RU2232473C2 - Data transfer method and system - Google Patents

Abstract

FIELD: radio engineering.

SUBSTANCE: proposed method and device are primarily used for long-distance communications with mobile objects affording desired correction of Doppler shifts by generating information signal incorporating at least one reference and one information channels; these channels are used to generate discrete states for shaping bit pattern.

EFFECT: enhanced data transfer speed and reliability of information signal correction.

37 cl, 29 dwg

Description

The invention relates to a method for transmitting information and to a system for implementing the method.

In many areas of technology, waves are used to transmit information, for example electromagnetic or acoustic, which, being emitted by the transmitter, reach the receiving device, propagating through a special waveguide or freely in a specific medium. If the receiver and transmitter are matched, for example, in frequency or in the selected frequency range, then a connection is established. Through such a connection, information can be transmitted in various ways.

The transmitted information (for example, speech, text, a sequence of numbers, music, image data and other data) can be converted or encoded for emission by the transmitter in the form of wave signals into the transmission medium. The receiver receives signals, converts them to their original form, i.e. decodes them, and gives the corresponding output information.

Depending on the form of encoding information distinguish between analog and digital transmission of information.

In analogue information transfer, the transmitted values are presented in a continuous spectrum of physical states. This usually occurs in the form of modulation of the amplitude, frequency, and / or phase of the carrier waves. During a given time interval, large amounts of information can be transmitted.

With the digital communication method, they are limited to certain discrete states. At present, there are still no restrictions on the speed of information transfer in the case of the use of electromagnetic waves, since the frequencies of the corresponding carrier waves are very high, and therefore various discrete states can be realized at extremely short time intervals.

In some information transmission media, for example in water, communication through electromagnetic waves is only conditionally possible, since these waves propagate in water only over very short distances. In this environment, however, it is possible to transmit information by means of sound waves that can propagate over long distances. Sound waves can be modulated in the same way as described above. These waves, however, are mechanical pressure waves, which differ from electromagnetic ones in the nature of propagation and, in addition, significantly lower frequencies, which ultimately affects the speed of information transfer. Regarding the nature of the propagation of mechanical waves, for example, there is a significant dependence of the speed of sound on certain environmental conditions.

The various problems that occur with the acoustic method of transmitting data are explained below by the example of the transmission of sound signals under water. When propagating in an aqueous medium, sound waves emitted by a transmitting hydrophone can be reflected or deviated, for example, from the surface of the water and / or, depending on the depth, from the bottom, various objects, air bubbles, suspended matter, as well as from layered inhomogeneities of the water channel . The various components of the sound waves arrive at the receiver with different amplitudes and phases, depending on the length of the propagation path, the angular relations and the acoustic properties of the respective boundary surfaces. The useful signal at the receiving point can, as a result of interference, be unpredictably amplified, weakened, distorted or compensated, and accordingly, reception, due to the so-called echo, can be disrupted.

For a more complete explanation of the problem, we first consider the simple case when a very short signal of a certain frequency is sent, the so-called continuous-wave pulse. Then, under the above conditions, not only a single signal can be received, but a whole group of time-shifted individual pulses of different levels. This effect is called channel response. For the described case, it is possible to distinguish between individual pulses when receiving and, for example, to select the best suitable pulse as a "signal proper", while other pulses can be considered as interference and processed accordingly. Such a possibility of distinguishing multipath pulses from each other is usually not possible for longer wave packets. This is because the receiver registers the total (or composite) signal, which, although it has the same frequency as the emitted signal, is a superposition of signals and noise with different phases and amplitudes, which leads to unforeseen jumps in phase and amplitude in the total signal. This negative effect, which makes it difficult or impossible to process the signal, is called "intersymbol interference". In the case of relative movements of the receiver and transmitter, an additional problem of frequency offsets arises due to the Doppler effect.

These problems complicate underwater communication, such as, for example, ultrasonic communication between divers and / or underwater vehicles. Also difficult is the remote control of underwater devices. Until now, the analog communication used is only conditionally practical. It was used and is often used to transmit speech signals, using the properties of the hearing organs to recognize familiar words and semantic context even with very noisy reception. Through appropriate training and agreement to use a limited list of words, some improvement in the degree of speech recognition can be achieved. This approach cannot, however, be applicable for transmitting, for example, computer data or other information algorithmically. For this reason, a suitable method of digital data transmission is also being searched in the field of acoustic communication.

Modern technical digital systems, especially for use underwater, are mainly based on the transmission of tonal signals of constant height, lying in a more or less narrow frequency band. To achieve the longest transmission distances and to eliminate information loss in acoustically "dead" zones, high energy signals are emitted in a wide frequency band in some systems. Regardless of whether signals are transmitted in a wide or narrow band, the coding method through successive “clicks” (amplitude-modulated signals) allows only a very limited transmission speed to be achieved, which makes it difficult or impossible to transmit large volumes of information, for example, images obtained underwater camera, etc. In addition to the relatively high energy costs, which also means "acoustic pollution", the currently known relatively inflexible systems also have big problems due to Doppler shifts.

In addition to the distortions and losses technically caused by the transmission, there are also significant difficulties in processing the information contained in complex signals to eliminate or attenuate (when receiving) various distortions and restore the signal parameters used for information coding. Among the communication methods, there is still no approach that allows a sufficient and optimal way to resolve the totality of the problems mentioned.

The objective of the invention is to provide a method and an appropriate system for transmitting data that would allow communication over long distances with a higher transmission speed.

In addition, the object of the invention is to provide a method and system for transmitting information that is resistant to the above interference and adaptable to a variety of communication conditions.

Particularly important is the task of creating a method and an appropriate system that ensures the efficient isolation and subsequent analysis, if possible, of those components that have undergone the least losses in the channel to best exclude the effects of intersymbol interference.

In addition, the object of the invention is to provide a method and an appropriate system for processing signals, providing as much as possible full compensation for Doppler shifts.

By improving the quality of signal processing, it is planned to create the prerequisites for a significant increase in the transmission speed and, if necessary, the transmission distance in difficult communication conditions, for example, between moving underwater objects.

These tasks are solved in the claimed method using the characteristics of paragraph 1 and in the system using the characteristics of paragraph 31.

In accordance with the invention, an information signal is generated consisting of at least two components — at least one reference component sent in the reference frequency channel and at least one information component transmitted in the information frequency channel, so that several frequency channels or frequency components. Through their simultaneous use, more units of information per unit of time can be transmitted. In the future, both the reference channel and the information channel provide discrete states forming a bit pattern. A significant difference compared with radio engineering is the fact that a non-high-frequency carrier wave modulated by a low-frequency wave is used. The generated information signal used to transmit information is an oscillation, which is a superposition of the reference channel and at least one information channel.

To form a bit pattern, in the simplest case, the frequencies (or tones) of the information channels can be turned on or off, and the presence or absence of the corresponding frequency components of the signal is evaluated as binary digital states (on / off), i.e. 1 or 0. Thus, one bit of information in each information channel can be transmitted. The components of the signal together form a bit pattern in which information can be encoded in any way.

In this simplest case, in the “on” states, other signal parameters may also vary, so that additional digital states can be distinguished.

Other preferred embodiments of the invention are presented in the dependent claims.

In accordance with paragraph 2, in the simplest way, the time sequence of the bit pattern is formed.

Clause 3 presents one of the preferred embodiments of the invention in which the frequency channels represent a harmonic sequence.

If, in accordance with paragraph 4, the information signal is generated as the sum of the reference channel, which is the fundamental tone (main wave) and at least one information channel, which are in harmonic relation to each other (for example, the information channel is the overtone of the fundamental tone) , or also, if all information channels are in harmonic relationships with the fundamental tone (main wave), then the individual frequencies (tones) or signal components make up a harmonic sequence or a consonant istemu. A feature of such a system is that the fundamental tone, having the lowest frequency and the greatest propagation range, can be emitted continuously during the transmission of information, creating a kind of continuous bridge between the emitter and receiver. The reference channel, formed in the form of a fundamental tone, in this case does not actually serve to transmit data, but serves as a constant support for coordinating the remaining information channels and, if necessary, as will be further presented, for coordinating the relative phase values, and also possible in quality of energy “recharge” in the case of using nonlinear effects to increase the transmission range of the entire frequency system. It should be noted that instead of the lower frequency, any other tone of a given frequency spectrum can be used as a reference channel (or the fundamental tone), if this is more preferable for certain environmental reasons or for a specific application.

By setting a certain diversity of the information frequency channels relative to the reference channel, it is guaranteed that the receiving device, which knows the corresponding diversity or proportionality coefficients, only needs to find the reference channel formed in the form of a fundamental tone in order to recognize all other active information channels on this basis and ensure their prompt coordination. This coordination process can be automated so that the system, without large additional costs, can be configured for a wide variety of communication conditions. The automatic recognition of the fundamental tone and the corresponding adaptive matching of the information channels from the receiver side creates significant advantages, especially for communication between moving objects, since, for example, problems caused by Doppler shifts in other methods disappear here, for example, if a harmonic system is used frequency channels.

If, in accordance with paragraph 5, the frequency of the reference channel in the process of data transfer varies over time, on this basis not only the adaptive system is constantly tuned to compensate for natural frequency shifts (Doppler effect, etc.). Much more important is that from the side of the transmitter, a controlled change in the frequency spectrum is realized absolutely consciously, without creating a danger of contact loss for the receiver.

If the change in the frequency of the reference channel occurs continuously or stepwise, according to paragraph 6, then one or more values of the gradients of the change in frequency are provided for use. This approach is hereinafter referred to as the method of frequency gradients (FH). By this method, it is achieved that, for example, reflections or interference signals can be eliminated. The change in the reference and information components based on the PG method is further defined as a method of variable multi-channel transmission (PMP).

In the case where the change in the components is carried out proportionally, this kind of the above method will be designated as FCG or pFMP, while the variation of the method with parallel component change will be designated as FGP or pFMP.

By applying the PG method, it is possible to carry out a more accurate and reliable signal analysis than by known methods, in particular using frequency channels with time-invariant frequencies.

Since in this case, the operating frequencies of the information frequency channels are constantly changing in time, all signal components arriving at a given moment in time to the receiver in different ways, have different (instantaneous) frequencies. Based on these frequency differences, the actual information frequency channels can be separated from the existing interference components in each case; this means that intersymbol interactions can be basically or even completely eliminated, while on the receiving side a much clearer display of the information signal emitted by the transmitting device can be restored.

Since in the FH method the frequency of the reference frequency channel and synchronously with this, according to a predetermined ratio, the frequencies of the information frequency channels can vary according to almost any dependence, the claimed method and system are very flexible. By means of deliberately introduced frequency offsets, it is possible to avoid overlapping signals of several transmitting systems and reduce the likelihood of unwanted eavesdropping.

If, in addition to the frequencies of the reference frequency channel and the information frequency channel, other signal parameters are also used to create a bit pattern, it is possible to provide a more complex character of coding in a simple way and increase the data transfer rate accordingly.

If, according to clause 7, the amplitude modulation of the information signal is performed, then at certain nodes of the amplitude oscillations used for modulation, time moments can be set at which, for example, individual information frequency channels can instantly change their information parameters without causing interference in the information signal at points such instant changes. By this, the quality of the transmission of information can also be improved.

If, in accordance with clause 8, bit patterns are generated in a predetermined time cycle, then they can be decoded in the receiver in a simple way, and the accuracy of information transmission increases.

If, in accordance with paragraph 9, a change in the bit pattern is performed within one time cycle, in particular, the first part of the time cycle can be used to recognize the information frequency channels used to transmit information, and the rest to form the actual bit pattern. In addition, the first part provides an additional reference signal (except for the reference frequency channel), with which the parameters of the signal components transmitted in the second part of the clock can be determined with high accuracy. Thus, the reliability of information transfer is improved.

A preferred embodiment of the invention is set forth in paragraph 10, according to which it is possible to tune to various communication conditions and various user needs.

By using the features of clause 11, an increase in data rate can be achieved.

On the basis of the achieved high quality of reception (due to the application of the PG method), in combination with the above method of turning on / off individual information components (creating binary states "on" / "off") or instead of this method, the information can modulate certain parameters or combinations of parameters in the signal. Since in addition to frequencies, there are also amplitudes and phase angles of the signal components in the received signal that have a more definite correspondence with the initially generated signal, almost all parameters can be used for modulation. This can be done, for example, by step changes.

A significant advantage of this method is that the signal-specific relationships in a given frequency system can be used for encoding. By means of such relations, it is achieved that bit patterns or symbols can be identified on the basis of one or two received clock cycles of the signal, without additional reference to an external reference value.

So, for example, phase angles can be set in a given time step as the current relationship between the corresponding parameters of the information component and the reference channel. This coding method is defined as a method of relative phase angles (OFU method). Background does not matter in this method; external time loses its significance for signal estimation. Instead, the internal (relative) system time takes effect, which can be counted, for example, on the basis of instantaneous values of cyclic time, which, from an external point of view, depends in each case on the current frequency value. Relative phase angles can be determined in a simple way, when, for example, during processing, all signal components, i.e. reference and information channels are normalized first by one single duration of the period. This principle is explained as follows. In the theory of signal processing, there are many approaches that use methods for displaying and converting a signal, which can be used to ensure that as a result of the application of the FG method, and especially the FHG, a number of interference effects can be eliminated; in this case, the differential phase angles can be determined with greater accuracy, which can be used for finer sampling (i.e., for encoding a large number of phase states), as well as for a further increase in the information transfer rate.

The next option is that the information is not encoded directly by the phase angle of the corresponding component with respect to the reference channel (fundamental tone), as the so-called “vertical” support inherent in the signal, but as the difference between the phase angles in the current and previous clock cycles of the same component , as the so-called "horizontal" inherent signal support. This method is defined as a differential phase method (DF method). In this case, each first beat of the corresponding component serves as a horizontal support. In very difficult communication conditions, it may be useful to use a combination of the DF method and the method according to paragraph 9. On the other hand, in the case of simple communication conditions, using only a horizontal support is sufficient. In this case, the reference frequency channel may serve as an information frequency channel. It should further be noted that both in the OFU method and in the DF method, the fact of the absence of one of the signal components (or, for example, not exceeding a certain amplitude threshold) can be used to encode an additional digital state.

If, in accordance with clause 12, the number of information channels varies depending on the properties of the transmitting channel, then this ensures that, in particular with decreasing distance between the transmitter and the receiver, additional, usually higher frequency (or those lying between the used) can be used components, for example, having consonant frequencies; and conversely, at very large communication distances, mainly lower frequencies are used. This ensures the optimal use of the nature of the propagation of wave signals, which is especially important when using sound signals. Thus, for example, in submarine channels, maximum bit rates and / or long communication ranges can be achieved, respectively. Needless to say, the method allows for flexibility of use, so that the settings used for certain communication conditions can be adopted as the main standard, if the specified operating mode is implemented.

In addition to specific states or proportions of the signal parameters, this method can be used to encode information using instantaneous values or certain dynamic characteristics of the signal.

If, as in clause 13, individual information channels are specified in a wider frequency band and do not overlap, the possibility of information coding using constant (non-step) phase shifts of the corresponding information component is created. This mode is defined as a phase gradient method or a phase velocity (FS) method. The diversity of information components relative to the reference channel is determined, in a typical case, as the distance between the characteristics of the average frequencies of the respective channels. When transmitting data, the frequencies of individual information frequency channels (within each time step) can vary slightly and smoothly (usually less than 0.5% of the current nominal frequency value), while a continuous gradual (or accelerated) phase shift of the current component relative to the phase of the reference channel or pitch. The receiving device not only recognizes whether a component of a certain frequency was transmitted in a given clock cycle, but also determines (if the component exists) the value of the relative phase angle and / or characteristic parameter that describes the phase change function depending on the current cyclic time for the reference channel (or pitch). In addition to eigenvalues describing states or proportions, for coding, smooth changes in time of such states or proportions can also be used. At the same time, various possibilities for changing and combining are created, which make it possible to increase the speed of information transfer, improve the adaptability of the system to various communication conditions, as well as optimize transceiver devices and their cost.

To simplify the processing of the signal, after its reception, the support component is separated from the at least one information component, as described in paragraph 16.

In accordance with paragraph 17, by pairwise processing the corresponding information component of the signal, together with the reference channel or the reference component (or one of the most suitable reference components), the possibility of compensating for Doppler effects is achieved. As an additional result, this signal processing step helps to stabilize the frequency. In the case of using the FGG method, this step can lead to the formation of stable, i.e. not varying in time intermediate frequencies.

Further development of the method in accordance with paragraph 18 provides the translation of the signal components into stable intermediate frequencies, which provide a preferred option for subsequent processing. One of the advantages achieved according to clause 18 is, for example, that constant intermediate frequencies can be formed in the optimal frequency window for subsequent filtering in accordance with clause 20, which makes it possible to use narrow-band filters.

When applying the FSH or PPMP methods, there is, as an alternative to the approaches specified in clauses 16 and 18, the possibility of forming stable intermediate frequencies by mixing the signal received in the current clock cycle with the signal received in the previous clock cycle, without the need for preliminary separation of the signal components and the application of the local oscillation frequency. This signal processing option is proposed in paragraph 19, in particular for use in combination with a differential phase method.

Further development of the method in accordance with paragraph 20 is aimed at isolating (for example, filtering out) from the spectrum of stable intermediate frequencies having a whole set of multipath components (or reflections) that are only most suitable for each component of the signal components and, at the same time, minimize the effect of interference on them from other components. The latter assumes that at this stage of processing the signal components themselves can also be separated from each other, if this has not yet happened (or has not fully occurred) according to the method variant according to paragraph 16.

For this purpose, in the simplest case, special filters can be used. In this case, unnecessary components can be eliminated (for example, filtered out), i.e. components that are not currently intended for processing. As a result of such processing, each information component of the signal will be clearly expressed by one component, on the basis of which it becomes possible to best restore information parameters (used to encode data), for example, amplitudes and / or phase angles. This description is only about the presentation of the basic principle. It goes without saying that more sophisticated signal processing methods can also be applied here, which, for example, in addition to identifying the components of the received signal, also produce the corresponding parameters.

With further improvement of the method according to paragraph 21, the elimination of errors in signal processing is achieved.

As a result of further development of the method according to paragraph 22, it is achieved that, for the current communication conditions, such multipath components (signal components) can constantly be identified on the basis of which it is optimal, i.e. in the best way, the parameters of the transmitted signal can be determined. As a rule, these are the most powerful signal components that make it possible to achieve the best signal processing quality. In the process of tuning to the channel, for example, the best filter settings can be determined for the most accurate filtering of the desired signal components and the optimal suppression of interference effects of other multipath components and the possible influence of other side frequency bands. The latter may in some cases contribute to an increase in the radius of reception of the signal and / or also to an increase in the speed of information transfer. The better and more reliable the received signals can be processed, the more possibilities there are for more finely distinguishing digital states or applying different combinations of parameters to encode information.

Using the current update of filter settings, in accordance with paragraph 23, optimal reception results can be achieved even in rapidly changing communication conditions, and the advantage of using this method is that there is no need to interrupt the actual transmission of information for tuning to the channel.

According to paragraph 27, optimization of one of the methods for compensating Doppler shifts is achieved.

The method according to paragraph 28 is preferably provided for processing received signals influenced by strong Doppler shifts, where each signal component is represented, however, mainly through only one multipath component.

Further preferred embodiments of the invention are set forth in the remaining dependent clauses.

With reference to the drawings below, various embodiments of the claimed invention are described in more detail.

Figure 1 represents one of the claimed options using one reference frequency channel and three information frequency channels.

Fig. 2a represents an information signal composed of the components of Fig. 1 and subjected to amplitude modulation.

Fig.2b represents a number of clocked information signals.

Figure 3 schematically represents one of the options for encoding information.

Fig. 4 represents the same coding as in Fig. 3, but in a variant of the parallel FH method.

Figure 5 represents the result of the analysis at time t _{i of the} signal, consisting of the leading and delayed interference component, using the method proportional to the FG with respect to the three information frequency channels in harmonic relation to each other.

FIG. 6 represents the basic principle of improving signal analysis with respect to interfering components of FIG. 5, respectively, using one reference and four information frequency channels.

7 schematically represents one of the applications of the step frequency offset with an additional change in the information frequency channels within the cycle, and in each case the first half of the cycle forms an additional horizontal support for the DF method.

On figa schematically presents one of the coding options with two stages of frequency change.

Figb is an example of the principle of pentary coding of information in one of the information frequency channels.

Figa and 96 represent two different phase gradients formed by the FSH method.

Figure 10 presents various phase gradients that can be formed using the disproportionate hCG method (above) and the hCG method (bottom).

11 represents the construction principle of one of the claimed transmission systems.

12 is a possible embodiment of one of the claimed transmission systems using amplitude modulation.

13 represents a construction principle of one of the claimed receiving devices according to the first embodiment.

Fig represents a possible variant of one of the claimed receiving devices with additional detection of phase angles according to the second embodiment.

Fig. 15 represents the result of the analysis at time t _{i of the} signal, consisting of the leading and delayed interfering components, using the parallel FH method with respect to three information frequency channels that are in harmonic relation to each other.

Fig. 16 schematically presents several examples of favorable arrangement of frequency channels for specific applications.

17 schematically represents one of the main signal processing processes in accordance with the claimed method.

Fig. 18 is an example for time-varying frequency channels of a received signal formed by the pFPS method, consisting of one reference and three information frequency components under practically ideal communication conditions (minimum intersymbol interactions).

Fig.19 represents the received signal according to Fig.18 after converting the first information component of the signal to an intermediate frequency.

Fig. 20 is an example of the fact that due to the changing properties of the multipath channel, significant fluctuations in the levels of various multipath components of this received component can occur.

Fig. 21 is an example (already shown in Fig. 20) after passing through a narrow-band filter.

22 schematically represents one of the main procedures of the proposed method, in which the channel is tuned.

23 schematically represents the main steps of signal processing in various preferred embodiments of the inventive method.

24 represents the construction principle of the inventive system for processing signals according to the third embodiment.

Fig represents the principle of construction of the inventive system for tuning on the channel.

Figure 1 shows the option of generating an information signal IS, consisting, for example, of one reference component, which is a reference channel VK, which in this case is made as the main tone GT, and, for example, three information components, which are information frequency channels I1, I2 , I3. Presented in figure 1 information frequency channels are harmonic overtones NK1, NK2 and NK3 of the fundamental tone GT, the sum of which is an information signal. From figure 1 it can be concluded that the information in each information channel can be encoded by changing binary states of the channel type "on" or "off", which corresponds, for example, to digital codes "1" or "0" (compare with fig.2b) .

Figure 2a shows the amplitude modulation of the information signal IS (see Fig. 1) in order to ensure a smooth (continuous) transition at the beginning and end of the clock cycle for the case when the information signal belonging to a particular information frequency channel changes in time.

A change of this kind is shown in FIG. 2b, and the value of the information signal varies from measure to measure so that, for example, in sector 1 there is an information signal consisting of the sum of the fundamental tone, second and third harmonic overtones (GT + HK2 + HK3), which in the next measure (in sector II) is converted by turning off the second and third harmonic overtones into a signal consisting of only the fundamental tone, in order to form another information signal in the next measure (corresponding to another coded bit pattern) representing the sum of the fundamental tone and the first overtone (see Sector III). Thus, one bit per unit time (cycle) can be transmitted on each of the information channels. As a result, a bit pattern is created from this, in which information can be encoded in any way. In general, depending on the number of information frequency channels used, the encoding system used, for example, a letter or any other character may be encoded.

It should be noted that when using 2, 4, 8, 12, 16 or another number of information channels, direct compatibility with various existing methods of electronic data processing can be achieved.

Figure 3 presents, for example, the word "DolphinCom" when transmitting it in the generally accepted ASCII code using four information channels. The system of frequency channels constituting the information signal consists in this example of one reference frequency channel (in the form of a fundamental tone) and four harmonic information channels (I1, I2, I3 and I4, in the form of overtones of the fundamental tone), which vary in time by means of a proportional CHG. The coding in this example is done only by “turning on” / “turning off” the overtones. The vertical lines show measures that always have the same duration. In each measure, a special bit pattern is formed, defined as a symbol. Two characters together form one letter in ASCII code. The word "DolphinCom" is presented. In general, any other code can also be used to encode the transmitted information, which gives the user maximum freedom for his own programming, and also makes this method compatible with almost all other computing systems. As shown in figure 3, the reference frequency channel changes its frequency smoothly and continuously, and the frequencies of the four information frequency channels (I1, I2, I3 and I4) change in proportion to the frequency of the reference frequency channel.

Figure 4 again shows the transmitted word "DolphinCom" in the ASCII code when using also four information channels, and the frequency offset of the reference frequency channel is also smoothly and continuously as in figure 3, where the frequency offsets of the information frequency channels (initially ordered in harmonic ratio frequencies), however, in contrast to the variant of FIG. 3, a parallel frequency offset of the reference frequency channel is performed.

Figure 5 shows how a significantly more accurate and reliable signal analysis can be carried out when, for example, a smooth and continuous frequency shift is performed in the reference frequency channel in accordance with the FG method. In this case, for the example shown in Fig. 5 and taking into account Fig. 3, three information frequency channels were selected on which, in addition to the actual signal frequencies, in each case (on the receiving side) one "leading frequency" and one " delayed frequency "as interference signals, the corresponding time shift of these interference signals on each of the information frequency channels is identical. To clarify the basic principle, the indication of measures was not performed. The vertical cut line (at time t _i ) shows that on the receiving side the frequencies of all information components are different from each other. It is especially important to note the fact that, based on these frequency differences, the actual signal frequencies can now be separated from the interference frequencies, and thus intersymbol interactions can either be completely eliminated or significantly weakened. It is also important in this connection that the amplitudes and phase angles of the received and thus “cleaned” signal components have an explicit relationship with the reference frequency channel. When applying the FH method, special frequency filters can be used to separate the actual signal frequencies from the corresponding interference frequencies. In Fig. 5, it can be clearly seen that the distance from the signal frequencies proper to the corresponding interference frequencies increases with an increase in the gradient of the change (offset) of the frequency df / dt, i.e. with an increase in the corresponding rate of change of frequency. Since in the system of FIG. 5, the frequencies of all information frequency channels vary (are shifted) proportionally, the frequency change gradients related to the higher frequency information channels increase their values; this increases the quality of separation of the actual frequencies of the useful signal from the interference frequencies.

In Fig. 6, this effect (or functional mechanism) is shown schematically for a system consisting of one reference frequency channel and four information frequency channels, each having two adjacent interference frequencies (multipath components). The dashed line in FIG. 6 represents a characteristic of a conventional filter. From FIG. 6, it can be clearly recognized that even for a constant filter bandwidth, an improved separation quality of multipath components for higher frequency information components is achieved. Compared to existing methods, a substantially higher resolution is generally achieved. It should be noted in particular that according to this method, first of all, the most high-frequency information channels, whose frequencies are most suppressed in the propagation medium and, therefore, are received with the least energy, can be better extracted (filtered) from the noisy signal. It follows that, for example, at interference frequencies that are close to the actual frequency of the signal, a correspondingly higher frequency gradient should be selected, i.e. the frequency shift rate increases, while lower gradients may be sufficient for large distances between the useful and interference frequencies. For this kind of adaptation, either a pre-prepared set of frequency offset patterns or on-line gradients can be used. The latter is easily feasible if the connection is used in two directions, i.e. if the receiver can emit, and the transmitter can also serve as a receiver. Thus, the channel response can be analyzed and the results of the analysis exchanged between, for example, the receiving and transmitting devices, or the corresponding selection of the bit pattern can be carried out, followed by setting the optimal gradient of the change (displacement rate) of the frequency. In general, such a gradient can degenerate to zero if the communication conditions become suitably favorable (stationary), and the interference becomes negligibly small.

In this regard, it should be noted that in order to maximize the information transfer rate, the change in the duration of the cycle can be carried out proportionally with the change in the frequency of the reference channel, since the analysis of the individual components contained in the composite signal requires no more than a certain number of oscillation periods.

Figures 7, 8a and 8b show further possibilities that arise in particular when applying a step change in the frequency of the reference frequency channels. These possibilities are particularly interesting alternatives to the FG method, where the time offsets between the actual signal frequency and the interference frequencies are quite large values, for example, of the millisecond range. In this case, a good separation of the actual signal frequencies from the interfering frequencies can be ensured if all the channels (tactfully) simultaneously and stepwise change their frequency to a higher or lower value, while keeping this value constant inside the measure itself. Also in this case, it is useful to carry out a frequency offset, if possible, so that the internal proportions in the signal are equally determined at all steps. This can be most easily achieved using proportional or parallel step offsets. This alternative is generally defined as a step frequency method (midrange method). From Figs. 7, 8a and 8b it is clear how, for example, with the help of an additional phase state, the pentary coding of individual information channels is performed. In addition, to increase reliability, at the beginning of each clock cycle, a reference signal is transmitted on all information frequency channels, which is respectively encoded in the second clock section. As shown in the notation on the side of FIGS. 7 and 8a, for example, five states can determine 5 different discrete states, namely the absence of a signal (0) and four digital steps obtained, for example, using the DF method. In total, in each clock cycle for a signal consisting of one reference frequency channel and three information frequency channels, 5 ³ = 125 combinations are formed that can be used for encoding.

Fig. 8b represents, as an example, the principle of pentary coding of the information frequency channel. The points defined by the designations D1-D4 are signal components whose amplitude exceeded the amplitude threshold ao and four different phase angles (or angular sectors).

Fig. 8b represents, as an example, the principle of pentary coding of the information frequency channel. The points defined by the designations D1-D4 are signal components whose amplitude exceeded the amplitude threshold ao and four different phase angles (or angular sectors) which can be digitally separated from each other. State D5 represents the case when the amplitude of the signal component does not exceed the threshold Ao.

Another way of encoding information is presented, for example, by the PG method and is explained using FIGS. 9a, 9b and 10. FIGS. 9a and 9b represent the so-called method of proportional phase gradients or velocities (SFC). Figure 10 additionally represents the relationship between components for disproportionate phase gradients (nFS).

The basic principle of the FS method can be explained most simply based on the following initial configuration.

Let a system be given in which information frequency channels continuously make up a harmonic series, i.e. frequency offsets are performed using the proportional-frequency method. Let’s take any information frequency channel in which the corresponding frequency in this interval (measure) is now formed not as an exact overtone to the fundamental tone, but is slightly shifted up or down relative to the required overtone frequency (usually less than 0.5% of this nominal value), as shown figure 10 in the upper row. In the general case, a frequency shift is made, which, however, depends on whether this frequency is slightly higher or lower than its "nominal" value, as a result of its addition to the fundamental frequency GT, a continuous and smooth change in the phase angle of this frequency becomes obvious, as it presented in figa and 9b. The frequency phase of the information frequency channel is either ahead of the fundamental tone, or accordingly lags behind it. In the corresponding measure, this creates a phase gradient, the direction of which can be easily recognized and also easily determined by one of the known algorithms. When calculating the phase gradient in this case, a linear characteristic of the phase change with respect to the instantaneous value of the duration of the period of the fundamental tone is formed. A prerequisite for such a linear characteristic is that with a constant frequency shift in the entire frequency system, the internal proportions set in the current beat, but easily changeable, continue to be observed; it also means that the relative frequency shifts with respect to the fundamental tone are unchanged. The fulfillment of this condition is defined as a proportional method of FS (pFS). In the case of PPS, there is a uniform rotation of the phase of the information frequency channel left or right with respect to the phase of the reference frequency channel.

This effect can be very advantageously used for encoding information, since, for example, when analyzing the direction of rotation in a signal, the direction of the phase gradient is determined much easier than the actual value of the phase shift. Mathematically, this only means determining the sign of the first derivative of the relative phase shift of the frequency of the information frequency channel with respect to the fundamental frequency, or in other words, determining whether the relative speed of the phase shift is greater or less than zero (see figure 10 above).

This approach can be applied for each information channel individually in each cycle. If this approach is used in combination with the simple “on” / “off” channel method previously described, then up to four different digital states are created for each information channel in relation to GT: 1) there is no signal, 2) a signal with a positive phase gradient , 3) a signal with a negative phase gradient and 4) a signal without a gradient, although in practice, the fourth state can hardly be used, in which case the four-position coding will be replaced by actually three-position phase-gradient coding which, in some circumstances, may still not be as reliable as a two-position, since one of the digital states corresponds to a singular value of the phase velocity (Null). This issue, however, depends on the current conditions (quality) of communication, as for the transmitter, as a rule, there are no difficulties with the formation of gradients of any accuracy. Theoretically, the information transfer rate could double in the first case compared to a simple “on” / “off”, and in the second case, increase by one third.

Alternatively, the corresponding number of channels can be saved, and the frequency spectrum of the system can be kept narrower, which provides many corresponding advantages that can be used by choice. In this case, the emitters may not be very broadband, which, when using cascades of emitters, can lead to savings of individual or multiple elements. This can be useful in order to, for example, reduce the cost of the device. On the other hand, with the same device configuration, higher system flexibility and adaptability can be achieved. Thus, optionally, the information transfer rate can be increased by the fact that when you refuse to use lower frequency channels, the cycle time decreases, while when you refuse to use high frequencies, longer communication ranges can be achieved. There are a number of significant reasons for the need to increase the number of digital states and, thereby, increase the density of information in each information frequency channel.

The PPS method can be improved so that, in addition to the direction, various slopes of linear phase gradients can also be used, which can be used for coding by means of different frequency shifts of the information frequency channels, and in a particular case, depending on the level of sampling achieved , other coding and combining capabilities may also appear.

In conjunction with the PG method, positive and negative phase gradients can also be formed using such changes in the frequency of information frequency channels that are not completely proportional to the change in the frequency of the fundamental tone (see figure 10 below).

In contrast to the pPS method, this option is called the nPS method, and “n” means “disproportionate”. The nPS method can be implemented in such a way that the frequency of the corresponding information frequency channel changes in this cycle somewhat faster or slower than prescribed, for example, The basic version of the proportional method of PG. It is easy to recognize that both variants of the FS method are most preferably used on the basis of harmonic frequency series in conjunction with the proportional FH method.

The principle of an alternative method for forming phase gradients consists, in other words, in such a modification of the proportional FG method that an additional small, usually linear, frequency gradient can be formed individually for each component. Unlike the PPS method, when adding with the fundamental tone, instead of a linear phase shift, a slightly different characteristic is formed, namely, usually a quadratic curve describing the accelerated phase angle shift, the direction and shape of which depends, however, on the position of the initial and final values of the corresponding frequency changes in relation to its nominal value (see figure 10). In the case of the NFS method, up to six configurations can already be distinguished based on the signs of the first and second derivatives of the relative phase angle as a function of the pitch cycle time. Together for both options in the case of the FS method, a total of up to eight different combinations of characters are formed.

It is also interesting in the nPS method that, in addition to the signs, certain relative phase angles can also be used, for example, the phases of the initial and final values or the cut-off points on the curve of the proper phase values in the sense of the OFF method.

Figure 11 presents the reference version of the encoding of information (modulation) by one of the transmitting devices. The scheme reflecting the basic principle has a device that feeds encoded information to encoder 3. The encoder converts the received information into the necessary code, in accordance with the number of reference and information frequency channels, and delivers the encoded information to the device (in the form of generators 5) generating at least one reference and at least one information frequency signal, which are controlled using a special control device 7. The wave components of the corresponding amplitude, frequency and the phases are fed to the summing device 9, which is controlled by the control device 7.

According to this embodiment, the signal generated by the summing device is supplied, if necessary, to a power amplifier 11, which transmits an information signal to an input of a radiator (or a cascade of radiators) suitable for use in a particular medium.

According to this embodiment, one oscillator per frequency channel is provided.

On Fig shows an embodiment of one of the preferred transmitting devices, performing, if necessary, amplitude modulation of the information signal. For this, before applying an information signal, consisting of separate components (reference and information frequency components), to the summing device, the information signal is first transmitted channel by channel to the respective modulators under control from the control device.

On Fig presents one of the embodiments of the receiving device of the inventive system. A converter adapted to receive signals in an appropriate transmission medium receives the incoming information signal shown in FIG. 13 as an acoustic signal and feeds it to an amplifier 23. After the amplifier, a filter 25 is provided for separating and analyzing individual frequency channels and, in particular , to pass the reference frequency channel. The signal coming from a filter, preferably a low-pass filter, is transmitted to a frequency detector of the reference channel 27, which estimates the value of the level and frequency of the reference frequency channel. These data are transmitted to the control module 29 of the receiving device. In parallel with this, before entering the low-pass filter 25, an information signal is also supplied to the controllable filters 31 provided one for each of the components. The output signals of the filter are analyzed in a controlled threshold device 33 and transmitted to a decoder 35, which converts the code into a source symbol.

On Fig presents the following embodiment having additional phase detectors, for example, for the method of PPS, nFS, differential phase or phase gradient method. For phase analysis, a phase detector 32 is provided, which is connected between the controlled filter 31 and the controlled threshold device 33, preferably in accordance with the number of effective frequency information channels, and for determining the mentioned phase values, the reference channel processed by the pitch detector is used as a support.

15-24, various embodiments of signal processing are shown in detail.

The inventive method includes two significant support functions that can be combined in any way or also be used independently. These support functions are defined as 1) full compensation for Doppler shifts, abbreviated as CDS, and 2) "channel cleaning", abbreviated as HK. The relevant basic principles are first explained separately; then, various possibilities for combining and improving the method and system are explained.

1. The solution to the problem of Doppler shifts

To explain the basic principles of CDS, an example is first chosen in which the frequencies of the reference and information channels make up a harmonic series, i.e. these frequencies are integer relations to each other. If the transmitter changes its frequency, then this happens according to the PPMP method. In the receiver, all frequency components of the signal are separated immediately at the first stage, for example, using a cascade of bandpass filters. For simplicity, we first accept the ideal transmission conditions when each component consists of only one single multipath component and all signal components are emitted and received at approximately the same power level. Such communication conditions may arise, for example, during the propagation of electromagnetic signals in the air. This example first explains how and what is achieved by fully compensating for Doppler shifts.

The problem of Doppler shifts is that due to the movements of the transmitter and receiver relative to each other, frequency shifts occur that cannot be accurately predicted in advance, because, for example, the speed of relative movements is not exactly known. Moreover, the position of the phases cannot be determined with sufficient accuracy, which leads to significant limitations for any form of communication using phase encoding. This problem can be partially solved by differential phase coding, in which the phase angle itself is not taken into account, but only its change from measure to measure. One-hundred-percent compensation for Doppler shifts can, however, be achieved if individual information components are processed appropriately together with the reference component. One useful solution contains "pairwise alignment by Doppler shifts", abbreviated pVDS, which can be implemented in different ways. One of the simplest features of this is explained in more detail in the following example:

Let any one, for example, the first information component (the first overtone of the fundamental tone), be presented selectively and representatively for all others to be processed in one of the time steps. The frequency of this component f _ik and, consequently, the speed, the change in the phase angle ω _ik represent values twice as large as the corresponding values of the fundamental tone f _b and ω _b . Let the signal be received in discrete form, then the send _b and send _ik signal components emitted by the transmitter can be represented as follows:

where N is the total number of digitized points (samples) in a given clock cycle, n is the number of the current sample, t _s is the time interval in which digitization took place, nt _s is the discrete time, E is the energy, θ _ik is the initial phase, θ _inf is the phase angle encoding a certain digital state on the information channel, and the factor k describes the steepness of the change in the frequency of the component when applying the proportional PMP method.

In the general case, k may be any suitable function of time, may take positive or negative values, or be equal to zero. The latter means that the use of constant carrier frequencies is considered here as one of the possible special cases.

Since the value of the initial phase of the reference frequency does not change on the transmitting side, it does not play any further role, therefore, the corresponding value in equation (1) is set equal to zero.

Due to Doppler shifts, the received signal components empf _b and empf _ik differ from those emitted by only one additional term:

where D is the Doppler coefficient describing the ratio of the relative velocity of the receiver-transmitter (positive sign for mutual approach and negative for removal) to the propagation speed of the signal in the transmitting medium.

Based on the various arguments of these equations, it becomes clear that the Doppler effect has such an effect on these components that the corresponding members of these arguments differ from each other exactly two times, which, in fact, is determined by the proportionality relation between the frequencies of these components. In this example, the proportionality coefficient is two.

Since the proportionality coefficient is a known quantity, the exact values of the Doppler shifts no longer play any practical role. Namely, if the reference component is transformed so that it receives the same frequency response as the analyzed information component, then the values of the Doppler shifts of these components will become equal. In the presented example, by squaring the support component, such a support Rf can be generated that has a Doppler shift such as for the information component. After squaring the support component, you will get:

получаем нормированный опорный сигнал Rf', значение фазы которой отличается от фазы информационной компоненты, представленной в уравнении (5), следующим образом:After the unused sideband of the spectrum is filtered, as well as after scaling the result by the coefficient

we get the normalized reference signal Rf ', the phase value of which differs from the phase of the information component presented in equation (5), as follows:

This reference signal can be used to some extent as an internal clock signal of a composite signal, with the help of which the phase angles of information components can be determined.

Similarly, the necessary reference signals can also be formed from the reference component for all other information components contained in the received signal. In this case, the supporting component should be raised to the appropriate degree (multiplied by itself many times) and, if necessary, filtered. In the general case, information components can also be converted in a similar way, which can be useful if the frequencies of information components are not chosen as in the previous case, but otherwise, for example, when they lie in the lower frequency range than the reference component, or if their frequencies are not integer to the frequency of the reference component. In the latter case, the above approach can be applied individually for each side of the pair, formed for the current processing (reference and one of the information components) repeatedly, and just as much as is necessary to fully cover both parts of the pair. Since after each multiplication (exponentiation) the number of frequency components of the spectrum increases rapidly, it is necessary to initially strive for such an arrangement of frequency channels so that as few steps as possible are required for pairwise processing of the received signal to equalize Doppler shifts.

In the general case, when using applications with phase-coded signals, when choosing approaches for pairing the Doppler shifts, it is also necessary to ensure that there is no loss of information when manipulating information components, for example, due to the occurrence of uncertainties (ambiguities) in the estimates of the phase position.

In the following, on the basis of the mentioned example, one of the simple possibilities is described, how according to the aforementioned approach, the position of the phase angle of the corresponding information component can be determined. For this, it must be described, for example, the possibility of decomposing the corresponding component on the basis of the quadrature functions of the reference signal Rf '[n].

Since the support Rf '[n] in this example is already presented in a cosine form, we write:

RfC [n] = Rf ’[n]

The corresponding sine component can be obtained, for example, by taking the first derivative of RfC [n] and the corresponding normalization of the amplitude.

The projection of the information component on the cosine component has the form:

где N1 описывает начало, а N2 - конец соответствующего такта.

where N1 describes the beginning and N2 describes the end of the corresponding measure.

Since in the second part of the sum, the values of the function oscillate at zero, the positive and negative quantities are mutually annihilated, i.e. this part tends to zero and therefore it can be neglected:

The projection of the received information component on the sine component is respectively recorded:

If we consider the quantities CQ and SQ as, for example, the coordinates of a point in a radial coordinate system, then the connecting line between this point, the origin and the abscissa encloses the desired angle Θ. It can be simply determined using the appropriate algorithm. An obvious way to determine the angle is shown as:

The phase of the received component is presented here as the difference between the initial phase of the transmitted wave and some coding phase, i.e. within each measure, it is time invariant. For the sake of completeness, a variant is presented where the difference between the phase in the current and previous measure can be used for encoding. Using indices i and i + 1, we obtain a variant of differential phase encoding:

Similarly, in each cycle, the phase angles of the remaining information components are determined. This gives the user the possibility of correspondingly more accurate (multi-stage) discretization of phase angles and increase the speed of information transfer. The above-described method for determining phase angles is defined here as a C-S projection.

The CDS method is the basis for the embodiment of the method according to paragraph 28. On Fig presents the most important elements of the processing performed by the method of CDS. It is shown that various (of the described) elements can be similarly applied to other forms of the method.

2. Channel cleaning (ChK)

HK includes the identification (in each case) of the most suitable multipath component and its separation from the remaining components while minimizing intersymbol interactions. It may include already partial compensation for Doppler shifts, which may already be sufficient for a number of different applications.

Let us consider as an example the case when Doppler shifts do not play a significant role; the received signal contains only interference represented by multipath components. Communication conditions of this kind are often found in acoustic communication between stationary or slowly moving underwater objects. Each of the signal components is then represented by a whole spectrum of multipath components (see the schematic representation in FIGS. 5 and 17a). The inventive method should, first of all, provide the possibility of such signal processing in which intersymbol interactions are significantly reduced.

Despite the fact that as a result of applying the PMF method, various multipath components arrive at a receiver with different values of instantaneous frequencies, it is hardly possible to isolate the most favorable components from the spectrum of the corresponding component immediately at the first stage of processing, since they are usually located very close to each other to a friend, and their frequencies are time-varying quantities (see Fig. 18). Even tracking filters can hardly be used for this purpose. Above, we proceeded, however, from the fact that the corresponding spectra of the reference and information components can be separated from each other (Fig.17b and Fig.17c).

After multiplying the reference and corresponding information components (Fig.17d), two spectra of intermediate frequencies are formed, which lie in different frequency ranges and have different frequency gradients (Fig.17e). The lower frequency range can be omitted with a low-pass filter for further processing. In this part of the spectrum, a decrease in the effect of Doppler shifts is also observed, while in the other part of the spectrum, these shifts increase their influence. If this second (side) frequency band does not cause interference, at this stage it can still remain in the spectrum (not be filtered out), while one of the filter steps can be saved.

In the next processing step, at least one remaining part of the spectrum is multiplied by the auxiliary frequency generated in the receiving device (Fig.17f), the characteristic of which is selected so that as a result of the multiplication, a part of the second intermediate frequencies no longer changes in time (Fig.17g) .

The characteristic of the corresponding frequency (H1; H2; ... HN) is either established on the basis of operational coordination of the signal structures used by the receiver and transmitter, or is determined from a preliminary sounding of the properties of the transmitting channel (see below “channel tuning”),

On Fig presented that this can also be achieved when at first only the supporting component is separated from the group of information components. By appropriately selecting the local oscillator frequency, the frequency provided for processing the information component (the first in the current example) can be stabilized.

The advantage of this approach is that by means of a suitable local oscillator frequency, the necessary parts of the unchanged (stable) in time frequencies can be placed in certain frequency windows and, at the same time, filtered using conventional filters, for example low-pass filters. (Fig.17h).

Fig. 20 shows in an example close to practical use that, due to the existence in the spectrum of stable intermediate frequencies (of a number of) multipath components of the signal, no conclusions can be made about the value of the phase angle, because different multipath components are rearranged at different points in time to different power levels.

For this reason, a second stage (if possible with the best quality) of filtering is provided, the band of which is set (for each component) on the basis of data on the channel properties obtained during the “channel tuning” procedure (description below) at the most powerful level multipath component. The dashed line (Fig.17h) shows that the slope of the leading and trailing edges of the filter can be set with the necessary slope. As a result, the effect of the remaining multipath components can be significantly reduced (Fig. 17i).

Fig.21 represents a practical example of how, as a result of such narrow-band filtering, the influence of fluctuating multipath components (see Fig.20) can be significantly weakened (suppression of these components) and an unambiguous assessment of the information parameter can be made. In general, the process described in this context is defined as “channel cleaning” with partial compensation for Doppler shifts.

Parameter Definition

Signal components prepared in this way and cleared of the channel interference can now be subjected to a detailed analysis of their parameters. In this case, both the amplitudes and phases of the signal components carrying information can be determined with high accuracy. Different amplitudes can vary, in the simplest case, using threshold devices. If it is necessary to carry out a C-S projection to determine the phase angles, the reference vibrations (sine and cosine components) required for the decomposition of the components can be artificially generated by the receiving device.

The latter does not present technical difficulties, since the settings of the last (narrow-band) filtering stage are known in the system; therefore, the frequencies of the signal components that carry information are known. Depending on the form of coding used by the user, the most suitable processing forms from among the known algorithms can be selected and applied.

In the method described above, the “channel cleaning” procedure of the LC is preferably used in conjunction with the pFMP method (see FIG. 5). This method can also be implemented without problems and in conjunction with the pasf method (see Fig. 15). In the case of using the paPMP method, the product of the reference and information components leads directly to the formation of stable intermediate frequencies. In this case, further multiplication by the auxiliary component becomes redundant. If this intermediate step is still necessary, for example, to move a certain frequency spectrum into a given frequency window, this can be done by multiplying by the corresponding auxiliary component of the constant frequency. This is, however, within the scope of the above description.

Channel cleaning is compatible with all forms of the PMD method, in which the gradient of the frequency change is a value greater than zero. In order to distinguish the described option from subsequent modifications, it will be called as the ChK-1 method. The preferred embodiment of the inventive method described here forms the basis of paragraph 18. The most important elements of this method are presented in FIG.

The above-described variant of channel cleaning can be transformed, for example, so that initially the multiplication of the information component by the reference component is not performed. In this case, stable intermediate frequencies are formed by multiplying the corresponding component by the corresponding auxiliary frequency. This approach gives the advantage that the spectra of stable intermediate frequencies do not have such complex spectra of multipath components as the received components. After separating (filtering) the best multipath component of each component (i.e., cleaning the channel), there is still the possibility of processing the information component using a support so as to achieve at least partial compensation for Doppler shifts, or as, by analogy with the example described for CDS , to evaluate the phase angle of the information component using a CS projection, with the formation of the sine and cosine components from the "cleaned" reference component. The choice of the frequency response of the corresponding support occurs, if necessary, during the multiplication with the corresponding auxiliary accompanying frequency, or after passing the last filtering stage by multiplying with the corresponding auxiliary constant frequency. In the latter case, only a single passage of the support component through the filter is necessary.

Using the described modifications, other preferred embodiments of the invention may be determined. A schematically simplified processing procedure is presented in FIG. 23 and is defined as a ChK-2 method.

If the Doppler effects do not play any role, the support component can be omitted or used as an additional information component. In this case, the ChK-2 method is used. The determination of the signal parameters should, however, occur as described for the ChK-1 method.

For completeness, we refer once again to the alternative solution described above, but not shown in the drawings, by which the stage of obtaining stable intermediate frequencies can be achieved in the framework of, for example, the PMF method without preliminary separation of components, simply by multiplying the signals received directly in adjacent clock cycles. This step also includes partial compensation for Doppler shifts. A feature in this case is that, depending on the steepness of the frequency change, the spectra of stable intermediate frequencies of the corresponding channels lie in different, although more or less close to each other adjacent frequency windows. However, as a result of such a transformation, a very complex signal structure is obtained. Especially if a large number of information channels are used, it is necessary to carefully monitor the possible imposition of multiplication products of the processed components. For separation of multipath components, for example, a cascade of narrow-band filters can be used.

As a result, it is necessary to once again point out that in the present description of the system only the principle of the proposed approach is explained. In practice, it is possible to use more complex methods of signal processing and analysis, which include the use of the above processing steps of a similar or other form while maintaining the principle.

Integrated Solutions

After we have separately described the methodological foundations for the full compensation of Doppler shifts and various options for cleaning the channel (including also partial compensation for Doppler shifts), we consider the practical case in which interference with the reception of a signal causes both a multitude of multipath components and Doppler shifts. Such a combination of interfering factors often complicates the connection between moving objects under water.

For this example, there is the possibility of obtaining a solution, for example, by combining the methods of CDS and ChK-2.

After separation of the reference and information components, the pairing of Doppler shifts is first performed, as described for the CDS method, in which at least one of the components of the selected pair is transformed (or, if necessary, both components are transformed) so that both components of the pair are exactly equal characteristics of the frequency change and therefore are equally "loaded" with Doppler shifts. If necessary, the frequency bands (components) of the signal that are not required for processing can be filtered out and the remaining signal components can be re-normalized.

After that, both components separately are usually multiplied with the same auxiliary frequency (which has the same gradient of frequency change as the corresponding component, but is parallel shifted by a certain amount), and stable intermediate frequencies are formed, which are then filtered to clean the channel by subsequent stages of filtration. For this, narrow-band filters are installed individually for each component. Ideally, the appropriate filter settings can also be taken into account when precisely matching the aforementioned auxiliary frequencies.

As a result, purification from intersymbol interactions of both the support component and the information component is achieved. After such cleaning (including, if necessary, analysis of threshold values), the analysis of the parameters can occur, for example, according to the described approach for CDS or ChK-2, and by fully processing the corresponding information component with the corresponding reference signal, full compensation for Doppler shifts is achieved.

The above form of execution of the proposed method creates the basis for a variant of the method according to paragraph 8. In Fig.23 it is designated as Kompl.1.

The next possible solution is to appropriately combine the CDS and ChK-1 methods (see a simplified representation of the processing circuit in Fig. 23). Also in this case, after separation of the support component from the information component, pairwise alignment of Doppler shifts occurs. After that, one of the components is shifted in parallel by a certain frequency value by multiplying it with the corresponding auxiliary constant frequency generated by the receiving device. Then both components are multiplied, and the processing step shown in Fig. 17d is achieved, i.e. stage of formation of stable intermediate frequencies. After that, filtering (at both stages) and analysis of the parameters according to the ChK-1 method are performed.

This second (complex) solution assumes that by projecting the information component onto a reference signal having an identical characteristic of Doppler shifts, the influence of frequency shifts caused by the movement of objects is completely eliminated. In this case, this reference signal is "consumed". However, it is no longer needed. A significant advantage of this approach is that to place the required part of the spectrum of intermediate frequencies exactly within a given frequency window, there is a need only to generate an auxiliary constant frequency. In the favorable case, the same auxiliary frequency can be used for all pairs of processed components. In principle, it is possible to use this auxiliary frequency at the final stage of channel cleaning as a reference signal for analyzing phase values. Since in practice they will try to configure narrow-band filters for each signal component individually (in this case, a combination of the reference and corresponding information components is already presented), then the current filter settings become known to the system in each case of settings, and therefore the receiving device can artificially generate reference signals (including quadrature sinusoidal and cosine components), if necessary, for performing anal for phase angle (compare with CHK-1).

Channel setup and channel setup

By repeatedly adjusting to the channel, based on the corresponding test signals, it is first necessary to establish that the signal structure is optimally selected for the respective communication conditions and / or it is ensured that at least the receiving device can separate the signal components to the required degree. If this prerequisite is fulfilled, then tuning to the channel can be carried out for all variants of the method, which correspond to point 1 and also all further options, including the procedure for cleaning the channel. To do this, you can first provide for the transmission of longer non-encoded signals, which otherwise, however, have the necessary characteristics used in the future to transmit information. In this case, the user can choose whether to apply all frequency channels at the same time, or tune to the channel based on test signals sent in turn and representing in each case either only the reference component or only one of the information components. Any of these approaches should, of course, be consistent with the selected version of the signal processing method. The received test signals go through all the processing steps provided for in an appropriate way to form stable intermediate frequencies. At this level, an analysis of the energy distribution in a given frequency spectrum is carried out individually for each of the signal components (or from mixed components formed by multiplying the reference and corresponding information components) intended for further processing. For such an analysis, for example, fast Fourier transform can be applied. Based on the processing results, the best (favorable) multipath component (usually having the highest power level) is selected, for which high-quality tuning of narrow-band filters is then performed, the results of which are stored in memory. After completing the appropriate settings for all signal components, the actual transfer of data (information) may begin. These settings are saved until the next channel setup.

Especially with acoustic data transmission in underwater channels, communication conditions are often unstable, time-varying. In such cases, it is recommended that the channel tuning procedure be repeated at intervals corresponding to the communication conditions, which means that the settings of narrow-band filters are regularly updated.

The use of longer non-encoded signals for tuning to the channel offers good static reliability, although this means that the data transmission itself should be interrupted from time to time. Such interruptions can, however, be avoided. A useful alternative is the method stated in paragraph 23. At the same time, on the basis of the received signals, in parallel with the process of the signal processing itself (or as an integral part of it), the channel is tuned in and the mentioned filter settings are updated. In this case, it is advisable to use for processing the results of receiving signals from multiple clock cycles. One of these alternative solutions sets, accordingly, more complex requirements for the signal processing system.

Determination of relative speed between receiver and transmitter

It is presented below that, based on the received signals, a determination of the current change in the distance between the receiver and the transmitter can be performed. Signal processing according to the above method involves the restoration in the best way of the parameters of the (sent) signal (for example, phase angles) for each of the signal components. The signal components are expediently processed together so that Doppler offsets are eliminated. The latter are considered as interference values. Doppler shifts, however, contain information in the form of the Doppler coefficient D = ν / c (compare with equations 4 and 5), which although it has nothing to do with the actual transmission of data, can, however, provide information about the instantaneous value of the relative velocity v between receiver and transmitter. The Doppler coefficient can be determined using the appropriate signal processing method. Since the propagation speed of signal c is a known quantity (for example, it can be measured during channel sensing), the value of v can be estimated or relatively accurately determined.

The following possible solution may be presented as an example.

For this, any one component of the received signal can be filtered out (it is advisable to select an uncoded reference component). If necessary, then again, in accordance with the ChK-2 method, it is possible to reduce the spectrum of the received signal to one multipath component. Since the signal structures are known to both the receiving and transmitting devices, and it is also possible to determine the phase angle in the process of signal analysis (using one of the familiar methods), the system can generate an amplitude-normalized reference signal, which with respect to its phase angle and frequency response - with the exception of the not yet known Doppler component - is identical to the corresponding component of the received signal. After projecting this component of the received signal onto the quadrature sine and cosine components of the reference signal and after filtering (low-pass filtering), the pure Doppler component can be obtained in the form of sinusoidal and cosine oscillations of equal amplitude. The arc tangent function defines the argument Dωnt _s . Since ωnt _s , is a known quantity, dividing these quantities gives D, and D multiplied by c gives finally ν. (See FIG. 23). For many applications, it may be useful to obtain this useful information without additional measurement costs.

It should be noted that knowledge of the Doppler component contributes to the further improvement of the signal processing itself. So, for example, auxiliary frequencies generated within the framework of the HF can be more precisely coordinated with the structure of the corresponding component of the received signal and, thus, undesirable (in the analysis of the signal), Doppler shifts can be weakened more efficiently or in a simpler way. By integrating such procedures, if necessary iteratively applying them, in addition to improving the processing results, at least intermediate optimization can also be achieved; Thus, with an increase in stability with respect to Doppler displacements, the fields of application, for example, of the ChK-1 and ChK-2 methods, can expand. In this case, it is possible to avoid, especially when multiplying the components of the received signal with each other, the rapid growth of computational costs, in some cases, you can refuse the intermediate stage of filtering and generally process faster. All simplifications of the method are useful for achieving a real-time processing mode. Even if the estimation of Doppler shifts first causes additional computational costs, in general, savings can still be achieved, since then the central signal processing procedures can be performed using less software and / or hardware.

Further, the aforementioned simplifications or improvements of the proposed method can be achieved in a simple way also in the case when information about the current Doppler shift is provided in one of the forms, for example, from the side of an external measuring system.

On Fig presents one of the possible functional diagrams of a signal processing device containing a filtering device consisting of two parallel controlled filters BPF1 and BPF2, which separates the received signal into a reference and information component.

Both parts of the signal are first fed to the frequency conversion device [consisting of Wandler1 (Converter 1) and Wandler2 (Converter 2)], in which the Doppler offsets are paired up before these two parts of the signal are multiplied by multiplier 1, for subsequent conversion with using auxiliary (heterodyne) frequency (supplied by the generator of the processing system) and multiplier 2 to the constant (no longer changeable in time) intermediate frequencies.

The filter LPF1 and, if necessary, also the second filter LPF2, which in each case are included after the respective multipliers, to remove from the spectrum unnecessary side frequency components for processing, serves as a means of suppressing interference components (in this form of the method implementation). Then, a multipath component suitable for processing is extracted from the spectrum using the BPF3 filter included in the processing chain; the signal is then fed to the parameter analysis device (in this form of execution, to the module for parameter analysis). If necessary, the parameter analysis module can be supplemented with a reference signal generator associated with BPF3.

At the output of the processing device, for each information component, signal parameters are used that are used to modulate the transmitted symbols (for encoding information).

25 is a functional diagram for tuning to a channel, applicable in a similar context. In contrast to Fig. 24, in this embodiment, the signal components after the LPF2 filter are supplied to a tuning device in which the FFT module (or the module for analyzing frequency spectra) and device 3 represent a processing device. The result is fed to a control device that makes the appropriate optimal BPF3 filter setting.

The following describes in detail further possibilities for using the proposed method and system.

When the distance between the receiver and the transmitter is reduced, additional frequency channels lying between the originally used frequency channels can optionally be used, higher frequency channels can also be used, or the entire frequency spectrum can be shifted entirely towards higher frequencies. At the same time, such an additional beneficial effect appears that with a decrease in the communication distance, the influence of interference factors also decreases. To implement this, the receiver and transmitter must be able to use a fairly wide range of frequencies, as well as the ability to accordingly reconfigure encoding devices. On the transmitter side, the recognition of additional frequencies entered can be done either automatically, or the transition to a new operating mode can be done by sending a message from the transmitter (for example, with the latest information packet). The individual tone channels must, however, lie so far apart that they can be separated by the receiver under appropriate communication conditions. On the other hand, increasing the communication distance may necessitate shifting the spectrum of the frequencies used to a lower frequency range or the need to increase the frequency distances between the channels (especially when amplifying the effects of interference) by proportionally stretching the spectrum of the signal or by lowering the intermediate frequency components.

The choice of frequency ranges, provided mainly from the main options so that the tones or frequencies of the channels have a consonant or integer ratio (overtone) to the fundamental tone having the lowest frequency, is aimed at achieving an energetically favorable ratio.

When applying harmonic frequency sequences, there is, among other things, the possibility of using nonlinear effects in the propagation of sound vibrations and at the same time achieving large transmission distances. Sound waves are longitudinal waves in which high-density regions are replaced by low-density regions. Since the speed of sound also depends on the density of the medium, the densified areas propagate faster. The fronts of initially sinusoidal oscillations become gradually asymmetric, i.e. sinusoidal oscillation is converted to a greater extent into a sawtooth oscillation. Physically, this means the transfer of energy to overtones. In water, this effect becomes sensitive at a distance of several kilometers. If overtones (one or more) are also sent along with the fundamental tone, then the latter, due to harmonic relations and the aforementioned nonlinear effects, receive additional energy from lower frequency tones. Subsequently, they disappear in the noise not so soon (they stay longer above the background noise level) and therefore spread over long distances. Since the distance of data transmission by the system is determined primarily by the highest frequency of the system, the communication range as a whole increases. It is advisable that the main tone (if possible, also other tones) emit with high power.

Due to the multivariance of the system, other properties of the transmission channel may also be used. Due to the layered heterogeneity of the aquatic environment, transmission channels are often created that have their own properties for the propagation of vibrations. Depending on the corresponding eigenvalues, slightly different modes can be excited, which, as a rule, have relatively low frequencies, but propagate over very large distances. In the information transfer system, in principle, there is the possibility of matching frequency ranges to such modes. The receiving and transmitting devices must be able to negotiate for such a mode of transmission and reception.

If the relative speed of the receiver with respect to the transmitter is so small that Doppler shifts can be neglected, corresponding uniform frequency changes can be made for the entire system as an alternative to the above-described proportional FG method. In this case, the so-called frequency offset curve can be set (depending on certain communication conditions) and summed with all frequency channels (see Fig. 4). This method is called the paGG method. A feature of this method is that due to the parallel displacement of the frequency channels, a constant frequency gradient is formed, i.e. the frequency displacement rate, and in the best case, the optimal resolution of the frequencies of the multipath (interfering) components for the corresponding channels along the entire frequency spectrum can be achieved. The method thus modified has, among other things, the advantage that the spectrum of frequencies occupied by the signal components does not expand with increasing reference frequency. Due to the more compact arrangement of the components in the spectrum, the upper tones are hardly at risk of being in the frequency range with too small a radius of propagation of oscillations, which could result in loss of information. The upper frequency ranges (more favorable in terms of data rate) can be more effectively used. The parallel method of PG can be easier to implement, because emitters and receiving sensors can often work only in a very limited frequency band, and the use of appropriate cascades is not always possible.

It goes without saying that also with the parallel FS method, the receiver should be informed how the individual frequency channels must be matched in relation to the pitch. In principle, switching between a parallel and proportional FH method should not be problematic, since with respect to determining the frequency characteristics with respect to the reference channel, it is only necessary to switch from the multiplication function to the addition function.

If phase jumps create problems at the junction points of the clocks, it may be useful to implement beat-based amplitude modulation. One of the further ways to minimize the effects of interference is to use a coding method that eliminates the existence of tones in two adjacent clock cycles of the same information channel. The same effect can be achieved by multiplexing with alternate use of even and odd information channels. In addition to other parameters, a speed characteristic (based on frequency values) can also be determined by means of special sensing or in the process of two-way communication. It can (and even should) be taken into account when generating signals.

If necessary, if current communication conditions allow, it is further intended to use the reference frequency channel additionally as an information frequency channel.

Claims (37)

1. A method of transmitting information in which at least one reference component (VK) and at least one information component (I1; I2; ...; IN) are included in the transmitting signal, the reference component (VK) and information components (I1; I2; ...; IN) form the corresponding discrete states to form a bit pattern, the bit pattern being set during a given time interval by manipulating the frequency, amplitude and / or phase of the information components, and also characterized in that at least one of k a signal component experiences a continuous change in its frequency during the transmission of a portion of information, and such a continuous change in frequency does not depend on the transmitted portion of information. 2. The method according to claim 1, characterized in that at least one reference component and at least one information component experience a continuous change in frequency over time, and between the frequencies of the reference and information components, a certain ratio is set in accordance with a predetermined function of time. 3. The method according to claim 2, characterized in that the frequency distances between the components are constant or undergo proportional changes. 4. The method according to any one of claims 1 to 3, characterized in that the frequency of at least one component in the process of transmitting a portion of information is continuously increasing. 5. The method according to any one of claims 1 to 4, characterized in that the frequency of at least one component in the process of transmitting a portion of information is continuously reduced. 6. The method according to any one of claims 1 to 5, characterized in that in order to minimize intersymbol interference, the gradients of the frequency changes are set depending on the position of the interference frequencies relative to the corresponding frequency of the signal component, and / or in order to avoid mutual interference of several data transmission systems, the corresponding gradients carrier frequency offsets. 7. The method according to any one of claims 1 to 6, characterized in that the initial frequencies of the components change their values from one data transmission interval to another. 8. The method according to any one of claims 1 to 7, characterized in that the frequency ranges, i.e. frequency bands of two or more components intersect. 9. The method according to any one of claims 1 to 8, characterized in that at least one component, preferably a reference frequency, lies in a separate frequency band. 10. The method according to any one of claims 1 to 9, characterized in that the bit pattern changes during one clock cycle. 11. The method according to any one of claims 1 to 10, characterized in that the number of information components (I1; 12; ...; IN) varies depending on the data transmission channel. 12. The method according to any one of claims 1 to 11, characterized in that the reference component (VK) is used as an additional information component (IN + 1). 13. The method according to any one of claims 1 to 12, characterized in that the reference component (VK) and at least one information component (I1; I2; ...; IN) are formed in the form of a sound wave or electromagnetic wave. 14. The method according to any one of claims 1 to 13, characterized in that for processing the information signal after reception, the support component (VK) is separated from at least one information component (I1; I2; ...; IN). 15. The method according to any one of claims 1 to 14, characterized in that the pairwise processing of the support component (VK) and at least one information component (I1; I2; ...; IN) is performed. 16. The method according to any one of paragraphs.14 or 15, characterized in that the information component and the reference component or pairwise processed reference components and information components are converted to constant intermediate frequencies (Z'1; Z'2; ... Z'N + X) preferably by multiplying with auxiliary frequencies (H1; H2; ... HN + X). 17. The method according to any one of claims 1 to 16, characterized in that preferably in the variant with a proportional change in the frequency of the signal components, constant intermediate frequencies are formed by pairwise processing, preferably by multiplying the signal received in the current clock with the signal received in the previous clock. 18. The method according to any one of paragraphs.15 or 17, characterized in that for separating the frequency components of the spectrum of the received signal, the effect is used that, depending on the magnitude of the frequency gradient used in the emitted signal, the differences in the propagation time of the multipath components contained in the received signal, after converting the received signal to intermediate constant frequencies, they are presented in the form of larger or smaller frequency differences, the most favorable for the subsequent processing of the frequency The components are extracted from the spectrum of intermediate frequencies (Z'1; Z'2; ... Z'N + X), preferably using filtering devices, and / or the corresponding information parameters of the signal are determined based on the analysis of such components. 19. The method according to any one of paragraphs.14-18, characterized in that, at predetermined intervals, the channel is tuned. 20. The method according to any of paragraphs.14-19, characterized in that in the process of data transmission, continuous identification of the most favorable multipath component of the spectrum of the received signal for subsequent processing is carried out and / or filter settings are updated based on the corresponding analysis of the spectra of intermediate constant frequencies, by which tuning to the channel is implemented as a procedure that is performed continuously and without stopping the actual process of data transfer. 21. The method according to any one of paragraphs.14-20, characterized in that the Doppler frequency shifts that occur during data transfer are determined preferably within the receiving system and are taken into account when generating auxiliary frequencies. 22. The method according to any one of paragraphs.14-21, characterized in that the pairwise processing is carried out using components generated within the system, which in each case have a corresponding frequency response. 23. The method according to any one of paragraphs.14-22, characterized in that in each case a) the support component VK is transformed into a transformed support component VK 'and at least one information component 11; 12; ...; IN into the transformed information component I1 '; I2 '; ...; IN ', as well as b) signal parameters, i.e. the corresponding information modulation parameters are estimated based on the projection I1 '; I2 '; ...; IN 'to the corresponding quadrature components (sine and cosine components) obtained from the corresponding VK'. 24. The method according to claim 23, wherein the reference component is converted into a reference component through the corresponding procedure, which has identical Doppler offsets with the information component processed in each case, so that the multiplication of these two components forms a spectral component (signal) constant frequency. 25. The method according to any one of paragraphs.23 or 24, characterized in that a) multiplying the transformed information component (I1 '; I2'; ...; IN ') with the reference component (RF) to form some first quantity (CQ); b) the transformed information component (I1 '; I2'; ...; IN ') is multiplied with the first time derivative of the reference component (RF) to form a second value (SQ); c) the relationship between these first and second quantities is determined in order to obtain a finite quantity, time-invariant, which depends only on the information parameter, also on the invariant time. 26. An information transmission system comprising a transmitting and receiving device, between which an information signal (IS) is transmitted, the transmitting device comprising means for creating a support component (VK) and at least one information component (I1; I2; ...; IN ), and at least one of these components undergoes a continuous change in frequency during the transmission of a portion of information, as well as a means for generating bit patterns, and the receiving device comprises means for receiving and processing a signal consisting of a reference component (VK) and at least one information component (I1; I2; ...; IN), in which at least one component undergoes a continuous change in frequency during transmission portions of information. 27. The system according to p. 26, characterized in that the transmitting device comprises at least one generator for the formation of a support component (VK) and at least one information component; the first control module, which is connected to the generator and sets the characteristic of the frequency change in time; an encoder or modulator connected to the control module for converting information into a specific signal, as well as a mixer in the connection circuit after the generator and encoder or modulator. 28. System according to any one of paragraphs.26 or 27, characterized in that the receiving device comprises at least one input, a processing device and at least one output, the processing device comprising in series connected the following means: means for separating and converting the signal components , in particular for their conversion to constant intermediate frequencies; means for separating or suppressing signal components that are interferences, as well as means for analyzing parameters. 29. The system according to any one of paragraphs.26-28, characterized in that the means for separating and converting the signal components contains at least one multiplier, through which pairwise multiplication of each of the information components is performed (I1; I2; ...; IN) with a VC reference component, wherein the multiplication products form the spectra of constant intermediate frequencies from which the desired signal components are separated (filtered) by means of a series-connected means for suppressing the components that are interference, containing at least one filtering device, which is then transmitted to a series-connected means for analyzing the parameters. 30. The system according to any one of paragraphs.26-29, characterized in that the separation means further comprises a filtering device with a control module, which is in the connection circuit in front of the multiplier and consists of at least two filter elements connected in series, by which at least one component of the signal is separated from the remaining components of the signal. 31. The system according to any one of paragraphs.29 or 30, characterized in that the means for separating and converting the signal components, in addition to the multiplier designed for pairwise processing of VK and IK, contains an additional device with a module producing auxiliary frequencies, and, if necessary, with an additional multiplier, which provides the conversion of the signal components - if necessary through intermediate stages, at which these components still have time-varying frequencies - to predetermined ranges of constants the covariant intermediate frequencies. 32. System according to any one of paragraphs 28 or 30, characterized in that the means for separating and converting the signal components comprises at least one multiplier and at least one module for generating auxiliary frequencies in the form of one or more generators or the interrogated memory device, by of which reference and information components are individually converted to constant intermediate frequencies lying in a given frequency range, a means is connected in series to these hardware o to suppress constituent noise, which contains at least one filtering device by which the desired signal constituents are extracted from the corresponding spectra of constant intermediate frequencies, thus, in fact, the signal is cleared of constituent interference, and then transmitted to the means connected in series for analysis of parameters. 33. The system according to any one of paragraphs.26-32, characterized in that the means for converting the frequency contains, in addition, at least one converter for equalizing Doppler shifts. 34. The system according to any one of paragraphs.26-33, characterized in that the means for suppressing components that are interferences, contains an additional controllable filter for suppressing components that are interferences. 35. The system according to any one of paragraphs.26-34, characterized in that the means for analyzing the parameters comprises at least one multiplier for pairwise processing in each case of one information component of the signal with at least one reference oscillation, which is generated either inside the system by a corresponding generator it is either read from a storage device, or provided by means of a reference component, and also contains a module that evaluates the values of information parameters. 36. The system according to any one of paragraphs.26-35, characterized in that it further comprises means for setting up in the connection circuit after means for converting the frequency and preferably before means for analyzing the parameters, and also comprises a module for analyzing the frequency spectra and a processing device connected to means for suppressing signal components that are interference. 37. The system according to any one of paragraphs.26-36, characterized in that it further comprises a module for aligning Doppler offsets, which is connected to at least one auxiliary frequency generator and / or an additional module for determining the rate of change of distance between the receiving and transmitting devices.

Images