provided in more detail on the basis of Figs. 15 to 24.

The procedures that follow the application can be implemented in different ways and include two main basic functions that can be combined or applied individually. The basic function may be designated as complete Doppler compensation, abbreviated as vDK, and "channel purification", abbreviated as KR. The basic principles will be explained in each of the first examples before considering the possibility of different technical and procedural variations and combinations.

1. Troubleshooting Doppler

To illustrate the basic principles of vDK, a simple example is chosen from the first example. The lowest frequency channel here supplies the frequency and reference components of the information channel from the harmonic series. That is, the component and frequency represent a whole-figure relationship to a reference frequency. If a change in frequency is made at the transmitter, this is due to pVMT. At the receiver, all signal components are separated from one another in the first stage, for example by a cascade of a band pass filter (BPF). For simplicity, in the first example, it is assumed that the ideal transmission state is maintained, so that each component consists only of the channel response, and all components can be transmitted and received at the same intensity. The above preferred state can be generated according to the propagation of electromagnetic waves in the air, for example. The above example will first be best used to illustrate how full Doppler compensation can be performed by appropriate processing.

Doppler problems include the fact that as a result of relative movement between a transmitter and a receiver, for example, the speed of relative movement is not known precisely, and therefore sometimes an unpredictable frequency shift occurs. This also means that the phase position of the information signal can no longer be accurately determined, which means a significant limitation in all forms of information transmission using phase encoding. The above problem can be somewhat reduced by differential phase encoding, which is to be considered here is not the original phase angle but the change between pulses, but it is not basically solved. However, nearly 100 percent Doppler compensation can be achieved if the individual information components are processed in a suitable way with the reference components. One advantageous solution involves a pair by pair abbreviated pDA, which can be implemented in different ways. One simple possibility is illustrated in detail in the following example.

For all other examples, the processing of the first information component is expressed in herr per time pulse, the frequency of which is f _ik and each velocity ω _ik is two of the corresponding values f _b and ω _b of the reference component. It is a ship. Assuming that the received signal is provided in digital form, the signal components send b and send ik coming from the transmitter have the following form.

Where N is the total number of scans made at a given pulse interval, n is the current number of scans at a particular moment, ts is the length of the time interval at which scanning is performed, nts is discrete time, E is energy, and ωik is the starting Is the phase, [omega] inf is the angle of the information component used for encoding, and the factor k represents the slope of the frequency displacement actively generated in the proportional VMT.

In general k may be any suitable time function and may be positive or negative or even zero. The latter means that the use of constant transmission frequencies is a consideration in certain cases.

Since the phase position of the reference frequency at the transmitter does not change and no longer plays a role, the corresponding value is set to zero in equation (1).

As a result of the Doppler imposition, the received signal components, empf b and empf ik, are distinguished from those transmitted by additional elements.

Where D is the Doppler coefficient, which includes the ratio of the relative speed between the transmitter and the receiver (positive signal for mutual access and negative signal for increasing distance) and the speed of the processed signal on the transmission medium.

Based on the underlined elements of the equation, the Doppler burden of the two signal components depends on the proportional factor, which also represents the ratio of the corresponding transmission frequency. In the present example, the above proportional factor is equal to two.

Since the proportional factor is known, in practice the exact amount of phase shift caused by the Doppler effect no longer plays a role. In particular, if the reference component is in each case transformed in such a way as to obtain the same frequency characteristic as the information signal to be analyzed, then exactly the same Doppler shift is derived from it. In the present example, this Doppler-identical reference RF can be generated from the reference component by multiplication by itself. According to the multiplication rule, it is derived as follows.

delete

로 스캔한 후, 최종적으로 평균(normed) 기준 신호 RF'을 획득하며, 방정식(S)에서 표시된 정보 성분의 위상과 관련하여 다르다. Filter and factor out unwanted sidebands

After scanning with N, we finally obtain the averaged reference signal RF 'and differ in relation to the phase of the information component indicated in equation (S).

The fact that the phase angle of the information component can be determined allows the reference signal to be used to a certain degree as a signal-internal clock.

In a similar manner, it is possible for the criteria necessary in each case to develop from the reference signal for all other information components included in the received signal. The only condition is that the multiplication is performed several times and the filtering is required. In general, information components can be transformed in the same way, which, for example, differs from the example selected here, if the frequency of the information components is located below the reference component or supported in a whole number relationship to them. If not, it may prove useful. In the latter case, the same procedure can be applied so that one side of each pair is formed from reference and information components as often as necessary until the information is each harmonic. However, due to the fact that in each multiplication the number of frequency components contained in the individual spectra is multiplied by the product, attempts should be made to position the channel so that as few steps as possible are required for pair by pair Doppler compensation.

In general, attention should be paid to all applications for phase-encoded signals regarding the selection of a suitable procedure for Doppler compensation in pairs, for example, where no information loss occurs during manipulation of the information component due to ambiguous phase position.

Based on the examples given, the possibility regarding the method of determining the phase position of the individual information components in a simple manner after the above-described signal processing will be described. To this end, for example, a breakdown of the corresponding information component can be carried out in the quadrature function of the reference signal Rf '[n], which is described below.

In the above example, since the reference Rf '[n] has already been provided in the form of cosine, it can be written as follows.

RfC [n] = Rf '[n]

The sinus square component RfS [n] can be obtained, for example, by the formation of the corresponding norm of amplitudes and the first derivative of RfC [n].

Now, the reference cosine square component is recorded for the projection of the information component.

Where N1 is the start and N2 is the end of the individual pulse.

In terms of the fact that the function value is oscillated around the zero value in the second summand, the positive and negative portions are totally balanced. Thus the element generally goes to zero value and can therefore be ignored without significant error.

Thus, for the projection of the information component received as the reference sine square component, the following is derived.

Thereafter, CQ and SQ are regarded as x and y coordinate points respectively in the rectangular coordinate system. The connecting line between the point and the origin and the abscissa includes a phase angle sought (Θ). This can easily be determined with a suitable algorithm. Exemplary representative forms are as follows.

The phase of the received information component appears as the difference between the start phase and the encoding phase of the transmitted wave. That is, in each of the pulses, the difference is invariant with time. For completeness, the phase difference between the preceding pulse and the current pulse can also be used for encoding. If the pulses are designated with index i or i + 1 respectively, then for differential phase encoding are derived as follows.

Inferring, it is possible to determine the phase position of another information component for each time pulse with a high degree of accuracy. Next, this gives the user the possibility of fine discretization of the phase angle and hence the increase in information rate. The above method related to phase angle determination is generally designated later as CS projection.

The vDK in particular forms the basis for the process according to claim 28. 23 is an overview of the most important elements in the flow chart of the vDK. The figure shows that many of the elements described above can be applied in a similar manner to other embodiments.

2. channel purificaiton (KR)

KR includes separation of signal descriptions and identification of the best channel response, with simultaneous minimization of intersymbol interactions. Next, it can include partial Doppler compensation, which can already be extended to a series of applications.

The Doppler effect can no longer actually work, but assumes that reception can be compromised by the overlap of several channel responses. Such transmission conditions frequently occur in acoustic communication between slow moving or stationary objects under water. Each signal component is represented by the full spectrum of the channel response (see schematic diagrams in FIGS. 5 and 17A). The process that follows the application should ensure that the intersymbol interaction is minimal.

In the results of the VMT, different channel responses arrive at receivers with different frequencies, but in the first phase of each component, it is practically impossible to filter out the most desirable channel responses from the associated spectrum, which is largely the response Is very close and the frequency is not fixed (see FIG. 18). In addition, it is difficult for the bandpass filter to be transmitted to be adjusted precisely enough. Initially it was assumed that the corresponding spectra of the reference and information components could be separated from one another (see FIGS. 17B and 17C).

After the multiplication of the reference component by each information component (FIG. 17D), two spectra are obtained at intermediate frequencies, respectively, which are different values and operate at different speeds (FIG. 17E). For example, there is the possibility of filtering in each low frequency band by a low pass filter for subsequent processing. For example, since sufficient computing capacity may be used, if the second sideband does not cause interference, the sideband may also be transmitted. That is, the filter step can be eliminated.

In the next processing step, the minimum value of the other frequency band is multiplied by the auxiliary frequency generated in the system (Fig. 17F), and the characteristic is selected in such a manner that a part of the second intermediate frequency is determined as a result of the multiplication. That is, the frequency associated with it will not undergo any time change (FIG. 17G).

The characteristics of each individual auxiliary frequency (H1; H2; ... HN) are derived from the tuning determined or operatively matched between the transmitter and the receiver for the signal structure used for the information transmission, or at the beginning of the information transmission. It is determined in the framework of the sampling of the transport channel (channel training see below) that is performed.

19 shows that the step can also be reached if only the reference component can be separated from the information component in the first example. By the selection of the appropriate heterodyne frequency, the frequency of the information components each provided for processing (in this embodiment, the first processing) can be stabilized.
The first advantage of the procedure is that the desired portion of the stable intermediate frequency is always located in a limited window by means of a suitable heterodyne frequency and thus can be optimally filtered by a fixed filter such as a low pass filter (FIG. 17H).

FIG. 20 shows that based on a close relationship to the practical implementation in the case of multiple channel responses, it is not possible to create any reliable state based on the stable intermediate frequency spectrum of the phase position, where various different channel responses are different. This is because it can be expressed as a substantial difference in time.

Thus, in the process of a channel training procedure in which the second filter step is in progress (see description below), the best possible separation sharpness is inserted where defined for each component for the fullest channel response. The dashed line in FIG. 17H shows that the flank of the filter can be very sharply defined. As a result, the influence of other channel responses can be minimized in the best way (FIG. 17I).

FIG. 21 shows that for the example very closely related to the actual implementation, in the result of the sharp filtering process from the multiple channel responses still varying in FIG. 20, a certain choice is possible and the influence of other parts can be suppressed. Shows that. Here, the overall process described can be designated as channel purification with partial Doppler compensation.

Parameter determination

The detailed signal analysis can be performed on the signal component purified to the maximum possible degree of the interference channel effect thus processed. In this situation, the amplitude as well as the phase of the signal portion carrying the information can be determined with maximum possible precision and reproducibility. Different amplitude values can be differentiated in the same way, for example by a threshold switch. If a C-S projection is performed to determine the phase angle, if appropriate, the necessary reference oscillation (or sine and cosine portions thereof) can be artificially generated. The latter is not technically a problem, since the system recognizes the setting of the last (sharp) filter stage and hence the frequency of the signal part carrying the information. However, depending on the type of encoding used, the user can select and execute the best from a wide repertoire of known algorithms.

In the form described above, KR may be preferably used together with pVMT (see FIG. 5). However, the KR can be adapted without any problem with paVMT (see FIG. 15). In the case of paVMT, the multiplication of the reference and information components directly results in a constant intermediate frequency, so that if appropriate, multiplication with the auxiliary frequency is unnecessary. Nevertheless, if the above intermediate step is intended to replace the frequency band associated with a particular filter window, for example, this can be easily performed by multiplying by one constant auxiliary frequency in each case. However, this is within the scope of the strategy described above.

KR is therefore suitable for all types of VMT where the slope of the actively generated frequency change is not zero. To identify the modifications described herein and the following modifications, the modifications will be designated KR1. A preferred embodiment of the process according to the present application forms the basis of claim 18. The most important element of this basic variant is again shown generally in FIG.

The above-described modification of KR can be changed, for example, in such a way that the reference and information components are not initially multiplied with each other. In this case, the formation of a stable intermediate frequency occurs directly in one step by multiplying the appropriate auxiliary frequency and the individual signal components. The above procedure provides the advantage that a stable intermediate frequency spectrum does not have more elements than the receive component. After filtering the best channel response for each component in each individual case (channel purification), phase angle determination is described in vDK by CS projection of the component that transmits the information to the cosine and sine components of the purified reference signal. In order to perform similarly to the example shown, or to achieve partial Doppler compensation, it is possible to process the signal portion carrying information along with the criteria. The corresponding frequency adaptation of the criterion may be carried out after the end of the last filter step, if necessary, during multiplication with a suitable and constant auxiliary frequency or by multiplication with a suitable and constant auxiliary frequency. In the second example, the reference component only needs to pass through the filter.

Advantageous embodiments are derived by the modifications described in the preceding paragraphs. A schematic simplified sequence scheme is designated in the overall figure (FIG. 23) as KR2.

However, if the Doppler effect does not do anything, the reference components can be excluded together or used as additional information components. In this case only KR2 is useful. However, parameter determination should be carried out in a manner similar to the procedure described for KR1.

For the sake of completeness, although not represented graphically, the criteria for an optional solution described at the outset must be made once again, where, for example, a stable intermediate frequency step in the framework of pVMT is also Only by multiplication of the received signal in sequential pulses can be achieved without prior component separation. The step similarly includes partial Doppler compensation. A feature of this case is the fact that, depending on the frequency stroke, the stable intermediate frequencies of the associated channels are located in separate windows located close to each other. Due to this transmission, a very complex signal structure is obtained. In particular, if very large numbers of information channels are used, care must be taken to avoid possible overlap of the vector cross product. A sharp cascade filter can be used to separate the channel response.

Finally, in the description of the filter system, a reference is made to the basic principles of the procedure. Indeed, it may be understood that more complex methods of signal processing and signal analysis are used, the meaning of which includes operational steps described in similar or different forms. In any case, the principle is the same.

Complete solutions

The basic principles of the technical procedure for full Doppler compensation and different variations of channel purification have been described separately (including partial Doppler compensation), examples of applications will now be considered, in which case the reception is not only by multiple channel responses Hindered by strong Doppler effect. Combinations of the above interfering elements, for example, make communication with or between objects moving under water difficult.

For this situation, possible solutions include, for example, combining both vDK and KR2.

After separation of the reference and information components, Doppler compensation is performed in pairs, at least one of the signal components considered in pairs, as first described by the vDK, or if appropriate, both have exactly the same frequency cycle, so Equally deformed in a manner that has a heavy Doppler burden. Optionally, each unwanted sideband may be filtered and the remaining reference signal portions may be averaged again.

Both components are typically multiplied by multiplication with the same auxiliary frequency (with the same slope as the related component, but replaced approximately simultaneously), thus transmitted at a stable intermediate frequency and then in each case to the next filter stage for purification Is processed. To this end, sharp filters can be individually adjusted for each component, if appropriate. In an ideal situation, the relevant filter settings can be considered in the fine tuning of the referenced auxiliary frequency.

As a result, a purified signal is obtained not only for the information component but also for the largest intersymbol interaction for the reference. After the above "purification" (including threshold analysis, if available), parameter determination is carried out, for example in accordance with the procedure described for vDK or KR2, thus including the criteria. By processing in pairs of individual information components, full Doppler compensation is achieved.

Embodiments of the process according to the above-described application form the basis for advantageous embodiments of the process according to claim 8. In FIG. 23 this was designated Komp.1.

Another possible solution involves a suitable combination of vDK and KR1 (see simplified diagram of the sequence plan of Komp1.2 in FIG. 23).

In this case, after separation of the information component and the reference, Doppler compensation in pairs is performed next. Thereafter, one of the two components is replaced in parallel by an appropriate amount by multiplication by the auxiliary frequency generated in the system, but in this case both partner components are multiplied in turn by the given amount, accordingly shown in FIG. 17G. Processing steps, i.e., a plan of stable intermediate frequencies, are achieved. The process then continues with two filter steps and parameter determination following KR1.

The second complete solution involves the fact that by the projection of the information component to the Doppler-identical reference, the effect of the frequency displacement caused by the movement is completely eliminated. However, this allows the criterion to be "used up", ie it is no longer needed. The main advantage of the process lies in the fact that only one constant auxiliary frequency in the system needs to be generated to ensure that the desired sidebands of the stable intermediate frequency are accurately positioned in the optimal frequency window for filtering. In the most preferred case, one frequency and the same frequency can be used for all component pairs. In principle, the possibility of using said auxiliary frequency arises from "purification" as a reference for phase analysis. In practice, however, filter settings are well known in the system, since efforts are made to adjust sharp filters for each signal component (in this case, already incorporating the corresponding processing combination of individual information and reference components). If necessary for analysis (see KR1), accurate matching criteria in the system (including sine and cosine square components) are possible without the problem of artificially generating the criteria.

Channel training and channel tuning

The channel training technique already described several times ensures that the signal structure best adapts to the transmission state in each case on the basis of a suitable test signal, first of all, and also ensures that the receiver can always perform component separation as needed. . If the previous state is realized, channel tuning is performed, which is absolutely recommended for variants involving not only the process according to claim 1 but all other channel purification processes. To this end, the selection that can be used is used to transmit a rather long signal with the features already provided for signal transmission without encoding. In this case, it is up to the user to determine whether all frequency channels will be used simultaneously or whether channel tuning will be performed on the basis of the test signal, in which case the test signals in turn comprise a reference component and one or more information component (s). Each procedure must match a variant of the signal processing procedure selected. The received test signal passes through all the processing steps provided in the relevant variations leading to the formation of a stable intermediate frequency. At this level, the analysis is provided for analysis of the energy density distribution in a given frequency spectrum (or mixed components formed from individual information and reference components, respectively). For this analysis, for example FFT can be used. Based on the evaluation results, in each case the best suitable channel response (typically the response with the highest energy) is selected respectively, for which the best possible setting of the "sharp" filter is executed and stored. Once the relevant settings for all components have been determined, the actual information transfer can begin. The filter setting is maintained until the next channel tuning.

Especially in the case of acoustic data transmission under water, the transmission status is often not temporarily stable. In this case, one possibility is to repeat the minimum channel tuning at a suitable time interval, ie to update the settings of the regularly sharp filter.

The use of longer encoded signals for channel tuning provides good statistical reliability, but also means that information transmission is interrupted approximately shortly. However, the interruption can be avoided. One preferred advantage is provided by the process according to claim 23. In this situation, a continuous update of the mentioned filter settings is performed in parallel with the signal processing procedure or on the basis of the progress signal received as a component thereof, and thus continuous channel tuning is performed. As a result, reception results from a plurality of pulses are appropriately incorporated into the calculation. The above alternative solution basically imposes significant requirements on the evaluation system.

Determination of relative speed between transmitter and receiver

Instructions may be derived from the received signal that may be helpful for the current change in distance between the transmitter and receiver. Signal processing following the above-described procedure aims at reconstructing (transmission) parameters (particularly phase angles) for each signal component in the best manner. To achieve this, the components are intended to be treated with each other in such a way that the Doppler portion is removed. This is treated as an interference value. The Doppler portion above contains information items in the form of Doppler coefficient D = v / c (see FIGS. 4 and 5) that are not critical for actual data transmission but can provide an indication of the relative speed v between the transmitter and receiver at a particular moment. Include. Doppler coefficients may be determined by suitable signal processing methods. The speed c of the signal propagation is known in general or can be measured in the framework of channel sampling, and v can be determined or evaluated with relatively high precision.

As an example one possible solution is outlined below;

To this end, any desired receiving component (uncoded reference component) can be derived. If necessary, this can be individually reduced along with KR2 back to the channel response. Since the signal structure used by the transmitter is known to the receiver, and because the phase position can be determined according to one of the preceding procedures, it is possible to generate an amplitude-standardized reference signal within the system. This allows comparing the reception characteristics related to the frequency characteristics and phase position except for the Doppler portion, which characteristics are not actually known. After projection of the receiving component into the sine and cosine components of the reference and filter (LPF), the pure Doppler portions are obtained in a simple form of sine and cosine oscillation, each having the same amplitude. The arc tangent function gives the polar angle Dωnt _s . Since D omega n _s is known, division D is derived and c multiplied by D finally gives v (see Doppler arrangement in FIG. 23 for a schematic sequence design).

It would be advantageous for many applications if the above advantageous additional information could be obtained without further measurement effort.

Recognition of the Doppler portion will also contribute to the improvement of the actual signal processing. Therefore, the auxiliary frequencies generated in the system within the frame of the KR can be more accurately tuned to the structure of each individual receiving component, which reduces unwanted Doppler effects in actual signal analysis in a better and simpler way. By incorporating the above measurements and, where applicable, by repeated applications, at least a temporary process optimization can be achieved in addition to the improvement of the evaluation results, which is an inevitable rapid data increase due to an increase in the Doppler resistance. Can be reduced, the intermediate filter can be removed, and the sequence as a whole can be faster. At the heart of the process, all simplifications are an advantage of online evaluation. Even if Doppler evaluation requires additional effort, overall savings can be achieved because the core routines of signal processing can be managed with less hardware and software capacity, if appropriate.

In addition, if the information on the current Doppler shift is provided in a form already prepared by an external measurement system, the improvement of the latter process and the simplification of the process according to the present application can be implemented in a simple manner.

In Fig. 24, the basic structure of the signal processing unit is shown. The basic principle characterizes a filter unit, the embodiment consisting of two filter elements BPF1 and BPF2 which can be connected and controlled in parallel, dividing the received signal into reference and information components.

The two signal parts are first led to a converter unit comprising converter 1 and converter 2 along the medium for frequency conversion. Doppler compensation here is generated in pairs before the two parts are purified by multiplicator 1 for transmission at a stable intermediate frequency by a heterodyne frequency carried by the generator or by a suitable auxiliary frequency.

As a medium for suppression of interference elements, the above embodiment is used in the first example of filter LPF1 and, if appropriate, the second filter, in which case the filters are connected downstream of the sequence of the multiplier and the filter is an unnecessary sideband Remove it. Next, the readout of the optimum signal part occurs by the sharp filter BPF3 connected in series, respectively. Next, the signal portion is transferred to a parameter analysis medium, i.e. in this embodiment a parameter analysis module. If necessary, the parameter analysis module can be supplemented by a generator for a reference signal connected to BPF3.

The signal parameter for encoding is output at the end of the processing unit for each information component.

25 shows a basic design for channel tuning that can be used to advantage in the same environment. By the difference from Fig. 24, in this embodiment, the signal component is induced in the medium for tuning in which the FFT unit forms a module for analysis of the frequency spectrum, followed by LPF2, and a unit designated as 3 is formed for evaluation. These results are driven into the control module and each starts the optimal filter setting for BPF3.

The possibilities or applications and applications of the process following the present application are described in detail below.

As another option, in the above process, in the case of a reduction in the distance between the transmitter and the receiver, it is possible for another frequency channel between the original frequency channel or even the high frequency channel to be used without any problem, or in the direction of the high frequency. Can replace the entire spectrum. In this situation, it is possible to use the weakening of the interference effect also with the decrease in the transmission distance. To accomplish this, transmitters and receivers need to be designed for a wide frequency spectrum and need to be equipped with the capacity to transform the encoder. In some of the receivers, the recognition with the new input frequency can be carried out automatically, or a change to the new mode of operation can be informed to the receiver by the transmitter in a suitable manner (e.g. the last information package). The individual tone channels should be far enough apart from each other so that they can easily distinguish the transmission state from the receiver. Conversely, an increase in distance may require an increase in the distance between the channels by shifting the entire spectrum to lower frequencies or by surrendering the upper channel or even omitting or dropping proportions of intermediate steps.

The preselection of the frequency band provided in the basic variant of the process aims to obtain a desired overall arrangement for the energy such that the tone or frequency is an integer multiple (harmonics) of the tone having a resonant or lower frequency.

By using the harmonic frequency series, the possibility of using the nonlinear effect of the propagation of speech is provided, and the possibility of signal transmission over a considerable distance is also provided. The sound wave is a longitudinal wave, where sections with high and low densities are alternated. However, since the speed of sound waves depends, in particular, on the density of the medium, the dense parts spread faster. The flank of the original sinus oscillation will become progressively asymmetric, ie the sinusoidal oscillation will be further modified for example in the direction of the serrated oscillation. In physical terms, this means that energy is transferred to harmonics. In water, the above effect can be recognized after passing several kilometers. If more than one harmonic is transmitted on the base tone, because of the harmonic relationship due to the mentioned nonlinear effects, each receives additional energy from the deep tone. As a result, they decay very quickly and remain long above the level of fundamental noise, thus achieving greater coverage. Since the range of the system as a whole is determined in the first example by the highest frequency band respectively, a large transmission radius is achieved overall. For this purpose, the basic tone should be permanent, and if possible all other tones should be transmitted with high energy.
Due to the high variability of the system, other characteristics of the transmission path can also be used. For example, as a result of the layer inhomogeneity of water, specific transmission channels with their oscillation characteristics are formed. Depending on the characteristic eigenvalues involved, different modes can be easily stimulated, which generally have a relatively low frequency, but can carry very far when compensated. In principle, the possibility that the frequency band of the transmission system is tuned in this mode is maintained. In addition to this, the transmitter and receiver should understand each other in a suitable manner.

If the relative speed between the transmitter and receiver is so small that the Doppler effect can be neglected, it is possible to implement a uniform frequency change for the entire system as an alternative to the proportional FGM described above. In this case, the particular "offset curve" or "melody" expressed in terms of the image is determined in each case after the analysis of the particular transmission state and all frequency channels added to the melody (see Figure 4). Or specified. The method is represented as parallel FGM. A feature of the method is that the result of the parallel bee is that the same slope is always derived for all frequency channels, i.e. for the same drift rate, and as a result the optimum separation of the actual signal from the interference component in the ideal state is achieved for the entire spectrum It is in the fact that it can be achieved. The process modified in this manner also has the advantage that the frequency spectrum does not spread with an increase in the reference frequency. Due to the more promulgated bundling, the upper tone does not easily cause the risk of being cut out by sliding into a frequency range with a very small propagation radius. Thus, the upper frequency range can be used efficiently, which frequency range is faster in terms of information transmission. Parallel FGM can be achieved more easily in practice because transducers can often operate in a limited frequency band and the use of the corresponding converter cascade is not always possible.

In general, in the case of a parallel FGM, the receiver should be informed in an appropriate manner as to how the individual frequency channels can be tuned in relation to the GT (base tone). In principle, the switchover between the proportional FGM and the parallel FGM is not a problem for the frequency tuning of the information frequency channel since it only involves switching between multiplication and addition.

In the case of phase jumps that cause problems in pulse transmission, the pulse-related amplitude modulation described above can be used. Another method of minimizing interference impacts includes using an encoding process that excludes the provision of two consecutive pulses, the tone, in the information channel. The same effect can also be achieved by multiplexing, for example the alternating operation of even and odd information channels. As with many other parameters, it is obvious that the frequency-related rate characteristics of individual transport channels can be determined by specific sampling or in the process of intercommunication, and must be taken into account in signal generation.

It can further be seen that the reference frequency channel can be used as an additional information channel if necessary and if a particular transmission state allows it.

Claims (47)

As a method for transmitting information, Generate at least one information signal IS consisting of a reference component BK and at least one information component I1; I2; ..., IN; At least one of the components changes frequency continuously with time during information transmission; The reference component (BK) and the information component (I1; I2; ...; IN) in each case form a discontinuous state for providing a bit pattern. The method of claim 1, Wherein the at least one reference component and the at least one information component continuously change in frequency with respect to time during information transmission, and a frequency interval defined between the reference component and the information component is determined according to a predetermined time function. Information transmission method characterized by the above-mentioned. The method of claim 2, The frequency interval may be constant with respect to time or vary in proportion to time. The method according to any one of claims 1 to 3, And the frequency of the at least one component continuously increases in the transmission interval. The method according to any one of claims 1 to 3, And the frequency of the at least one component continuously decreases in the transmission interval. The method according to any one of claims 1 to 3, The slope of the frequency change is adjusted as a function of the interference frequency position of the individual frequencies of the components in order to minimize intersymbol interactions and / or moderate frequency drift to prevent mutual overlap of the various transmission systems. Transmission method. The method according to any one of claims 1 to 3, And the initial frequency of the components changes at every transmission interval. The method according to any one of claims 1 to 3, And a frequency band of two or more components, which are in a range in which the frequencies change. The method according to any one of claims 1 to 3, At least one of said components, in particular at least one reference component (BK), is arranged in an individual frequency band. The method according to any one of claims 1 to 3, And wherein said bit pattern is determined by a change in frequency, amplitude and / or phase angle or dynamic phase characteristics at a given time clock. The method according to any one of claims 1 to 3, And wherein said bit pattern changes within a time clock. The method according to any one of claims 1 to 3, And the number of said information components (I1; I2; ...; IN) varies as a function of said transmission path. The method according to any one of claims 1 to 3, The reference component (BK) is used as an additional information component (IN + 1). The method according to any one of claims 1 to 3, Said reference component (BK) and said at least one information component (I1; I2; ...; IN) are formed as acoustic waves or electromagnetic waves. The method according to any one of claims 1 to 3, In order to process the information signal after reception, the reference component (BK) is separated from the at least one information component (I1; I2; ...; IN). The method according to any one of claims 1 to 3, And the reference component (BK) and the information component (I1; I2; ...; IN) are processed in pairs. The method of claim 15, The information component and the reference component, or the reference component and the information component processed into pairs, are multiplied by auxiliary frequencies H1; H2; ...; HN + X so that the standard intermediate frequencies Z'1; Z '2; ...; 2' N + X) information transmission method characterized in that. The method according to any one of claims 1 to 3, With respect to the proportional change of the frequency channels, the standard intermediate frequencies are formed by pairwise processing, in particular by multiplying a signal received with a preceding pulse and a current time pulse. The method of claim 16, The execution time difference of the multipath portion included in the received signal is individually used as a function of the rise of the frequency slope used in the transmission signal after the transmission to the standard intermediate frequency is expressed in the form of a frequency difference; From the spectra of the standard intermediate frequencies Z'1; Z'2; ...; 2'N + X, an optimal signal component is selected based on the filter devices respectively and / or from the associated signal components. Information transmission method, characterized in that the parameters are determined. The method of claim 15, Method for transmitting information, characterized in that channel tuning is performed at a specific time interval. The method of claim 15, During the information transmission, the identification of the best received component and / or the update of the filter settings are performed based on a suitable analysis of the spectrum of the standard intermediate frequencies, resulting in continuous channel tuning without interruption of the actual information transmission. Information transmission method characterized by the above-mentioned. The method of claim 15, The transmission-induced Doppler frequency shift is determined internally in the system in consideration of the generation of auxiliary frequencies. The method of claim 15, And in each case said processing is performed in pairs based on internally generated components with appropriate frequency characteristics. The method of claim 15, a) The reference component BK is transmitted to the converted reference component BK ', and the at least one information component I1; I2; ... IN is converted into the converted information component I1'; I2 '; ...; IN '); b) determining signal parameters related to information encoding based on the projection of the transformed information component I1 '; I2'; ...; IN 'into the sine and cosine component of the transformed reference component BK' Information transmission method characterized in that. The method of claim 24, Wherein said reference component is transformed into a Doppler-identical reference component (RF) having an information component to be processed in each case by a suitable transformation such that the product of the two components produces a frequency-stable signal. . The method of claim 24, a) the transformed information components I1 '; I2'; ...; IN 'are multiplied to generate a first value CQ; b) the transformed information component (I1 '; I2'; ...; IN ') is multiplied by a reference component (RF) derived in time to generate a second value (SQ); c) a correlation is formed between the first value and the second value to obtain a final value that is invariant with time that is affected only by information parameters invariant with time. An information transmission system designed to carry out the method according to claims 1 to 3, further comprising a transmitter and a receiver, wherein an information signal (IS) is transmitted between the transmitter and the receiver. The transmitter processes the reference component (BK) and at least one information component (I1; I2; ...; IN) in order to continuously change frequency with time and to provide a bit pattern; The receiver comprises means for obtaining an information signal (IS) consisting of at least one information component (I1; I2; ...; IN) and one reference component (BK), wherein the at least one component An information transmission system in which the frequency changes continuously with time. The method of claim 27, wherein the transmitter, At least one generator for providing said reference component (BK) and said at least one information component; A first control module connected to said generator for determining said frequency response; An encoder or modulator connected to the control module for conversion of the information by signaling means; And An information transmission system comprising a mixer connected downstream of said generator and said encoder or modulator. The method of claim 27, The receiver comprises at least one input, one processing unit and at least one output, the processing unit means for separating and converting the signal components for transmission at a standard intermediate frequency, separating or suppressing interference portions. And means for serially connecting the means and the parameter analyzing means. The method of claim 29, The separating and converting means comprises at least one multiplier for multiplying at least one information component (I1; I2; ...; IN) with the reference component (BK), wherein the multiplication is a standard intermediate. The downstream means for forming a spectrum of frequencies and for suppressing the interference portions comprises at least one filter unit, the filter unit filtering the desired signal portions and forwarding them to the downstream means for parameter analysis. Information transmission system. The method of claim 29, The separating and converting means comprises a filter unit having a control module connected upstream of the multiplier and comprising at least two filter elements connected in parallel, wherein at least one component is initiated from the other signal parts by the filter unit. Information transmission system, characterized in that separated to. The method of claim 30, The separating and converting means, in addition to a multiplier for processing BK and IK in pairs, provides an additional multiplier for transmitting the signal component within a predetermined range of standard intermediate frequencies by a frequency invariant intermediate step, and for providing an auxiliary frequency. An information transmission system comprising a unit having a module. The method of claim 29, The separating and converting means comprises at least one module for providing an auxiliary frequency called in the form of at least one multiplier and one or more generators or memory units, wherein the separating and converting means are provided in which the reference and information components are predetermined. Means for transmission separately from each other within a range of standard intermediate frequencies, having at least one filter unit downstream of the separation and conversion means and connected with means for suppressing the interference portion, the filter unit being the desired signal portion And each filtered and separate interference portion from the individual spectra of said standard intermediate frequency to provide to said downstream means for parameter analysis. The method of claim 29, The separating and converting means further comprises at least one converter for Doppler compensation. The method of claim 29, And said interference portion suppressing means further comprises controllable filters. The method of claim 29, Said parameter analyzing means comprises at least one multiplier for processing one information transfer signal component in pairs by at least one reference oscillation provided internally by the generator or from a memory or by a reference component, and And an analysis module. The method of claim 29, Further comprising a tuning unit connected downstream of said separating and converting means and also upstream of said parameter analyzing means, a module for frequency spectrum analysis and a module for suppressing interference parts. Information transmission system. The method of claim 32, And a Doppler analysis module connected to at least one of said auxiliary frequency providing modules and / or an additional evaluation module for determining the rate of change of distance between said transmitter and said receiver. A transmitter configured as part of an information transmission system according to claim 27. A receiver configured as part of an information transmission system according to claim 27. delete delete delete delete delete delete delete

Images