RU2251802C1 - Method for receiving multiple-beam signal, method for searching for and method for generating soft decisions in case of receiving multiple-beam signal and devices, containing implementations of said method - Google Patents

Abstract

FIELD: radio engineering.

SUBSTANCE: implementation of soft decisions generating method in case of receiving multi-beam signal allows substantial decrease of complication level of receiver, because it contains lesser amount of one-beam receivers, than a prototype.

EFFECT: increased interference resistance and increased capacity of communications system during receiving of multiple-beam signal due to efficient periodic procedure of renewal of multiple-beam signal components when receiving estimates of components search, also considering mutual influence of signal components.

6 cl, 13 dwg

Description

The invention relates to the field of radio engineering, in particular to methods for receiving a multipath signal in communication systems with code division multiplexing and can be used in receiving devices of the base and mobile (subscriber) stations.

In communication systems with moving objects, the signal propagation channels between the receiver and the data transmitter are multipath and non-stationary. The time delays of the components of the multipath signal are in a certain time domain of uncertainty (the multipath region). The effectiveness of communication systems is largely determined by the ability of the search, time synchronization and data processing algorithms to provide the necessary quality of received information in multipath non-stationary channels.

Code separation of channels in communication systems is, for example, using code spreading pseudorandom sequences (PSP) and can be used in the receiving devices of the base and mobile (subscriber) stations.

The delay interval between any two beams in the multipath region is conveniently measured in the PSP chips.

The PSP chip is the duration of one elementary time interval of a pseudo-random sequence.

In communication systems with code division multiplexing under multipath conditions, to improve the quality of communication, multipath (RAKE) receivers are used, which produce a weighted summation of the output signals of a set of single-beam receivers tuned to the signals of the detected rays.

A known method of multipath reception described in patent EP 1041725 A2 "RAKE receiver", H 04 B 1/707, October, 04, 2000.

The method described in the patent is as follows. Periodically search for the temporal positions (delays) of the beam signals of the base stations, obtaining initial estimates of the temporal positions of the beam signals. At each search period, the power of the maximum beams of the base stations is estimated and the base station with the highest beam power is selected for the communication session. The time delays of the signals of the detected beams of the selected base station are estimated. In single-beam receivers, for all beam signals of a selected base station, soft decisions about information symbols are obtained and these soft decisions are combined by weighted summation.

The approach proposed in this patent allows one to efficiently provide reception and improve the quality of received information by periodically analyzing the multipath region and using the base station for which the signal has the highest energy for a communication session.

The known method of multipath reception described in patent EP 0877493 A2 "Synchronization in a multipath spread spectrum communication receiver", H 04 B 1/707, November, 11, 1998.

The method described in the patent is as follows. A multipath signal is received on two antennas spaced apart in space. In the field of multipath, in several stages, a preliminary search is made for the signals of rays spaced from each other by a chip or more. At each stage, the component that has the greatest correlation response is isolated. The analysis interval for each time delay when searching for beam signals is selected depending on the correlation time, which is estimated by the average fading period. For detected signals of beams spaced from each other by a chip or more, the time delay is monitored, soft decisions about information symbols are obtained, and these soft decisions are combined by weighted summation.

The approach proposed in this patent allows one to efficiently provide information reception, firstly, due to spatial diversity, and secondly, due to the use of multi-beam signal components uncorrelated for processing, separated in time by a chip or more, and thirdly, due to the use of adaptive analysis interval when searching for ray signals.

A known method of multipath reception, is shown in US patent No. 5490165 "Demodulation Element Assignment in a System Capable of Receiving Multiple Signals", H 04 B 1/69, Feb.6, 1996.

The described method is as follows. A preliminary search is made for the temporal positions (delays) of the ray signals, obtaining initial estimates of the temporal positions of the ray signals. For each detected beam, the procedure for tracking the time delay of the signal is performed and an updated estimate of the time delay is obtained. In single-beam receivers for detected ray signals with an accurate estimate of the time delay, soft decisions about information symbols are obtained and these soft decisions are combined by weighted summation.

A known method of receiving a multipath signal described in the patent of the Russian Federation No. 2168274.

The described method is as follows. Periodically, a preliminary search is made for the temporal positions (delays) of the ray signals in the multipath interval, obtaining initial estimates of the temporal positions of the ray signals.

For each subsequent search period, a multipath region is formed as an interval of time delays relative to the time position of the signal with the maximum level detected during the initial search. The boundaries of the multipath region are determined by the characteristics of the communication channel.

For each beam signal, its time delay is estimated. For ray signals, soft decisions about information symbols are obtained and combined by weighted summation.

The approach proposed in this patent makes it possible to efficiently ensure the reception of information in a multipath signal by adapting to changing propagation conditions in the communication channel by periodically determining the multipath region, periodically searching and using updated beams on each period.

The disadvantage of the proposed signal processing approaches in the mentioned patents is that during the search for the beam signals the effect of the components of the multipath signal on each other is not taken into account. As a result of this, the probability of erroneous detection of ray signals increases. In addition, the number of ray signals that are used to obtain soft decisions about information symbols is not optimized, which leads to excessive requirements for hardware implementation without increasing the quality of the information that is extracted.

Closest to the proposed solution is a method of receiving multipath signals and a device for its implementation, described in RF patent No. 2120180.

The described method for receiving multipath signals of the prototype is as follows:

The time domain of multipath is determined.

The number and time delays of the components of the multipath signal are periodically determined, for which the following operations are performed:

- A signal is searched in the multipath region and a search estimate of the number and time delays of the multipath signal components is determined.

- Form the updated number and time delays of the component of the multipath signal, for which:

- form a set of components consisting of components of the multipath signal of the previous period and the maximum power of the search component;

- from the formed set of components exclude the minimum component power. The updated time delays of the multipath components are equal to the time delays of the constituent components.

- Find the time delay component of the multipath signal of the current period, updating the updated time delay component of the multipath signal.

Soft decisions about information symbols are formed using the found time delays of the multipath component.

An enlarged structural diagram of a device for implementing the prototype method is shown in figure 2. A detailed structural diagram of the prototype is shown in figure 1.

In figure 1, the following notation is used:

1-1 - 1-L - data receivers,

2-1 - 2-L - multipliers,

3 is a diagram for determining weighting factors,

4 - adder

5 is a decision scheme,

6-1 - 6-M - cluster receivers,

7-1 - 7-M - multipliers,

8 is a diagram for detecting and analyzing a cluster of rays,

9 - search receiver,

10 - switch

11 - control unit.

In figure 2 is indicated:

11 - control unit

12 is a search block,

13 is a multi-channel data receiver,

14 is a scheme for combining soft decisions,

15 is a block forming soft decisions.

The enlarged structural diagram of the prototype device includes a search unit 12, which includes a search receiver 9 and a detection and analysis circuit of a cluster of rays 8, a multi-channel data receiver 13, containing L data receivers 1-1 - 1-L and M - receivers of a cluster of rays 6-1 - 6-M and switch 10, control unit 11, and soft decision combining scheme 14, which includes 2-1 - 2-L multipliers, 7-1 - 7-M multipliers, weighting coefficient determination scheme 3, adder 4, and decision circuit 5.

In this case, the first and second inputs of the search unit 12 and the multi-channel data receiver 13 are combined and are signal inputs of the device. The output of the search unit 12, which is the output of the search estimate of the number and time delays of the multipath component, is connected to the first input of the control unit 11.

The first outputs of the multi-channel data receiver 13, which are the outputs of the correlation responses of the symbols of the set of components of the multipath signal of the previous period, are connected to the second inputs of the control unit 11.

The first output of the control unit 11, which is the control output, is connected to the third input of the search unit 12.

The second outputs of the control unit 11, which are outputs of the updated number and time delays of the multipath components, are connected to the third inputs of the multi-channel data receiver 13.

The second output of the multi-channel data receiver 13, which is the output of the correlation responses of the symbols of the components of the multipath signal with updated updated time delays, is connected to the input of the soft decision combining circuit 14. The multi-channel data receiver 13 and soft decision combining circuit 14 are combined into a soft decision forming unit 15, output which is the output of soft device solutions.

The device prototype receiving multipath signals, presented in figure 2, works as follows.

An input complex multipath signal containing in-phase and quadrature components (SS and FAC) is fed to the inputs of the search unit 12 and to the inputs of the multi-channel data receiver 13, with each data receiver processing a single beam signal. The signals from the first output of the multi-channel data receiver 13, which are the correlation responses of the symbols of the components of the multipath signal with updated updated time delays, are fed to the second input of the control unit 11. The signals from the second output of the multi-channel data receiver 13, which are the correlation responses of the symbols of the components of the multipath signal with updated updated time delays, input to the circuit combining soft decisions 14.

In the soft decision combining scheme 14, the correlation responses of the multipath signal components with the updated updated time delays of the multi-channel data receiver 13 are multiplied by the formed weight coefficients so that a larger signal corresponds to a larger signal. Then, the weighted output signals of the multiplication schemes are summed up, as a result of which a soft decision is formed about the received information signal.

The search block 12 for control signals from the first output of block 11 periodically analyzes the multipath region, and at each period determines the number and time delays of the multipath component.

The maximum power of the search component is fed to the input of the control unit 11, where it is compared with the minimum power component of the multipath signal of the previous period, received from the multi-channel data receiver 13.

If the maximum power of the search component is greater than the minimum power of the component of the multipath signal of the previous period, then a set of components is formed consisting of the components of the multipath signal of the previous period and the maximum power of the search component. At the same time, the minimum power component of the previous period is excluded from the formed set of components. Thus, a set of components of the multipath signal of the current period is formed with updated time delays, which are further refined by the delay tracking system in the multi-channel data receiver 13.

The method of searching for the multipath signal of the prototype according to the patent of the Russian Federation No. 2120180 is as follows:

- Form the complex correlation responses of the pilot signal, determining the correlation of the input signal with the reference signal shifted by the discrete search time within the multipath region.

- Determine the values of the decisive function for the given discrete time delays of the multipath region by summing the squares of the in-phase and quadrature parts of the corresponding complex correlation responses of the pilot signal.

- Compare the values of the generated decision function with a given threshold h.

- Form estimates of the search for time delays of the multipath signal components by exceeding the threshold h.

The search time discrete is selected, for example, to be equal to half the memory bandwidth chip.

The threshold h is selected in proportion to the noise power.

To implement the search method in the prototype, a search unit 12 is used, which includes a search receiver 9 and a detection and analysis scheme for the cluster of rays 8. The block diagram of the search unit 12 is shown in FIG. 3, where it is indicated:

16-1 - 16-K - quadrature correlators,

17 - pseudo-random sequence generator (PSP),

18 is a diagram of the formation of the threshold,

19 is a control circuit,

20 - multiplexer.

The search unit 12 contains K quadrature correlators 16-1 to 16-K, the first and second inputs of which are combined with the first and second inputs of the threshold formation circuit 18 and are signal inputs of the device. The third input of each quadrature correlator 16-1 - 16-K is the input of the SRP signal and connected to the corresponding output of the SRP 17 generator.

The fourth input of each quadrature correlator 16-1 - 16-K, which is the input of the control signal, is connected to its corresponding first output of the control circuit 19. The second output of the control circuit 19, which is the output of time shifts of the PSP generator 17, is connected to the input of the PSP generator 17 The output of each quadrature correlator 16-1 to 16-K, which is the output of the values of the decisive functions for the given time delays of the multipath region, is connected to the corresponding input of the multiplexer 20.

The output of the multiplexer 20, which is the output of the values of the critical functions from the multipath region, is connected to the third input of the detection and analysis circuit of the beam cluster 8. Simultaneously, the output of the multiplexer 20 and the outputs of the detection and analysis circuit of the beam 8 are connected to the input of the control unit 11.

The output of the threshold formation circuit 18, which is the output of the threshold value h, is connected to the second input of the beam cluster 8 detection and analysis circuit.

The input of the control circuit 19 and the first input of the detection and analysis circuit of the cluster of beams 8, which are inputs of control signals, are connected to the first output of the control unit 11.

The search unit 12 operates as follows.

The input signal is supplied to the first and second inputs K of the parallel quadrature correlators 16-1 - 16-K and to the first and second inputs of the threshold formation circuit 18. The reference signal from the PSP 17 generator is supplied to the third input of the quadrature correlators 16-1 - 16-AK. For adjacent quadrature correlators, this signal is shifted by a discrete search time equal to, for example, half the memory bandwidth chip. The time shifts (delays) of the reference signals of the PSP generator 17 sets the signal of the time shifts of the PSP generator of the control circuit 19. In each quadrature correlator 16-1 - 16-K, the input pilot signal is multiplied by the reference signal. The multiplication results are accumulated, forming the in-phase and quadrature components of the complex correlation responses of the signal. The squares of the in-phase and quadrature parts of the corresponding complex correlation responses are summarized, as a result of which the values of the decisive function are generated for the given discrete time delays of the multipath region. The values of the generated decision function through the multiplexer 20 are fed to the third input of the detection and analysis of the cluster of rays 8 and to the input of the control unit 11.

The control circuit 19, upon a signal from the output of the control unit 11, carries out a time shift of the pseudo-random sequence generator 17 and resets the adders of the quadrature correlators 16-1 to 16-K.

According to the input signal, the threshold formation circuit 18 forms a threshold h, which is selected, for example, in proportion to the noise power. The value of the threshold h goes to the second input of the detection and analysis of the cluster of rays 8.

In Scheme 8, the values of the decision functions are compared with the threshold h. When threshold h is exceeded, a decision is made about the presence of a signal for given discrete time delays of the multipath region, and estimates of the search for time delays of the multipath signal components are generated. The generated estimates are sent to the control unit 11.

The method of forming soft decisions when receiving a multipath signal of the prototype according to the patent of the Russian Federation No. 2120180 is as follows:

- Form a sequence of correlation responses of information symbols for each specified component of the multipath signal.

- Carry out a weighted summation of the correlation responses of the information symbols of all the specified components of the multipath signal, obtaining the united soft decisions about information symbols.

Correlation responses of information symbols for each selected component of a multipath signal are generated, for example, by determining the correlation of the input signal with the known SRP at symbol duration intervals.

The weights when summarizing the correlation responses of the information symbols of the specified components of the multipath signal are selected in such a way that a higher weight coefficient corresponds to a signal with a higher power.

The block diagram of the device for implementing the prototype method of forming soft solutions is shown in Fig.4.

The soft decision forming unit 15 includes a multi-channel data receiver 13 and a soft decision combining scheme 14.

In this case, the first and second inputs of the multi-channel data receiver 13 are the signal inputs of the device. The first outputs and third inputs of the multi-channel data receiver 13 are connected to the control unit 11. The second output of the multi-channel data receiver 13, which is the output of the sequence of correlation responses of the symbols of the specified multipath signal components, is connected to the inputs of the soft decision combining circuit 14, the output of which is the output of the combined soft decisions devices.

Block 15 operates as follows:

In the multi-channel data receiver 13, a sequence of correlation responses of information symbols of each specified component of the multipath signal is formed. For this, the initial estimates of the time delays of the component are refined using a time delay tracking system. Next: determine the correlation of the input signal with a known SRP corresponding to the specified time delay of the multipath component. The generated correlation responses of the refined components of the information symbols of the multipath signal are fed to the soft decision combining scheme 14, where they are weighted summarizing, obtaining the combined soft decisions about the information characters that are output to the device.

The weights when summing the correlation responses of the information symbols of the specified components of the multipath signal are chosen in such a way that a greater weight coefficient corresponds to a signal with a higher power.

This method of receiving a multipath signal has the following disadvantages.

According to the description of the prototype at each period, the search receiver sequentially scans the multipath region. Moreover, in the field of multipath, the search for the number and time delays of the components of the multipath signal is evaluated without taking into account the influence of lateral emissions of the correlation responses of the multipath components on each other.

As a result of the update procedure, one of the weakest components of the aggregate of the previous period is replaced by the maximum component of the aggregate of the current period. This leads to some energy gain. However, there may be cases in which several newly discovered constituent components of the current period outperform the constituent components of the previous period. Therefore, this approach when processing the input multipath signal leads to energy losses, because some of the more powerful components of the multipath signal of the current period are not used. In addition, the maximum constituent component of the current period may have a temporary position (delay) close to the temporal position of the constituent components of the previous period. This leads to the irrational use of single-beam receivers. As a result, all the selected channels of the multi-channel data receiver can be tuned to the same most powerful component of the multipath signal combination.

In addition, the number of ray signals that are used to obtain soft decisions about information symbols is not optimized, which leads to excessive requirements for hardware implementation without increasing the quality of the information that is extracted.

The problem that the invention solves is to increase the noise immunity and increase the capacity of a communication system when receiving a multipath signal due to an effective periodic procedure for updating the components of a multipath signal when receiving an estimate of the components search, taking into account the influence of the multipath signal components on each other during the search procedure, optimizing the number of components, which are used to obtain soft decisions about information symbols.

To solve this problem, a method of receiving a multipath signal, which consists in the fact that when receiving periodically determine the number and time delay component of the multipath signal, for which:

- determine the time domain of multipath,

- conduct a signal search in the field of multipath and determine the search estimate of the number and time delays of the components of the multipath signal,

- form the updated number and time delay component of the multipath signal,

- find the time delay component of the multipath signal of the current period, constantly updating the updated time delay component of the multipath signal,

- using these time delays, form soft decisions about information symbols,

additionally introduce the following operations:

- generate updated numbers and time delays of the multipath component from the components of the multipath signal of the previous period and those search components whose time delays differ from the time delays of each component of the multipath signal of the previous period by more than a predetermined value Th, and

- if the number of components of the formed population does not exceed a given number L, then the updated number of components is equal to the number of components of the formed population, and the updated time delays of the components of the multipath signal are equal to the time delays of the components of the population,

- if the number of components of the formed population is greater than a given number L, then the power of the components of the population is estimated and L of the most powerful components is selected, the time delays of which are updated time delays of the multipath component, in this case, the updated number of components is L.

The time domain of multipath is defined as the time domain of delays, the beginning of which is t1 less than the minimum time delay of the components of the multipath signal of the previous period, and the end is t2 more than the maximum time delay of the components of the multipath signal of the previous period.

The value of Th is chosen, for example, equal to half the chip memory bandwidth.

The value of L is selected, for example, equal to the maximum number of multipath components used to demodulate data.

The power of the multipath components can be estimated by the decisive search function.

To solve the same problem, in the method of receiving a multipath signal, a method for searching for a multipath signal is used, which consists in the fact that

- form the complex correlation responses of the pilot signal, determining the correlation of the input signal with the reference signal corresponding to the given discrete time delays of the multipath region,

- determine the values of the decisive function for the given discrete time delays of the multipath region by summing the squares of the in-phase and quadrature parts of the corresponding complex correlation responses of the pilot signal,

- compare the values of the generated decision function with a given threshold h,

additionally introduced the following operations:

- form the totality of the values of the decisive function that exceeded the threshold, and the totality of the corresponding time delays (positions),

- stage-by-stage estimates of the search for the time delays of the components of the multipath signal are formed, and at each stage, an estimate of one component is obtained, for which:

- determine the element with the maximum value from the set of values of the decisive function of the current stage, while the set of values of the decisive function of the first stage is the formed set of values of the decisive function,

- compare the maximum value of the decisive function of the current stage with a threshold h,

- in the case of not exceeding the threshold, the formation of an estimate of the search for time delays of the components of the multipath signal is stopped,

- in case of exceeding the threshold:

the time delay corresponding to the element with the maximum value of the decisive function of the current stage is considered an estimate of the search for the time delay of the components of the multipath signal,

correct the elements of the set of values of the decisive function of the current stage,

form a set of values of the decisive function of the subsequent stage, excluding from the adjusted set of values of the decisive function of the current stage the element with the maximum value.

The threshold h is selected, for example, proportional to the noise power.

Correction of elements of the set of values of the decisive function of the current stage is performed, for example, by subtracting the correction value from the elements of the set, the time delays of which differ from the time delay of the maximum element by no more than a given time threshold η.

The correction value is selected, for example, proportional to the maximum element and inversely proportional to the module of the difference in time delays of the maximum and the adjusted element of the set of values of the decisive function of the current stage.

To receive a multipath signal, it is proposed to use a method of forming soft decisions, which consists in the fact that

- form a sequence of correlation responses of information symbols for each component of the multipath signal,

- carry out a weighted summation of the correlation responses of information symbols of all components of the multipath signal, obtaining the united soft decisions about information symbols,

additionally introduced the following operations:

before the formation of a sequence of correlation responses of information symbols

- carry out the selection of the components of the multipath signal, for which:

- evaluate the power of the components of the multipath signal,

- perform the ranking of the components of the multipath signal in decreasing order of their power,

- sequentially for all ranked components of the multipath signal, except the first,

form the ratio of the power of the component to the sum of the power of all previous ranked components,

comparing the generated ratio with a given threshold ν,

if the threshold is exceeded, the component is selected to form soft decisions,

in case of not exceeding the threshold, the component selection procedure is completed,

- the formation of a sequence of correlation responses of information symbols and their weighted summation is carried out for the selected components of the multipath signal.

The threshold ν can be selected, for example, by allowable energy loss due to the limitation of the components used in demodulation.

The correlation responses of information symbols for each selected component of the multipath signal are generated, for example, by determining the correlation of the input signal with the known SRP over the symbol duration interval.

Weighting coefficients when summarizing the correlation responses of information symbols of the selected components of the multipath signal are selected as complex conjugate estimates of the complex envelope of the information symbols of the components of the multipath signal.

To increase the noise immunity and increase the capacity of the communication system when receiving a multipath signal using the periodic procedure for updating the multipath signal components, a multipath signal receiving device is proposed comprising a search unit, a control unit, a soft decision forming unit, the first and second inputs of the search unit and soft decision forming unit combined and are the inputs of the in-phase and quadrature components of the input multipath signal, the third input of the search unit is connected to the first m is the output of the control unit, which is the control output, the second outputs of the control unit, which are the time delay outputs of the updated components of the multipath signal of the current period, are connected to the third inputs of the soft decision forming unit, the first output of the soft decision forming unit is the output of soft decision about information symbols and the output of the device into which, according to the invention, a signal component updating unit is additionally introduced, the output of the search unit being the output estimates of the time delays and the values of the deciding function of the search for the multipath component, connected to the first input of the signal component update unit, and the fourth input of the soft decision forming unit, the output of the signal component update unit, which is the time delay output of the updated multipath component, is connected to the input of the control unit, the second output of the soft decision forming unit is the output of the pilot symbol correlation responses and time delays of the multipath signal component n edyduschego period and is coupled to a second input of block update signal component.

In the device for receiving a multipath signal, it is proposed to use a multipath signal search unit containing K quadrature correlators, a threshold generating circuit, a pseudo-random sequence generator, a control circuit, a multiplexer, the first and second inputs of the correlators being combined with the first and second inputs of the threshold generating circuit and are common mode inputs and of the quadrature components of the input multipath signal, the third input of each quadrature correlator is the input of the pseudo-random signal sequence and is connected to the corresponding output of the pseudo-random sequence generator, the fourth input of each quadrature correlator, which is the input of the control signal, is connected to the corresponding first output of the control circuit, the second output of the control circuit, which is the output of time offsets of the pseudo-random sequence generator, is connected to the generator input pseudo-random sequence, the output of each quadrature correlator, which is the output of the values I decide functions for the given time delays of the multipath region, connected to the corresponding input of the multiplexer, the input of the control circuit is an input of control signals, into which according to the invention a ray detection circuit, a first, second and third key, a first and second comparator, a logical node, a counter, a node are introduced control, OR circuit, node for determining the maximum element, the first and second subtracter, module former, register, memory element of the initial set of decision functions, memory element of the original set spacing of time delays, a memory register of decision search functions, a memory register of search estimates of time delays, a shaper of a correction signal, the output of the multiplexer, which is the output of the values of the decision function of the multipath region, is connected to the first input of the beam detection circuit, the second output of the control circuit, which is the output time shifts of the pseudo-random sequence generator, connected to the second input of the ray detection circuit, the output of the threshold formation circuit, which is the output of threshold, connected to the third input of the ray detection circuit and the first input of the first comparator, the first output of the ray detection circuit, which is the time delay output, is connected to the second input of the memory element of the original set of time delays, the second output of the ray detection circuit, which is the output of decisive values functions that exceed the threshold is connected to the second input of the memory element of the initial set of decision functions, the third output of the ray detection circuit, which is the output of a signal equal to h if detected components that exceed the threshold are connected to the first input of the control unit, the fourth output of the ray detection circuit, which is the output of the signal indicating the threshold is exceeded for a given discrete time delay, is connected to the second input of the control unit, the first output of the control unit, which is the output of the recording signal connected to the first input of the memory element of the original set of decision functions and with the first input of the memory element of the original set of time delays, the output of the original memory element with the aggregate of decisive functions, which is the output of the decisive function of the current stage, is connected to the first input of the node for determining the maximum element, with the first input of the second subtractor, with the first inputs of the first and second keys, the second input of the node for determining the maximum element is the control input and connected to the fifth output of the node control, the output of the node determining the maximum element, which is the output address of the selected maximum element, is connected to the fifth input of the control node, the second output of the node is controlled I, which is the output of the signal indicating the presence of a maximum, is connected to the second input of the first key and the first input of the logical node, the output of the first key is connected to the second input of the first comparator, the output of the first comparator, which is the output of exceeding the threshold, is connected to the second input of the logical node, the first the counter input, with the second controlled inputs of the second and third keys, with the first input of the register, the third output of the control node, which is the output of the number of detected components of the current search period, with is single with the second input of the counter, the output of the counter is connected to the first input of the OR circuit, the second input of which is connected to the output of the logical node, the output of the OR circuit, which is the output of the end indicator of the search procedure, is connected to the third input of the control unit, the fourth output of the control unit, which is the output of the control signal for generating a search estimate, is connected to the first input of the memory register of the decision functions of the search, the first input of the register of the memory of the evaluation of the search for time delays and to the third input of the signal generator la corrections, the second input of the memory register of the decision functions of the search and the first input of the driver of the correction signal are combined and connected to the output of the second key, which is the output of the maximum value of the decisive function of the current stage, the second inputs of the register of the evaluation of the search for time delays and the register are combined and connected to the output of the third key, the first input of which is combined with the first input of the first subtractor and connected to the output of the memory element of the original set of time delays, which is the output of the value the time delay corresponding to the maximum element, the output of the register, which is the output of the time delay values, is connected to the second input of the first subtracter, the output of the first subtractor, which is the output of the time delay difference, is connected to the input of the module shaper, the output of the module of the signal difference of the module shaper is connected to the second input of the correction signal generator and the first input of the second comparator, the second input of the second comparator is a threshold input, the output of the second comparator is is the output of exceeding the threshold and connected to the fourth input of the control unit and the fourth input of the correction signal generator, the output of the correction signal generator, which is the output of the correction signal, is connected to the second input of the second subtractor, the output of which is connected to the second input of the memory element of the initial set of decision functions, register output the memory of the evaluation of the search for time delays and the output of the memory register of the crucial search functions are combined and are the output of the search block.

In the device for receiving a multipath signal, it is proposed to use a soft decision generating unit containing a multi-channel data receiver and a soft decision combining circuit, the first and second inputs of the multi-channel data receiver are inputs of the in-phase and quadrature components of the input signal and the first and second inputs of the device, the output of the multi-channel data receiver is connected with the input of the soft decision combining scheme, which is the input to the sequence of correlation responses of information and the symbol lot of the multipath component components, the output of the combined soft decisions of the soft decision combining circuit is the first output of the block into which, according to the invention, the memory register, the first, second and third multiplexers, the power estimation unit, the key, the first, second and third memory nodes, a counter are additionally introduced , adder, first and second comparators, OR circuit, NOT circuit, ranking node, relation former, control node, and the third block input is the time delay inputs and the first inputs of the second multip lexor, the fourth input of the block is the input of the values of the search function of the current stage and the first input of the first multiplexer, the output of the multi-channel data receiver is connected to the input of the power estimation node, which is the input to the sequence of correlation responses of the pilot symbols of the multipath component of the signal, in addition, the output of the multi-channel data receiver is the output of the sequence of correlation responses of the pilot symbols and time delays of the multipath component and the second output of the device VA, the second inputs of the first and second multiplexers are control and connected to the first output of the control unit, the output of the power estimation unit, which is the output of the power estimation of the multipath component, is connected to the first input of the third multiplexer, the output of the first multiplexer is combined with the output of the third multiplexer, which is the output of the signal for estimating the power of the component of a multipath signal, and connected to the first input of the third memory node, the output of the second multiplexer, which is the time output the delay of the updated components of the multipath signal, connected to the first input of the first memory node, the second input of which is combined with the second input of the third multiplexer and connected to the second control output of the control node, the third output of the control node is the output of the control signals and connected to the second input of the third memory node, the first input of the first comparator and the first input of the memory register, the output of the third memory node, which is the output of the power rating of the current component, is connected to the first input of the node operation, the first input of the adder and the first input of the ratio former, the output of the ranking node, which is the output of the ordered addresses of the memory cells, is connected to the first input of the control node, the second input of the ranking node is connected to the fourth control output of the control node, the fifth output of the control node, which is the output a recording signal, connected to the second input of the second memory node, the first input of which is connected to the output of the adder, the output of the second memory node, which is the output of the total powers of all of the previous ranked components, connected to the second input of the adder and the second input of the ratio former, the output of the relationship former is connected to the first input of the second comparator, the second input of which is a threshold input, the output of the second comparator, which is the output of the threshold exceeding signal, is connected to the input of the circuit NOT, with counter input and with the first key input, the second key input is connected to the output of the first memory node, the key output is connected to the second input of the memory register, which is a time delay input and components of the multipath signal, the output of the memory register, which is the output of the set of time delays of the selected components, is connected to the third input of the multi-channel data receiver, the output of the circuit is NOT an inverse output and connected to the first input of the OR circuit, the counter output is the signal output equal to the number of threshold exceeded , and connected to the second input of the first comparator, the output of the first comparator is connected to the second input of the OR circuit, the output of which is the output of the signal indicating the completion of the rejection procedure a component for generating the soft decisions, and is coupled to a second control input node.

A comparative analysis of the method of receiving a multipath signal, the search method for receiving a multipath signal, the method of generating soft decisions when receiving a multipath signal with a prototype shows that the proposed invention is significantly different from the prototype, as it allows to increase noise immunity and increase the capacity of the communication system when receiving a multipath signal.

A comparative analysis of the proposed methods with other technical solutions in the art did not allow to identify the features claimed in the distinctive parts of the claims. Therefore, the inventive method for receiving a multipath signal, a search method for receiving a multipath signal, a method for generating soft decisions when receiving a multipath signal meet the criteria of "novelty", "technical solution of the problem", "significant differences" and has non-obvious solutions.

A comparative analysis of the multipath signal reception device, the multipath signal search receiver and the soft decision driver when receiving the multipath signal with the prototype shows that the proposed inventions are significantly different from the prototype, as they lead to increased noise immunity and increase the capacity of the communication system when receiving a multipath signal due to the introduction of new blocks and connections of device blocks.

A comparative analysis of the claimed devices with other technical solutions in the art did not allow to identify the features claimed in the distinctive parts of the claims. Therefore, the inventive device for receiving a multipath signal, a multipath signal search receiver and a soft decision driver when receiving a multipath signal meet the criteria of "novelty", "technical solution of the problem", "significant differences" and have non-obvious solutions.

Graphic materials presented in the application materials:

Figure 1 is a detailed structural diagram of a device for receiving a multipath signal (prototype).

Figure 2 is an enlarged structural diagram of a device for receiving a multipath signal (prototype).

Figure 3 is a structural diagram of a search unit (prototype).

Figure 4 is a structural diagram of a block forming soft decisions (prototype).

5 is a structural diagram of the proposed device receiving a multipath signal.

6 is an embodiment of a signal component update unit.

7 is an embodiment of a power estimation unit.

Fig. 8 is an embodiment of a selection node.

Fig.9 is a structural diagram of the proposed search unit when receiving a multipath signal.

10 is an embodiment of a beam detection circuit.

11 is an embodiment of a correction signal driver.

12 is an embodiment of a logical node.

Fig is a structural diagram of the proposed block forming soft decisions when receiving a multipath signal.

The proposed method of receiving a multipath signal is as follows.

The number and time delays of the components of the multipath signal are periodically determined, for which the following operations are performed:

- Determine the time domain of multipath.

- A signal is searched in the multipath region and a search estimate of the number and time delays of the multipath signal components is determined.

- Form the updated number and time delays of the component of the multipath signal, for which:

- form a set of components consisting of components of the multipath signal of the previous period and those search components whose time delays differ from the time delays of each component of the multipath signal of the previous period by more than a predetermined value Th;

- if the number of components of the formed population does not exceed a given number L, then the updated number of components is equal to the number of components of the formed population, and the updated time delays of the components of the multipath signal are equal to the time delays of the components of the population;

- if the number of components of the formed population is greater than a given number L, then the power of the components of the population is estimated and L most powerful components are selected whose time delays are updated time delays of the multipath component. In this case, the updated number of components is L.

- Find the time delay component of the multipath signal of the current period, constantly updating updated time delays of the component of the multipath signal.

- Using these time delays, soft decisions about information symbols are formed.

The time domain of multipath is defined as the time domain of delays, the beginning of which is t1 less than the minimum time delay of the components of the multipath signal of the previous period, and the end is t2 more than the maximum time delay of the components of the multipath signal of the previous period.

The value of Th is chosen, for example, equal to half the chip memory bandwidth.

The value of L is selected, for example, equal to the maximum number of multipath components used to demodulate data.

The power of the components of the multipath signal is estimated, for example, by the decisive search function.

To implement the proposed method for receiving a multipath signal, the device shown in Fig. 5 is used, where it is indicated:

11 - control unit

12 is a search block,

15 - block forming soft decisions,

21 is a signal component update unit.

The multipath signal receiving device comprises a search unit 12, a control unit 11, a soft decision generation unit 15, a signal component update unit 21. The first and second inputs of the search unit 12 and the soft decision generation unit 15 are combined and are inputs of the in-phase and quadrature components of the input multipath signal ( signal inputs of the device).

The third input of the search unit 12 is connected to the first output of the control unit 11, which is the control output. The second outputs of the control unit 11, which are the time delay outputs of the updated multipath signal components of the current period, are connected to the third inputs of the soft decision forming unit 15. The first output of the soft decision forming unit 15 is the output of soft decisions about information symbols and the output of the device.

The output of the search unit 12, which is the output of estimates of the time delays and the values of the decisive function of the search for the components of the multipath signal of the current period, is connected to the first input of the update component of the signal component 21 and the fourth input of the soft decision forming unit 15.

The output of the update component of the signal component 21, which is the time delay output of the updated components of the multipath signal of the current period, is connected to the input of the control unit 11.

The second output of the soft decision forming unit 15 is the output of the pilot symbol correlation responses of the updated multipath signal components and the updated time delays of the multipath signal component of the previous period and is connected to the second input of the signal component update block 21.

The proposed device for receiving a multipath signal works as follows.

The input complex signal of the device, containing the in-phase and quadrature components, is fed to the first and second inputs of the search unit 12 and to the first and second inputs of the soft decision forming unit 15.

In the search unit 12 periodically analyze the multipath region. Moreover, at each current search period, for example, as in the prototype, by correlation processing and comparing the values of the decisive search function with a threshold, the search estimate of the number and time delays of the components of the multipath signal is determined. The estimation of the time delays of the multipath signal components and the corresponding values of the decisive search function are received at the first input of the signal component update block 21 and at the fourth input of the soft decision block 15.

The multipath region for the current search period is determined in the control unit 11 by the signal that is supplied to its input from the output of the update unit of the signal component 21.

The boundaries of the multipath region of the current search period are determined, for example, as follows: for the lower boundary, take the time delay located at t1 before the time delay of the first detected component of the multipath signal (components with the least time delay), and take the time delay located at t2 as the upper boundary later than the time delay of the last detected multipath component.

Information about the size and temporary position of the multipath region from the first control output of the control unit 11 is supplied to the third input of the search unit 12.

In the block update component of the signal 21 form the updated number and time delay component of the multipath signal:

- For the current search period, by estimating the time delays and the values of the decisive search function of the components of the multipath signal of the search unit 12 and the correlation responses of the updated components of the multipath signal with the specified time delays of the soft decision forming unit 15.

- For the first search period according to the estimated time delays of the component of the multipath signal of the search unit 12.

The updated time delays of the components of the multipath signal of the current period from block 21 are received at the input of the control unit 11, from the second output of which the set of updated time delays of the components of the multipath signal of the previous period is supplied to the third input of the soft decision forming unit 15.

Based on the generated time delays in the soft decision forming unit 15, the pseudorandom sequences of the data receivers 13 are tuned, the time delays of the updated components of the multipath signal of the current period are refined, and the correlation responses of the symbols of the refined components are generated.

From the second output of the soft decision forming unit 15, the updated time delays and the corresponding correlation responses of the pilot symbols of the updated multipath components are supplied to the second inputs of the signal updating unit 21.

In the soft decision forming unit 15, a weighted summation of the generated correlation responses of the information symbols of the refined components of the multipath signal is obtained, obtaining the united soft decisions about the information symbols that are output to the device.

The weights when summing the output signals of the soft decision forming unit 15 are selected, for example, as complex conjugate estimates of the complex envelope of the information symbols of the updated components of the multipath signal with the specified time delays of the data receivers.

The control unit 11 is a typical unit and can be implemented by various means, for example, on modern digital signal processing (DSP) microprocessors - TMS320Cxx, Motorola 56xxx, Intel, etc.

An embodiment of the updating unit of the signal components 21 of the proposed multipath signal receiving device is shown in FIG. 6.

In Fig.6, the following notation is used:

22, 32, 36, 41 - multiplexers,

23, 33, 35 - keys,

24 - subtractor,

25 - register

26 is an OR-NOT circuit,

27 - trigger

28 - shaper module

29, 39 - comparators,

30 - circuit NOT

31 is a counter

34 - adder

37, 38, 42 - memory nodes,

40 - node assessment power

43 - node selection

44 - control unit.

The nodes 22-31, which are part of block 21, determine the search components of the current stage, the time delays of which differ from the time delays of each component of the multipath signal of the previous period by more than a predetermined value Th.

In the memory node 37, a set of time delays of the components of the multipath signal of the previous period and those components of the search of the current stage, the time delays of which differ from the time delays of each component of the multipath signal of the previous period by more than a predetermined value Th, are formed and stored.

The power estimation node 40, the multiplexer 41, the memory node 42, the multiplexer 32, and the key 33 are used to form in the memory node 42 a combination of power estimates (decisive functions) of the multipath signal component of the previous period and power search estimates (decisive function) of the multipath signal components of the current period, the time delays of which differ from the time delays of each component of the multipath signal of the previous period by more than a predetermined value Th.

The adder 34, the memory node 38, the key 35 and the comparator 39 generate signals about the number of components in the memory nodes 37 and 42 and a control signal about the fact that the signal exceeds the number of components of a given number L.

The selection node 43 generates the addresses of the set of updated time delays L. the most powerful components of the multipath signal recorded in the memory node 37.

The update unit of the signal component 21 operates as follows.

From the search unit 12, the first inputs of the multiplexer 22, the first inputs of the multiplexer 32, and the first inputs of the control unit 44 receive, respectively, estimates of the search for the time delays of the multipath signal components, estimates of the power search (decision function), and an estimate of the search for the number of components of the multipath signal of the current period.

From the soft decision forming unit 15, the time delays and correlation responses of the pilot symbols of the updated multipath signal components of the previous period are received respectively at the first input of the multiplexer 36 and the first input of the power estimation unit 40.

In the power estimation unit 40, according to the correlation responses of the symbols of the updated components and the reset signals from the second output of the control unit 44 to the second input of the power estimation unit 40, a power estimate (decisive function) of the updated components of the multipath signal of the previous period, which is transmitted to the first input of the multiplexer, is generated 41.

From the third output of the control unit 44, a switching signal is supplied to the second input of the multiplexer 41, through which the multiplexer and the first input of the memory node 42 produce power estimates (decisive functions) of the updated multipath signal components of the previous period.

In the memory node 42, according to the write signal coming from the fourth output of the control node 44 to the second input of the memory node 42, a set of power estimates (a decisive function) of the updated components of the multipath signal of the previous period is formed, and according to the read signal coming from the same output of the control unit 44 , this combination from the output of the memory node 42 is supplied to the first input of the selection node 43.

From the first output of the control unit 44, a switching signal is supplied to the second input of the multiplexer 36, through which the multiplexer 36 and the first input of the memory node 37 receive estimates of the updated time delays of the multipath component of the previous period.

In the memory node 37, according to the recording signal from the third output of the control node 44 to the second input of the memory node 37, a set of estimates of the updated time delays of the components of the multipath signal of the previous period is formed.

The following describes the procedure for generating a set of time delays of the previous period and those components of the search for the current stage whose time delays differ from the time delays of each component of the multipath signal of the previous period by more than a predetermined threshold Th.

From the first output of the control unit 44, a switching signal is supplied to the second input of the multiplexer 22 and to the second input of the multiplexer 32, by which the output of the multiplexer 22 evaluates the time delay search components of the multipath signal of the current period, and the output of the multiplexer 32 estimates the power search (decisive functions) of the corresponding components of the multipath signal.

At the same time, from the first output of the control unit 44, the corresponding clock signals, write and read signals, and the first input of the counter 31 receive clock signals and search estimates for the number of components of the multipath signal of the current period to the second input of the register 25.

Elementally, evaluations of the search for time delays of the components of the multipath signal of the current period from the output of the multiplexer 22 are received at the first input of the key 23 and the second input of the subtractor 24. At the first input of the subtractor 24 from the output of the memory node 37, by the read command from the third output of the control unit 44, the estimates of the updated time delays of the component of the multipath signal of the previous period.

Switching of the multiplexer 22 and reading from the memory node 37 is carried out according to the control signals from the control node 44 so that the time delay search estimate of each component of the multipath signal of the current period is sequentially compared in the subtractor 24 with all estimates of the updated time delay of the multipath signal component of the previous period.

The output signal of the subtractor 24, which is the value (value) of the difference in the search for time delays of the components of the multipath signal of the current period and the evaluation of the updated time delays of the component of the multipath signal of the previous period, is input to the shaper module 28, in which the module of this signal is generated. The generated signal from the output of the shaper module 28 is fed to the first input of the comparator 29, the second input of which serves a predetermined threshold signal Th. The value of the threshold signal Th is chosen, for example, to be equal to half of the memory bandwidth chip.

When the output signal of the driver of the module 28 of the signal threshold Th at the output of the comparator 29 generate a signal equal to, for example, a logical unit. Otherwise, a signal is generated equal to, for example, logical zero. This signal is fed to the input of the circuit NOT 30. From the output of the circuit NOT 30, an inverse signal is fed to the first input of the register 25 with a serial input and a parallel output, for example, implemented on the basis of standard microcircuits of the 561PR1 series or 1564IR8 series. The register 25 is pre-reset by power on or by the signal indicating the end of the current comparison cycle, which is formed at the output of the counter 31 during operation. The signal indicating the end of the current comparison cycle from the output of the counter 31 is fed to the third input of register 25 and to the second input of trigger 27.

The signal indicating the end of the current comparison cycle can be generated using standard programmable counters, for example, 564IE10 series or others.

The initial state of the counter 31 is set by a control command from the first output of the control unit 44 to the control input of the counter according to the known number of components of the totality of the previous period, which are stored in the memory node 37. In this case, a clock signal is sent to the counting input of the counter 31, and to the second input of the register 25 provides a recording signal.

The signal indicating the end of the current comparison cycle is formed, for example, at the time of the appearance of the logical unit signal at the output of the high order of the counter 31, which corresponds to the completion of reading from the memory node 37 all the components of the totality of the previous period.

With this organization of the procedure for generating a set of time delays, the counter 31 for each read cycle from the memory node 37 generates a signal indicating the end of the current cycle.

Consistently, according to the results of the comparison in the subtractor, 24 estimates of the time delay search for each component of the multipath signal of the current period with all estimates of the updated time delays of the components of the multipath signal of the previous period at the output of the circuit HE 30 generate a comparison signal equal to logical zero if the estimates of the time delays of the component of the multipath signal are apart from a friend no less than Th, and equal to a logical unit otherwise. The generated comparison signals are sequentially recorded in the register 25. From the parallel outputs of the register 25, the comparison signals are fed to the inputs of the OR-NOT 26 circuit, the output signal of which is written to the trigger 27 by the signal indicating the end of the current comparison cycle, and the register 25 is reset.

The output signal of the trigger 27 is supplied to the second control input of the key 23, to the second control input of the key 33 and to the first input of the adder 34. The output signal of the OR-NOT 26 circuit (trigger 27) is equal to a logical unit if the estimate of the time delay of the analyzed component of the signal of the current period is separated from all estimates of the updated time delays of the components of the multipath signal of the previous period by at least Th. Otherwise, the output signal of the OR-NOT 26 circuit (trigger 27) is equal to logical zero.

According to a signal equal to a logical unit, the search estimate of the time delay of the analyzed component of the signal of the current period from the output of the multiplexer 22 through the key 23 is transmitted to the first input of the memory node 37 and stored. The corresponding estimate of the power search (decisive function) of the multipath signal of the current period from the output of the multiplexer 32 through the key 33 is transmitted to the first input of the memory node 42 and stored.

By a signal equal to a logical zero, the keys 23, 33 are blocked, and they are not written to memory nodes.

Upon completion of the procedure for comparing all estimates of the search for time delays of the components of the multipath signal of the current period with all estimates of the updated time delays of the components of the multipath signal of the previous period in the memory node 37, a set of time delays of the components of the multipath signal of the previous period and those search components whose time delays differ from time delays each of the components of the multipath signal of the previous period by more than a predetermined threshold Th. At the same time, in the memory node 42, a set of estimates of the power search (decision function) of the corresponding components of the multipath signal is formed.

For each cycle of the comparison procedure, the output signal of the trigger 27, equal to a logical unit or logical zero, is fed to the first input of the adder 34. The second input of the adder 34, the first input of the comparator 39 and the second input of the control unit 44 receives a signal from the output of the memory unit 38, which corresponds to the number of components of the aggregate of the current period. In the adder 34, for each comparison cycle, the number recorded in the memory node 38 and the signal from the output of the trigger 27 are summed. The result of the summation is transmitted to the first input of the memory node 38, which is stored by the recording signal coming from the fourth output of the control unit 44 to the second input of the memory node 38.

After the completion of the procedure for forming the constituent components of the current stage by the read signal coming from the fourth output of the control unit 44 to the second input of the memory unit 38, the summation result from the output of the memory unit 38 goes to the input of the comparator 39. In the comparator 39, this signal is compared with a given number L, which is fed to the second input of the comparator 39 and to the second input of the key 35.

At the output of the comparator 39, a control signal is generated for the procedure for generating a set of updated time delays, which is transmitted to the first input of the key 35 and to the third input of the control unit 44.

If the signal from the output of the memory node 38 exceeds a predetermined number L, then the output of the comparator 39 produces a signal equal to in this case the logical unit, which is fed to the first input of the key 35 and to the third input of the control unit 44. To the second input of the control unit 44 from the output node memory 38 receives a signal that corresponds to the number of components of the total population. Based on these signals, the control unit 44 generates control signals for the selection unit 43 and the memory unit 42, which respectively come from the fifth output of the control unit 44 to the second input of the selection unit 43 and from the fourth output of the control unit 44 to the second input of the memory unit 42.

The control signals from the fourth and fifth outputs of the control unit 44 contain a clock signal, write and read signals, signals corresponding to the number of components of the total population and a given number L. Using these signals from the memory node 42 to the selection node 43, the set of power search estimates is read ( decisive function) of the corresponding components of the multipath signal and form the addresses of the time delays L of the most powerful components of the multipath signal recorded in the memory node 37. The generated addresses from the output of the node 43 selection supplied to the fourth input control unit 44. Control unit 44 reads from the host memory 37 the updated plurality of time delays of the current period, it is formed by the addresses. An updated set of time delays of the current period from the output of the memory node 37 and the corresponding addresses from the sixth output of the control unit 44 are received on the control unit 11.

If the signal from the output of the memory node 38 does not exceed a predetermined number L, then according to the control command from the third output of the control node 44, the entire set of time delays of the multipath signal component of the current period is read from the memory node 37 to the control unit 11.

The control unit 44 can be implemented in various ways, for example, on modern digital signal processing (DSP) microprocessors - TMS320Cxx, Motorola 56xxx, Intel, etc.

An embodiment of a power estimation unit 40 of the proposed multipath signal receiving device is shown in FIG. 7.

The power rating unit 40 operates as follows.

The first inputs of the adders with a reset of 46, 50 nodes for evaluating the power of components 45-1, 45-M from the second output of the soft decision forming unit 15 receive the in-phase and quadrature components of the correlation responses of the updated multipath signal components (A _ci A _si i = 1 .. .M). The second inputs of the adders with a reset of 46, 50 and 49 nodes for evaluating the power of the components 45-1, 45-M from the control unit 44 receive respectively the signals reset 1 and reset 2. In the adders with reset 46, 50 at time intervals between the reset signals produce coherent accumulation of in-phase and quadrature components of the correlation responses of the updated components of the multipath signal. When generating a signal, reset 1 for adders with reset 46, 50 takes into account the coherence interval of the multipath signal. The generated signals from the outputs of the adders with a reset 46, 50 are received respectively at the inputs of the quadrants 47 and 51, where they generate signals equal to the squares of the in-phase and quadrature components of the correlation responses. These signals are respectively supplied to the first and second input of the adder 48, the output of which forms a sequence of sums of squares of the in-phase and quadrature components of the correlation responses, which is fed to the input of the adder with a reset 49. In adders with a reset 49 nodes for evaluating the power of components 45-1, 45- M produce incoherent accumulation of the sums of squares of the in-phase and quadrature components of the correlation responses and form a set of estimates of the power (decisive function) of the components of the multipath signal of the previous ne of the rhode, which is input to the multiplexer 41. The characteristics of the signals reset 1 for the adders with reset 46, 50 and reset 2 for the adder with reset 49 are selected so as to provide conditions for evaluating the power of the components of the multipath signal, the same as in the search unit 12 This is necessary to obtain adequate estimates, which are used for comparison in the selection node 43.

An embodiment of a selection node 43 of the proposed multipath signal receiving device is shown in FIG.

First, it should be noted that to implement a specific solution, the use of a memory element 55 in the selection node 43 is redundant. However, this approach allows you to get a universal structure that can be used in the proposed device for solving various problems.

The selection node 43 for the total population of N estimates of the search power (decisive function) of the component of the multipath signal generated in the memory node 42, generates addresses of the updated time delays for L ranked most powerful components of the population. Moreover, N> L.

The selection node 43 operates as follows.

The first input of the memory element 55 from the output of the memory node 42 receives the total set of N estimates of the search power (decisive function) component of the multipath signal. According to the recording signal from the control unit 44 to the second input of the element

memory 55, this set is remembered. It should be noted that the memory element 55 has a structure identical to the memory node 42 and 37, that is, the corresponding parameters of the constituent components are recorded at the same addresses.

Further, all operations are performed with the components of the population recorded in the memory element 55.

The selection procedure is carried out in L stages. For each stage, which consists of K cycles, the maximum element of the population of the element of memory 55 is determined and the address corresponding to it is stored in the address register 60. The maximum element of the population of the element of memory 55 is zeroed and the next stage is performed for the changed set of elements. The whole procedure takes NxL cycles.

From the control unit 44, a clock signal, a read signal, and addresses of the constituent components are received at the second input of the memory element 55. The first input of the key 57 receives the addresses of the components of the population. At the first input of the counter 61, at the second input of the registers 56, 58 and at the first input of the register 52, a clock signal is received. The second input of the counter 61 receives a signal corresponding to the number of components (cycles) N recorded in the memory node 42 of the aggregate. The second input of the counter 62 receives a signal corresponding to a given number (steps) L.

At the first stage, the register 52 is pre-zeroed by a signal that is supplied to its second input from the control unit 44.

Each subsequent stage is implemented as follows.

From the control node 44 to the second input of the memory element 55 and the first input of the key 57 sequentially received addresses of the components of the population. The read signal from the memory element 55 to the first input of the register 56 sequentially reads the estimates of power (decisive function) of the components of the multipath signal. According to the recording signal received at the second input of the register 56 from the control unit 44, the read power (decision function) estimates of the component of the multipath signal are recorded.

According to the read signal, an estimate of the power (decisive function) of the analyzed component from the output of the register 56 goes to the first input of the key 53 and to the second input of the comparator 54, the first input of which from the output of the register 52 receives a zero signal for the first cycle or some intermediate largest current estimate of the component power for other cycles. If the signal at the second input of the comparator 54 is larger than the signal at its first input, then at the output of the comparator 54 a signal is generated equal to the logical unit, which is fed to the second input of the key 53, to the second input of the key 57, to the third inputs of the register 52 and 58 .

According to this signal, an estimate of the power (decisive function) of the component of the multipath signal from the output of the register 56 through the key 53 is fed to the fourth input of the register 52, which is recorded in it instead of the previously recorded lower power rating.

The corresponding address through the key 57 is fed to the first input of the register 58, which is recorded in it instead of the previously recorded address and transmitted to the first input of the key 59.

If the signal at the second input of the comparator 54 is less than the signal at its first input, then a signal equal to logical zero is generated at the output of the comparator 54, and the keys 53, 57 are not switched. Registers 52 and 58 remain in the state corresponding to the previous cycle.

Thus, on each cycle, the address corresponding to the larger component of the changed set of elements is selected.

The counter 61, according to the number of clock symbols at the first input and the signal corresponding to the number of components of the set N, generates a signal for the end of the current cycle and can be implemented, for example, on the basis of a standard programmable counter, for example, 564IE10 series, etc.

The signal for the end of the current cycle is, for example, the moment of occurrence of a single signal in the high order of the counter 61. This signal from the output of the counter 61 is fed to the first input of the counter 62, to the second input of the key 59, to the second input of the register 52 and to the first input of the key 63, the second input of which receives a signal equal to zero. The output signal of the key 63 is supplied to the first input of the memory element 55.

At the end of the current cycle, the register 52 is reset and the output of the key 63 sends a zero level signal to the first input of the memory element 55.

The address corresponding to the constituent component of each stage with the maximum power estimate is supplied through the key 59 to the first input of the address register 60 and to the second input of the memory element 55.

According to the write signal from the control unit 44 to the second input of the address register 60, the address corresponding to the component of each stage of the population with the maximum power rating is stored in the address register 60.

Thus, at each stage, the address corresponding to the larger element of the changed population is selected.

In the memory element 55 at the address corresponding to the component of the population with the maximum power rating, a zero value is recorded.

The counter 62, based on the number of signals that end the current cycle at the first input and the signal corresponding to the number of steps L, generates a signal to end the selection procedure and can be implemented, for example, on the basis of a standard programmable counter of the 564IE10 series, etc.

The signal of the end of the selection procedure is, for example, a signal of a logical unit at the high order of the counter 62, which occurs when the number of signals of the end of the current cycle coincides with the signal corresponding to the number of steps L. This signal from the output of the counter 62 is sent to the control unit 44. The signal the end of the selection procedure in the control unit 44 form a read signal, which is fed to the second input of the address register 60. By this signal from the address register 60 to the control unit 44 read the addresses L of the most powerful Ponent updated set of time delays.

The address register 60 with serial input and parallel output can be implemented, for example, on the basis of standard microcircuits of the 561PR1 series or 1564IR8 series.

The proposed method of searching for a signal in the field of multipath is as follows:

- Form the complex correlation responses of the pilot signal by determining the correlation of the input signal with the reference signal corresponding to the given discrete time positions (delays) in the multipath region (shifted by the search time discrete within the multipath region).

- Determine the values of the decisive function for the given discrete time delays of the multipath region by summing the squares of the in-phase and quadrature parts of the corresponding complex correlation responses of the pilot signal.

- Compare the values of the generated decision function with a given threshold h.

- Form the totality of the values of the decisive function that have exceeded the threshold, and the totality of the corresponding time delays.

- Gradually form estimates of the search for time delays of the components of the multipath signal. At the same time, at each stage, one component is evaluated, for which:

- determine the element with the maximum value from the set of values of the decisive function of the current stage, while the set of values of the decisive function of the first stage is the formed set of values of the decisive function,

- compare the maximum value of the decisive function of the current stage with a threshold h,

- in the case of not exceeding the threshold, the formation of an estimate of the search for time delays of the components of the multipath signal is stopped,

- in case of exceeding the threshold:

the time delay corresponding to the element with the maximum value of the decisive function of the current stage is considered an estimate of the search for the time delay of the components of the multipath signal,

correct the elements of the set of values of the decisive function of the current stage,

form a set of values of the decisive function of the subsequent stage, excluding from the adjusted set of values of the decisive function of the current stage the element with the maximum value.

The search time discrete is selected, for example, to be equal to half the memory bandwidth chip.

The threshold h is selected, for example, proportional to the noise power.

Correction of elements of the set of values of the decisive function of the current stage is performed, for example, by subtracting the correction value from the elements of the set, the time delays of which differ from the time delay of the maximum element by no more than a given time threshold η.

The correction value is selected, for example, proportional to the maximum element and inversely proportional to the module of the difference in time delays of the maximum and the corrected element of the set of values of the decisive function of the current stage.

The time threshold η is selected, for example, equal to two discrete search time.

To implement the proposed method of searching for a signal in the multipath domain, the device shown in Fig. 9 is used, where it is indicated:

16-1 - 16-K - quadrature correlators,

17 - generator PSP,

18 is a diagram of the formation of the threshold,

19 is a control circuit,

20 - multiplexer,

64 is a beam detection circuit,

65, 73, 76 - the first, second, third keys,

66, 83 - the first and second comparators,

67 - logical node

68 - counter

69 - control unit,

70 is an OR diagram

71 - a memory element of the original set of decisive functions,

72 - node determining the maximum element,

74 - memory register of critical search functions,

75 - memory element of the original set of time delays,

77 - memory register estimates search time delays,

78, 82 - the first and second subtractors,

79 - shaper module

80 - register

81 - shaper correction signal.

A device that implements the search method contains K quadrature correlators 16-1 to 16-K, the first and second inputs of which are combined with the first and second inputs of the threshold formation circuit 18 and are signal inputs of the device. The third input of each quadrature correlator 16-1 - 16-K is the input of the SRP signal and connected to the corresponding output of the SRP 17 generator.

The fourth input of each quadrature correlator 16-1 - 16-K, which is the input of the control signal, is connected to its corresponding first output of the control circuit 19. The second output of the control circuit 19, which is the output of time shifts of the PSP generator 17, is connected to the input of the PSP generator 17 and the second input of the ray detection circuit 64. The output of each quadrature correlator 16-1 - 16-K, which is the output of the values of the decisive function for the given time delays of the multipath region, is connected to the corresponding input multip lexor 20.

The output of the multiplexer 20, which is the output of the values of the decisive function from the multipath region, is connected to the first input of the ray detection circuit 64.

The output of the threshold formation circuit 18, which is the output of the threshold value h, is connected to the third input of the ray detection circuit 64 and the first input of the first comparator 66.

The input of the control circuit 19, which is the input of the control signals, is connected to the first output of the control unit 11.

The first output of the ray detection circuit 64, which is the time delay output, is connected to the second output of the memory element of the original set of time delays 75. The second output of the ray detection circuit 64, which is the output of decision function values that exceed the threshold, is connected to the second input of the memory element of the original set decision functions 71. The third output of the ray detection circuit 64, which is the output of a signal equal to the number of detected components that have exceeded the threshold, is connected to the first input of the control unit 69 The fourth output of the ray detection circuit 64, which is the output of the threshold signal for a given discrete time delay, is connected to the second input of the control unit 69. The first output of the control unit 69, which is the output of the write signal, is connected to the first input of the memory element of the initial decision set functions 71 and with the first input of the memory element of the original set of time delays 75. The output of the memory element of the original set of decisive functions 71, which is the output of the decisive function of the current of the first stage is connected to the first input of the node for determining the maximum element 72, with the first input of the subtractor 82, with the first inputs of the first key 65 and the second key 73. The second input of the node for determining the maximum element 72 is a control input and is connected to the fifth output of the control node 69. Output node determining the maximum element 72, which is the output address of the selected maximum element, connected to the fifth input of the control node 69.

The second output of the control unit 69, which is the output of the signal indicating the presence of a maximum, is connected to the second input of the key 65 and the first input of the logical node 67. The output of the key 65 is connected to the second input of the first comparator 66. The output of the comparator 66, which is the output of exceeding the threshold, is connected to the second input of the logical node 67, the first input of the counter 68, with the second controlled inputs of the second 73 and third 76 keys and with the first input of the register 80.

The third output of the control node 69, which is the output of the number of detected components of the current search period, is connected to the second input of the counter 68. The output of the counter 68 is connected to the first input of the OR circuit 70, the second input of which is connected to the output of the logical node 67. The output of the OR circuit 70, which is the output of the indicator of the end of the search procedure, connected to the third input of the control unit 69.

The fourth output of the control unit 69, which is the output of the control signal for generating a search estimate, is connected to the first input of the memory register of the search decision functions 74, with the first input of the register of the memory of the search for time delays 77 and with the third input of the shaper of the correction signal 81. The second input of the decision memory register search functions 74 and the first input of the correction signal generator 81 are combined and connected to the output of the second key 73, which is the output of the maximum decisive function of the current stage. The second inputs of the memory register search evaluation time delays 77 and the register 80 are combined and connected to the output of the third key 76, the first input of which is combined with the first input of the first subtractor 78 and connected to the memory element of the original set of time delays 75, which is the output of the time delay value, corresponding to the maximum element. The output of the register 80, which is the output of the time delay values, is connected to the second input of the first subtractor 78. The output of the first subtractor 78, which is the output of the time delay difference, is connected to the input of the shaper of module 79. The output of the difference signal module of the shaper of module 79 is connected to the second input of the shaper correction signal 81 and the first input of the second comparator 83. The second input of the comparator 83 is a threshold input. The output of the second comparator 83 is the threshold exceeding output and is connected to the fourth input of the control unit 69 and to the fourth input of the correction signal generator 81. The output of the correction signal generator 81, which is the output of the correction signal, is connected to the second input of the second subtractor 82, the output of which is connected to the second input memory element of the initial set of decision functions 71. The output of the memory register of the evaluation of the search for time delays and the output of the memory register of the decision functions of the search are combined and are the output of the search block 12.

The proposed device signal search 12 operates as follows.

From the totality of the values of the decisive search function that exceeded the threshold, a set of estimates of the search for the values of the decisive function of the components of the multipath signal and the set of time delays corresponding to them are formed.

When forming an estimate of the search for temporary positions (delays) of the component of a multipath signal, the influence of lateral emissions of the correlation responses of ray signals on each other is taken into account.

The in-phase and quadrature components of the input complex signal are respectively supplied to the first and second inputs K of the parallel quadrature correlators 16-1 - 16-K and to the first and second inputs of the threshold formation circuit 18. To the third input of the quadrature correlators 16-1 - 16-K from the corresponding the outputs of the generator PSP 17 receives a reference signal shifted by the discrete search time within the multipath region, for example, equal to half the chip PSP. The time shifts of the reference signals of the PSP generator 17 are set by the time shift signal of the PSP generator, which is supplied from the second output of the control circuit 19 to the input of the PSP generator 17 and to the second input of the ray detection circuit 64. To the fourth input of each quadrature correlator 16-1 - 16-K s the corresponding first output of the control circuit 19 receives a control signal, for example, a reset signal.

In each quadrature correlator 16-1 - 16-K, the input signal is multiplied by a reference signal, the multiplication results are accumulated and the correlation responses of the in-phase and quadrature components of the pilot signal are generated within the multipath region, the squares of the in-phase and quadrature parts of the corresponding complex correlation responses of the pilot signal are formed. Summing the squares of the in-phase and quadrature parts of the corresponding complex correlation responses, determine the values of the decisive function in the given discrete time positions of the multipath region, which are input to the multiplexer 20.

From the output of the multiplexer 20, the values of the decisive function in the given discrete time positions of the multipath region are supplied to the first input of the ray detection circuit 64.

According to the control signal from the output of the control unit 11, the control circuit 19 performs a time shift of the PSP generator 17 within the multipath region and resets the adders of the quadrature correlators 16-1 - 16-K.

According to the input signal, the threshold formation circuit 18 forms a threshold h, which is selected, for example, in proportion to the noise power estimate and is supplied to the third input of the ray detection circuit 64 and to the first input of the comparator 66.

In the ray detection circuit 64, the values of the decisive function for the given discrete time delays of the multipath region are compared with the threshold h and the values of the decisive function that exceed the threshold h and the corresponding time positions are generated, the signal corresponding to the number of detected components N, which exceeded the threshold, and an indication signal for exceeding threshold h at a given discrete time position.

Values of the decisive function that exceed the threshold h and the corresponding time positions from the second and first output of the ray detection circuit 64 are respectively received at the second input of the memory element of the original set of decision functions 71 and at the second input of the memory element of the original set of time delays 75. At the first input of the node control 69 from the third output of the ray detection circuit 64 and the second input of the control node 69 from the fourth output of the beam detection circuit 64 respectively receive a signal equal to the number of detected components N, exceeds the threshold, and indication signal exceeding a predetermined threshold h discrete time positions.

The signal indicating the excess of the threshold h at the first output of the control unit 69 generates a write signal that is transmitted to the first input of the memory element of the original set of decision functions 71 and to the first input of the memory element of the original set of time delays 75. By this signal, the values of the decision function exceeding the threshold h , and the corresponding temporary positions are recorded respectively in the memory elements of the original set of decision functions 71 and the memory element of the original set of time delays 75 and form the original ovokupnost decision function values that exceed the threshold h, and an initial set of corresponding time position (latency) of the current search period.

Using the initial sets, step-by-step estimates for the search of time delays of the components of the multipath signal are generated for each search period. Moreover, at each stage, one component is evaluated.

The initial set of values of the decisive function of the first stage is the formed set of values of the decisive function.

For each current stage, by a command to read from the control unit 69, from the output of the memory element of the initial set of decision functions 71, the initial set of decision functions is element-wise read and fed to the first input of the node for determining the maximum element 72, to the first input of the subtractor 82, to the first input of the key 73, and to first key input 65.

In the node for determining the maximum element 72 by the control signals received at its second input from the fifth output of the control unit 69, an element of the set of values of the current stage is selected that has the maximum value of the decisive function. As the node for determining the maximum element 72, for example, the selection node 43 described above can be used. To determine the maximum element for the node 72 in the control node 69, the control signals are generated the same as for the selection node 43 of the update unit of the signal components 21, except for the signal corresponding to the number of the largest elements. In this case, the number of the largest elements in the set of decision functions is equal to unity (L = 1).

With such an implementation, in the node for determining the maximum element 72, the address of the constellation element is formed in the memory element of the initial set of decision functions 71, which has a maximum value of the decision function. The address of the selected maximum element from the output of the node for determining the maximum element 72 goes to the fifth input of the control node 69, which is transmitted from its first output to the first inputs of the memory element of the initial set of decision functions 71 and to the first inputs of the memory element of the original set of time delays 75.

From the memory cell of the element of the initial set of decision functions 71 with the address of the value of the maximum decision function to the first input of the key 73 and to the first input of the key 65, the value of the maximum decision function of the current stage is read, and from the memory cell with the corresponding address of the memory element of the original set of time delays 75 to the first the input of the subtractor 78 and the first input of the key 76 read the temporary position corresponding to the value of the maximum decisive function.

The second input of the key 65 and the first input of the logical node 67 from the second output of the control unit 69 receives a signal indicating the presence of a maximum equal to, for example, a logical unit. According to this signal, the maximum value of the decisive function of the current stage from the first input of the key 65 is supplied to the second input of the comparator 66, in which it is compared with the threshold h.

According to the result of the comparison, a control signal (exceeding the threshold h) is formed in the comparator 66. This signal from the output of the comparator 66 is fed to the first input of the counter 68, to the second input of the logical node 67, to the second controlled inputs of the key 73 and 76 and to the first input of the register 80. The signal comes to the second input of the counter 68 from the third output of the control unit 69 corresponding to the number of detected components N of the multipath signal of the current search period.

If the maximum element of the totality of the current stage does not exceed the threshold h, then in the comparator 66 generate a control signal equal to, for example, a logical zero. At this signal, the keys 73 and 76 are closed, the register 80 is reset. The control signal from the comparator 66 and the signal indicating the presence of a maximum from the second output of the control unit 69 at the output of the logical node 67 generate a signal equal to the logical unit, which is fed to the second input of the OR circuit 70, the first input of which receives the signal from the output of the counter 68. In this case, regardless of the signal from the output of the counter 68, the output signal of the OR circuit 70 is equal to a logical unit and is fed to the third input of the control unit 69, through which control signals are generated in it. The generated control signals are supplied from the fourth output of the control unit 69 to the first inputs of the memory register of the deciding search functions 74 and to the first inputs of the memory register of the evaluation of the search for time delays 77 and from the first output of the control unit 69 to the first inputs of the memory element of the initial set of decision functions 71 and the memory element initial set of time delays 75. Based on these signals, the formation of an estimate of the search for time delays of the multipath signal components is stopped and the contents of the memory registers 74, 77 are read into the block the formation of soft decisions 15 and into the block updating the signal components 21.

If the maximum element of the totality of the current stage exceeds the threshold h, then in the comparator 66 generate a control signal equal to a logical unit. According to this signal, the maximum element of the original population through key 73 enters the second input of the memory register of the deciding search functions 74 and the first input of the shaper of the correction signal 81, and the value of the corresponding time delay enters through the key 76 to the second input of the memory register of the time delays of the search estimate 77 and the second input of the register 80, in which are stored. The value of the time delay corresponding to the maximum element is read from the output of the register 80 to the second input of the subtractor 78.

The time position (delay) corresponding to the element with the maximum value of the decisive function is considered an estimate of the search for the time delay of the multipath signal component. For the next stage, the elements of the set of values of the decisive function of the current stage are corrected.

The output of the logical node 67 by the signal from the comparator 66 (equal to a logical unit) and the signal indicating the presence of a maximum from the control unit 69 generate a signal equal to the logical zero, which is fed to the second input of the OR circuit 70, the first input of which receives the signal from the output of the counter 68 .

The counter 68 summarizes the number of excesses of the threshold h in the comparator 66 and, when the accumulated total value equal to the number of detected components N is accumulated, it generates a signal equal to a logical unit that corresponds to viewing all the elements of the original set of the current period. Otherwise, the signal at the output of counter 68 is logic zero. By a signal equal to a logical unit, the formation of an estimate of the search for time delays of the multipath signal components is stopped and the contents of the memory registers 74, 77 are read into the soft decision generation unit 15 and the signal component update unit 21.

According to the control signal from the first output of the control unit 69, the values of the decisive function are read out from the memory element of the original set of decisive functions 71 to the first input of the subtractor 82 and the corresponding time delays from the memory element of the original set of time delays 75 to the first input of the subtractor 78.

At the output of the subtractor 78, a difference is formed between the temporary positions of each element of the totality of the current stage and the temporary position of the maximum element. The difference signal for each element of the totality of the current stage is input to the shaper of module 79. From the output of the shaper 79, the difference signal module goes to the first input of the second comparator 83 and to the second input of the shaper of the correction signal 81. The time threshold η is applied to the second input of the comparator 83, which is selected , for example, equal to two discrete search times. If the signal module of the difference is not greater than the threshold η, then a control signal is generated at the output of the comparator 83, for example, a signal equal to a logical unit that is supplied to the fourth input of the control unit 69 and to the fourth input of the correction signal generator 81. This signal is generated in the node 69 a read signal and a write signal, which respectively come from the fourth output of the node 69 to the third input of the driver of the correction signal 81 and from the first output of the node 69 to the first input of the memory element of the original set of decision functions 71.

Using the control signal and the read signal in the shaper of the correction signal 81, a correction value (correction signal) is generated.

The correction value is selected, for example, in proportion to the value of the maximum element and inversely to the modulus of the difference in the time positions of the maximum and the adjusted element of the set of values of the decisive function of the current stage.

Correction of elements of the set of values of the decisive function of the current stage is performed, for example, by subtracting the correction value from the corresponding element of the set.

The correction value (correction signal) from the output of the shaper of the correction signal 81 is fed to the second input of the subtractor 82, in which the correction value of the decision function of the element of the totality of the current stage is carried out. The adjusted value goes to the second input of the memory element of the original set of decision functions 71.

According to the recording signal, the adjusted value of the decisive function of the element of the population is recorded in the memory element of the original set of decisive functions 71 at the address of the corrected element of the population.

If the modulus of the difference signal is greater than the threshold η, then the correction of the corresponding element of the set of values of the decisive function of the current stage is not performed.

At the end of the analysis of all elements of the set of values of the decisive function of the current stage, the value of the maximum element in the memory element of the original set of decisive functions 71 is reset.

Thus, at each current stage, a set of values of the decisive function and the corresponding time delays are formed for analysis at a subsequent stage.

An embodiment of the ray detection circuit 64 is shown in FIG. 10.

The ray detection circuit operates as follows.

The comparator 84 compares the values of the decisive function in the given discrete time positions of the multipath region, coming from the output of the multiplexer 20, with the threshold value h coming from the threshold 18 circuit, and the signal for exceeding the threshold h at a given discrete time position is generated. This signal is supplied to the control unit 69 and is used as a write signal. When the threshold is exceeded, the values of the decisive function through the key 86 go to the second input of the memory element of the original set of decisive functions 71, and the corresponding time positions (delays) from the control circuit 19 through the key 87 go to the second input of the memory element of the original set of time delays 75.

In the adder 85, the number of facts of exceeding the threshold h is accumulated and a signal corresponding to the number of detected components N is generated, which is supplied to the first input of the control unit 69.

An embodiment of the correction signal driver 85 is shown in FIG. 11.

The proposed driver signal correction works as follows.

By the control signal from the control unit 69, the value of the decisive function of the maximum element of the population of the current stage is recorded from the output of the key 73 in the register 88, and the signals of the time difference module of the maximum element and the analyzed element of the population are sequentially written from the output of the driver of the module 79 in the register 92. These signals are read from the outputs of the registers 88, 92 to the inputs of the shaper of the relation 89, in which the relations of the values of the decisive function of the maximum element of the population are sequentially generated a phase difference to the signal module maximum time delay element and the element being analyzed together. The signal ratio is scaled in the multiplier 90 by multiplying by the coefficient α, selected, for example, according to the results of computer simulations. According to the signal from the comparator 83, which is equal to a logical unit for the elements of the population, the temporary positions of which differ from the temporary position of the maximum element by no more than a predetermined time threshold η, the output signal of the multiplier 90 through the key 91 is fed to the subtractor 82.

An embodiment of the logical node 67 is presented in Fig.12.

The standard logical element AND 93 and the element NOT 94 are included in the logical node 67 in such a way that the signals from the control unit 69 and the comparator 66 generate a control signal to terminate (logical unit) or continue (logical zero) of the procedure for forming an estimate of the search for time delays of components multipath signal.

The memory registers 74, 77 can be implemented on the basis of a register with serial input and parallel output, for example, on standard microcircuits of the 561PR1 series or 1564IR8 series.

Counter 68 can be implemented, for example, on the basis of a standard programmable counter series 564IE10 and others.

The control node 69 can be implemented in various ways, for example, on modern digital signal processing (DSP) microprocessors - TMS320Cxx, Motorola 56xxx, Intel, etc.

The proposed method for forming soft decisions when receiving a multipath signal is as follows.

- Carry out the selection of the components of the multipath signal, for which:

- evaluate the power of the components of the multipath signal;

- perform the ranking of the components of the multipath signal in decreasing order of their power;

- sequentially for all ranked components of the multipath signal, except the first;

form the ratio of the power of the component to the sum of the power of all previous ranked components;

comparing the generated ratio with a given threshold ν;

if the threshold is exceeded, the component is selected to form soft decisions;

in case of not exceeding the threshold, the component selection procedure is completed.

- Form a sequence of correlation responses of information symbols for each selected component of the multipath signal.

- Carry out a weighted summation of the correlation responses of the information symbols of all the selected components of the multipath signal, obtaining the united soft decisions about information symbols.

The threshold ν is chosen, for example, according to permissible energy losses due to the limitation of the components used in demodulation, for example, for losses of 0.1 dB ν = 0.03.

Correlation responses of information symbols for each selected component of a multipath signal are generated, for example, by determining the correlation of the input signal with the known SRP at symbol duration intervals.

The weighting coefficients when summarizing the correlation responses of the information symbols of the selected components of the multipath signal are selected, for example, as complex conjugate estimates of the complex envelope of information symbols of the components of the multipath signal.

To implement the method of forming soft solutions, the device shown in Fig.13, where indicated:

13 is a multi-channel data receiver,

14 is a scheme for combining soft decisions,

95 - memory register,

96, 103, 109 - the first, second, third multiplexers,

97, 102 - the first and second comparators,

98 is an OR circuit,

99 - counter

100 - circuit NOT

101 is the key

104, 106, 110 - the first, second and third memory node,

105 - adder

107 - shaper relationships,

108 - power rating unit,

111 - ranking node,

112 - control unit.

The soft decision forming device comprises a multi-channel data receiver 13, the first and second inputs of which are the inputs of the in-phase and quadrature components of the input signal components and the first and second inputs of the device. The third input of the device is the inputs of time delays and the first inputs of the second multiplexer 103. The fourth input of the device is the input of the values of the deciding function of the search for the current stage and the first input of the first multiplexer 96. The output of the multi-channel data receiver 13 is connected to the input of the soft decision combining circuit 14, which is the input to the sequence correlation responses of information and pilot symbols of a component of a multipath signal, and with the input of the power estimation unit 108, which is the input of nosti correlation responses of the pilot symbols of the multipath signal components. In addition, the output of the multi-channel data receiver 13 is the output of the sequence of correlation responses of the pilot symbols and time delays of the multipath component and the second output of the device. The first output of the device is the output of the combined soft decisions of the soft decision combining scheme 14. The second inputs of the first multiplexer 96 and the second multiplexer 103 are control and connected to the first output of the control unit 112. The output of the power estimation unit 108, which is the output of the power estimation of the multipath component (decisive search function), connected to the first input of the third multiplexer 109. The output of the first multiplexer 96 is combined with the output of the third multiplexer 109, which is the output of the signal power of the component, and is connected to the first input of the third memory node 110. The output of the second multiplexer 103, which is the time delay output of the updated multipath component, is connected to the first input of the first memory node 104, the second input of which is combined with the second input of the third multiplexer 109 and connected with the second control output of the control unit 112. The third output of the control unit 112 is the output of the control signals and is connected to the second input of the third memory node 110, the first input of the first comparator 97 and ne the first input of the memory register 95. The output of the third memory node 110, which is the output of estimating the power of the current component, is connected to the first input of the ranking unit 111, the first input of the adder 105 and the first input of the ratio generator 107, in which the power relations of the current component to the sum of the powers of all are formed previous ranked components. The output of the ranking unit 111, which is the output of the ordered addresses of the memory cells, is connected to the first input of the control unit 112. The second input of the ranking unit 111 is connected to the fourth control output of the control unit 112. The fifth output of the control unit 112, which is the output of the write signal, is connected to the second the input of the second memory node 106, the first input of which is connected to the output of the adder 105. The output of the second memory node 106, which is the output of the total powers of all the previous ranked components, is connected to the second input the adder 105 and the second input of the ratio shaper 107. The output of the power ratio shaper 107 is connected to the first input of the second comparator 102, the second input of which is the input of the threshold value ν. The output of the second comparator 102, which is the output of a comparison signal of the generated power ratio of the current component to the sum of the powers of all previous ranked components with a given threshold ν, is connected to the input of the circuit NOT 100, to the input of the counter 99 and to the first input of the key 101. The second input of the key 101 is connected with the output of the first memory node 104. The output of the key 101, which is the time delay output of the multipath component, is connected to the second input of the memory register 95. The output of the memory register 95, which is the output of the scoop time delays of the selected components is connected to the third input of the multi-channel data receiver 13. The output of the circuit NOT 100 is an inverse output and is connected to the first input of the OR circuit 98. The output of the counter 99, which is the output of the signal equal to the number of excesses of the threshold ν, is connected to the second input the first comparator 97. The output of the first comparator 97 is connected to the second input of the OR circuit 98, the output of which is the output of the signal indicating the completion of the selection procedure for components to form soft decisions and connected to the second input th node control 112.

A device for forming soft solutions works as follows.

In the proposed device for the formation of soft decisions (demodulation) of the information signal of the current period, the components of the multipath signal are preliminarily selected. The selection procedure is performed from the ranked in decreasing order of the population of the multipath signal components found in the search unit 12 and the components processed in the soft decision combining scheme 14. The selection procedure is carried out for each search period.

As a selection criterion, the components of the multipath signal that are used in the formation of soft solutions take the fact that some predetermined threshold value is exceeded by a value equal to the ratio of the power of the component under study to the total power of all previous components, ranked by decrease in their power.

The first and second input of the multi-channel data receiver 13 receives respectively the in-phase and quadrature component of the input multipath signal component. From the outputs of the control unit 11, the first inputs of the multiplexer 103 receive time delays of the updated components of the multipath signal of the current period.

From the first output of the search unit 12 to the first input of the multiplexer 96 receives the values of the decisive search function of the current period, equivalent to the power of the respective components. From the output of the multi-channel data receiver 13, the sequence of correlation responses of the pilot symbols of the updated multipath components of the signal arrives at the input of the node for estimating the power 108, the sequences of the correlation responses of information and pilot symbols of the updated components of the multipath signal are received at the input of the soft decision circuit 14, the sequence of corrections is received from the signal update block 21 correlation responses of pilot symbols and time delays of the specified multipath signal components.

The multichannel data receiver 13 of the proposed device can be implemented similarly to the multichannel data receiver of the prototype. The time delay of the specified components of the multipath signal in the prototype device is formed in the tracking schemes for the delay of the data receivers, the output signal of which can be used in the proposed device.

At the power estimation node 108, according to the correlation responses of the pilot symbols of the updated multipath components, a power estimate (a decisive search function) of the refined components of the multipath signal is generated. The generated component power estimation signal is supplied from the output of the power estimator 108 to the first input of the multiplexer 109. For example, the power estimator 108 can be used, for example, the power estimator 40 described earlier.

According to the control signals received from the first output of the control unit 112 to the second inputs of the multiplexers 96 and 103, the input signals of the multiplexers are element-wise sent respectively to the first input of the memory unit 110 and to the first input of the memory node 104.

According to the recording signal from the second output of the control unit 112 to the second input of the memory node 104, the time delays of the updated multipath components of the current period and the evaluation of the search for time delays of the components of the multipath signal from the output of the multiplexer 103 are recorded in the memory.

According to the control signal from the second output of the control unit 112 to the second input of the multiplexer 109, the signal for evaluating the power of the specified components through the multiplexer 109 is element-wise transmitted to the first input of the memory unit 110.

According to the write signal from the third output of the control unit 112, which is supplied to the second input of the memory unit 110, the power estimates of the component are recorded in memory.

Thus, the time delay of the updated components of the multipath signal of the current period is stored in the memory node 104, and the corresponding values of the power estimates of the components of the multipath signal of the current period are stored in the memory node 110. In this case, the time delays and the corresponding values of the component power estimates are stored in the memory cells of the nodes 104 and 110 at the same addresses.

The read signal from the third output of the control unit 112 to the second input of the memory unit 110, reads the power estimates of the updated components of the multipath signal of the current period, which from the output of the memory unit 110 go to the first input of the ranking unit 111.

In the ranking node 111 according to the control commands received from the fourth output of the control node 112 to the second input of the node 111, the addresses of the memory cells are ordered so that the estimates of the power of the components placed in them are arranged in descending order. The ordered addresses of the memory cells come from the output of the ranking node 111 to the first input of the control node 112.

As the ranking node 111, for example, the selection node 43 described above can be used. Moreover, in the control unit 112, for the ranking unit 110, control signals are generated that are the same as for the selection unit 43 of the update unit of the signal components 21, with the exception of the signal corresponding to the number of largest elements. In this case, the number of the largest elements of the population is N (i.e., L = N).

From the memory cells of the node 110 located at the addresses that come from the third output of the control node 112 to the second input of the memory node 110, the power estimates of the components of the multipath signal are read elementwise in descending order. The read estimates of the power of the components from the output of the memory node 110 go to the first input of the adder 105 and to the first input of the ratio former 107. At the same time, from the memory cells of the memory node 104 located at the addresses that come from the second output of the control node 112 to the second input of the memory node 104, the corresponding time delays of the components of the multipath signal, which are output from the output of the memory node 104 to the second input of the key 101, are read elementwise.

The read signal, which comes from the fifth output of the control unit 112 to the second input of the memory node 106, sends a signal to the second input of the adder 105 and the second input of the ratio former 107 from the output of the memory node 106 equal to the sum of the powers of all previous ranked components of the population.

In the adder 105, the power of the current component is summed with the total power of all the previous ranked components of the population from the memory node 106. The result of the summation from the output of the adder 105 is sent to the first input of the memory node 106. By the write signal from the fifth output of the control unit 112, which is received at the second input of the node memory 106, the summation result is stored.

Thus, in the adder 105 and the memory node 106 carry out the elementwise accumulation (summation) of the values of the estimates of the power of the components of the multipath signal. In this case, for each analysis step, the current sum of the power estimates of the ranked components of the multipath signal is received at the second input of the ratio shaper 107 from the output of the memory node 106, and the value of the power of the current component is received at the first input of the ratio shaper 107 from the output of the memory node 110.

For each step of the analysis, a ratio of the power of the current component a to the sum of the power of all previous ranked components b is formed in the ratio generator 107.

The received signal from the output of the shaper of the ratio 107 enters the first input of the comparator 102, the second input of which receives the specified threshold signal ν.

The threshold ν is chosen, for example, according to the permissible energy loss due to the limitation of the components used in demodulation, for example, for losses of 0.1 dB, ν = 0.03.

The comparator 102 compares the generated ratio of the power of the current component to the sum of the powers of all previous ranked components with a given threshold ν, and the comparison result is fed to the input of the circuit NOT 100, to the input of the counter 99 and to the first input of the key 101.

The inverse signal from the output of the circuit NOT 100 goes to the first input of the circuit OR 98.

If the threshold is exceeded, a control signal is generated in the comparator 102, for example, in the form of a logical unit. According to this signal, the corresponding time delay of the multipath component through the key 101 is supplied to the second input of the memory register 95. The recording signal, which is received from the third output of the control unit 112 to the first input of the memory register 95, records the time delay of this component of the multipath signal in the memory register 95.

At the output of the counter 99, for each analysis step, a signal is generated equal to the number of threshold exceeded ν, which is fed to the second input of the comparator 97. A signal equal to the number of components recorded in the memory node 110 is received at the first input of the comparator 97 from the third output of the control unit 112. the results of the comparison at the output of the comparator 97 form a signal equal to a logical unit if the signal equal to the number of threshold exceeded ν is not less than the number of components recorded in the memory node 110 and equal to logical zero otherwise. This signal from the output of the comparator 97 is fed to the second input of the OR circuit 98, at the output of which a signal is generated indicating the completion of the selection procedure for components to form soft decisions. This signal is equal to a logical unit in the case of not exceeding the threshold ν or when the number of exceeding the threshold ν is not less than the number of components recorded in the memory node 110, and comes from the output of the OR circuit 98 to the second input of the control unit 112.

The signal indicating the completion of the selection procedure for the components to form soft decisions in the control unit 112 generates a read signal, which from the third output of the node 112 is supplied to the first input of the memory register 95. The signal from the memory register 95 reads the set of time delays of the selected components, which is fed to the third input of a multi-channel data receiver 13.

In the scheme for combining soft decisions 14 according to the correlation responses of the selected components of the information symbols of the multipath signal of the current period, weighted summation is performed, obtaining the combined soft decisions about the information symbols of the selected components of the multipath signal that are output to the device. The weights for a weighted summation of the correlation responses of information symbols of the refined components of the multipath signal are selected, for example, as complex conjugate estimates of the complex envelope of the information symbols of the refined components of the multipath signal, which are sent to the soft decision combining unit from the output of the multi-channel data receiver 13.

The memory register 95 can be implemented on the basis of a register with serial input and parallel output, for example, on standard microcircuits of the 561PR1 series or 1564IR8 series.

The control unit 112 can be implemented in various ways, for example, on modern digital signal processing (DSP) microprocessors - TMS320Cxx, Motorola 56xxx, Intel, etc.

Claims (16)

1. The method of receiving a multipath signal, which consists in the fact that when receiving periodically determine the number and time delays of the components of the multipath signal, for which they determine the time domain of multipath, search for the signal in the multipath region and determine the search for the number and time delays of the multipath signal, form updated number and time delays of the multipath components, find the time delays of the multipath components of the current period, constantly updating e time delays of the multipath component, using these time delays, form soft decisions about information symbols, characterized in that the generation of updated numbers and time delays of the multipath signal components is carried out from the multipath signal components of the previous period and those estimates of the time delay search of the multipath signal components, which differ from the time delays of each component of the multipath signal of the previous period by more than a predetermined value of Th, and e and the number of components of the formed population does not exceed a given number L, then the updated number of components is equal to the number of components of the formed population, and the updated time delays of the components of the multipath signal are equal to the time delays of the components of the population, if the number of components of the formed population is greater than the specified number L, then the power of the components of the population is estimated and choose the L most powerful components whose time delays are updated component time delays. a blank signal, in which case the updated number of components is L. 2. The method according to claim 1, characterized in that the time domain of multipath is defined as the time domain of delays, the beginning of which is t1 less than the minimum time delay of the components of the multipath signal of the previous period, and the end is t2 more than the maximum time delay of the components of the multipath signal of the previous period . 3. The method according to claim 1, characterized in that the value of Th is chosen equal to half the chip PSP. 4. The method according to claim 1, characterized in that the value of L is chosen equal to the maximum number of multipath components used to demodulate data. 5. The method according to claim 1, characterized in that the power of the components of the multipath signal is evaluated by the decisive search function. 6. A search method for receiving a multipath signal, which consists in generating complex correlation responses of the pilot signal, determining the correlation of the input signal with the reference signal corresponding to the given discrete time delays of the multipath region, determining the values of the decisive function for the given discrete time delays of the multipath region, summing the squares of the in-phase and quadrature parts of the corresponding complex correlation responses of the pilot signal, the values of the generated function with a given threshold h, characterized in that a set of values of the decision function exceeding the threshold and a set of time delays corresponding to them are formed, and estimates of the search for time delays of the components of the multipath signal are formed in stages, and at each stage an estimate of one component is obtained, for which they determine an element with a maximum value from the set of values of the decisive function of the current stage, while the set of values of the decisive function of the first stage is the formed set of of the beginnings of the decisive function, compare the maximum value of the decisive function of the current stage with the threshold h, if the threshold is not exceeded, the formation of the time delay search estimate for the components of the multipath signal is stopped, if the threshold is exceeded, the time delay corresponding to the element with the maximum decisive function of the current stage is considered a search estimate time delay components of a multipath signal, correct the elements of the set of values of the decisive function of the current stage, form the aggregate the range of values of the decisive function of the next stage, excluding from the adjusted set of values of the decisive function of the current stage the element with the maximum value. 7. The method according to claim 6, characterized in that the threshold h is selected proportional to the noise power. 8. The method according to claim 6, characterized in that the correction of the elements of the set of values of the decisive function of the current stage is carried out by subtracting the correction value from the elements of the set, the time delays of which differ from the time delay of the maximum element by no more than a predetermined time threshold η . 9. The method according to claim 6, characterized in that the correction value is selected proportional to the maximum element and inversely proportional to the module of the difference in time delays of the maximum and the adjusted element of the set of values of the decisive function of the current stage. 10. A method for generating soft decisions when receiving a multipath signal, which consists in forming a sequence of correlation responses of information symbols of the components of a multipath signal, carry out a weighted summation of the correlation responses of information symbols of the components of a multipath signal, obtaining the united soft decisions about information symbols, characterized in that up to the formation of a sequence of correlation responses of information symbols select components of multiple chevogo signal, which produces power estimate multipath signal components operate ranking component of the multipath signal in descending order of the power series for ranked component multipath signal except the first form the ratio of the power component to the sum of the powers of all previous ranked component compared formed ratio with a predetermined threshold ν , if the threshold is exceeded, the component is selected to form soft decisions; if the threshold is not exceeded, the comp selection procedure It is completed, the formation of a sequence of correlation responses of information symbols and their weighted summation are carried out for the selected components of the multipath signal. 11. The method according to claim 10, characterized in that the threshold ν is selected according to the allowable energy loss due to the limitation of the components used in demodulation. 12. The method according to claim 10, characterized in that the correlation responses of information symbols for each selected component of the multipath signal are formed by determining the correlation of the input signal with the known SRP over the symbol duration interval. 13. The method according to claim 10, characterized in that the weighting coefficients when summing the correlation responses of the information symbols of the selected components of the multipath signal are selected as complex conjugate estimates of the complex envelope of the information symbols of the components of the multipath signal. 14. A device for receiving a multipath signal, comprising a search unit, a control unit, a soft decision forming unit, wherein the first and second inputs of the search unit and the soft decision forming unit are inputs of the in-phase and quadrature components of the input multipath signal, the third input of the search unit is connected to the first output of the unit control, which is the control output, the second outputs of the control unit, which are the outputs of the time delays of the updated components of the multipath signal of the current period, are coupled with the third inputs of the soft decision forming unit, the first output of the soft decision forming unit is the output of soft decisions about information symbols and the output of the device, characterized in that the signal component update unit is additionally introduced, the output of the search unit being the output of estimates of time delays and values deciding functions of searching for a component of a multipath signal, connected to the first input of the signal component update unit, and the fourth input of the soft decision forming unit, the output of the unit updates of the signal component, which is the time delay output of the updated multipath component, is connected to the input of the control unit, the second output of the soft decision forming unit is the output of the correlation responses of the pilot symbols and time delays of the multipath signal component of the previous period, and is connected to the second input of the signal component update unit . 15. A multipath signal search unit comprising K quadrature correlators, a threshold generating circuit, a pseudo-random sequence generator, a control circuit, a multiplexer, the first and second inputs of the correlators being the first and second inputs of the threshold generating circuit and inputs of the in-phase and quadrature components of the input multipath signal, the third the input of each quadrature correlator is the input of the pseudo-random sequence signal and is connected to the corresponding output of the pseudo-random generator sequence, the fourth input of each quadrature correlator, which is the input of the control signal, is connected to the corresponding first output of the control circuit, the second output of the control circuit, which is the output of time offsets of the pseudo-random sequence generator, is connected to the input of the pseudo-random sequence generator, the output of each quadrature correlator, which is the output of the values of the decisive function for given time delays of the multipath region, is connected to the corresponding input of the multiplexer, the input of the control circuit is an input of control signals, characterized in that the ray detection circuit, the first, second and third key, the first and second comparator, logic node, counter, control node, OR circuit, node for determining the maximum element, are introduced and a second subtractor, module shaper, a register, a memory element of the initial set of decision functions, a memory element of the original set of time delays, a memory register of decision search functions, an estimation memory register by ska time delays, the driver of the correction signal, the output of the multiplexer, which is the output of the values of the decisive function of the multipath region, is connected to the first input of the beam detection circuit, the second output of the control circuit, which is the output of time offsets of the pseudo-random sequence generator, is connected to the second input of the beam detection circuit , the output of the threshold formation circuit, which is the output of the threshold value, is connected to the third input of the ray detection circuit and the first input of the first computer arator, the first output of the ray detection circuit, which is the output of time delays, is connected to the second input of the memory element of the original set of time delays, the second output of the ray detection circuit, which is the output of the values of the decisive functions that exceed the threshold, is connected to the second input of the memory element of the original set of decisive functions, the third output of the ray detection circuit, which is the output of a signal equal to the number of detected components that have exceeded the threshold, is connected to the first input of the control unit, h the fourth output of the ray detection circuit, which is the output of the threshold exceeding indication signal for a given discrete time delay, is connected to the second input of the control unit, the first output of the control unit, which is the output of the recording signal, is connected to the first input of the memory element of the initial set of decision functions and to the first the input of the memory element of the original set of time delays, the output of the memory element of the original set of decision functions, which is the output of the maximum decision function of the current stage, connected to the first input of the node for determining the maximum element, with the first input of the second subtractor, with the first inputs of the first and second keys, the second input of the node for determining the maximum element is a control input and connected to the fifth output of the control node, the output of the node for determining the maximum element, which is the address output of the selected maximum element, connected to the fifth input of the control unit, the second output of the control unit, which is the output of the signal indicating the presence of a maximum, is connected to the second input of the first key and the first input of the logical node, the output of the first key is connected to the second input of the first comparator, the output of the first comparator, which is the output of exceeding the threshold, is connected to the second input of the logical node, with the first input of the counter, with the second controlled inputs of the second and third keys , with the first input of the register, the third output of the control node, which is the output of the number of detected components of the current search period, is connected to the second input of the counter, the output of the counter is connected to the first input with OR, the second input of which is connected to the output of the logical node, the output of the OR circuit, which is the output of the indicator of the end of the search procedure, is connected to the third input of the control node, the fourth output of the control node, which is the output of the control signal for generating the search estimate, is connected to the first input of the register the memory of the decision functions of the search, with the first input of the register of the memory of the evaluation of the search for time delays and with the third input of the driver of the correction signal, the second input of the register of the memory of the decision functions of the search the first input of the correction signal generator is combined and connected to the output of the second key, which is the output of the maximum value of the decisive function of the current stage, the second inputs of the search delay evaluation register memory register and the register are combined and connected to the output of the third key, the first input of which is combined with the first input of the first subtractor and connected to the output of the memory element of the original set of time delays, which is the output of the value of the time delay corresponding to the maximum element, the output p a histra, which is the output of the time delay values, connected to the second input of the first subtractor, the output of the first subtractor, which is the output of the time delay difference, is connected to the input of the module shaper, the output of the module of the difference of the shaper of the module is connected to the second input of the correction signal shaper and the first input of the second comparator, the second input of the second comparator is a threshold input, the output of the second comparator is the output of exceeding the threshold and is connected to the fourth input of the control unit the fourth input of the correction signal generator, the output of the correction signal generator, which is the output of the correction signal, is connected to the second input of the second subtractor, the output of which is connected to the second input of the memory element of the initial set of decision functions, the output of the memory register for evaluating the search for time delays and the output of the register of decision memory search functions are combined and are the output of the multipath signal search unit. 16. A soft decision forming unit comprising a multi-channel data receiver and a soft decision combining circuit, the first and second inputs of the multi-channel data receiver are inputs of the in-phase and quadrature components of the input signal and the first and second inputs of the soft decision forming unit, the output of the multi-channel data receiver is connected to the input soft decision combining scheme, which is the input to the sequence of correlation responses of information and pilot symbols of the multipath signal component a, the output of the combined soft decisions of the soft decision combining circuit is the first output of the soft decision forming unit, characterized in that the memory register, the first, second and third multiplexers, the power estimation unit, the key, the first, second and third memory nodes, a counter are additionally introduced the adder, the first and second comparators, the OR circuit, the NOT circuit, the ranking node, the relation shaper, the control node, and the third input of the soft decision forming unit is the time delay inputs and the first inputs of the second mul typlexer, the fourth input of the soft decision forming unit is the input of the values of the search function of the current stage and the first input of the first multiplexer, the output of the multi-channel data receiver is connected to the input of the power estimation unit, which is the input to the sequence of correlation responses of pilot symbols and time delays of the multipath signal components, except Moreover, the output of a multi-channel data receiver is the output of a sequence of correlation responses of pilot symbols and computer time delays the multipath of the multipath signal and the second output of the soft decision forming unit, the second inputs of the first and second multiplexers are control and connected to the first output of the control unit, the output of the power estimation unit, which is the output of the power evaluation of the multipath component, is connected to the first input of the third multiplexer, the output of the first the multiplexer is combined with the output of the third multiplexer, which is the output of the signal for evaluating the power of the components of the multipath signal, and connected to the first input of the third about the memory node, the output of the second multiplexer, which is the time delay output of the updated multipath component, is connected to the first input of the first memory node, the second input of which is combined with the second input of the third multiplexer and connected to the second control output of the control node, the third output of the control node is the output control signals and is connected to the second input of the third memory node, the first input of the first comparator and the first input of the memory register, the output of the third memory node, which is the output of estimating the power of the current component is connected to the first input of the ranking node, the first input of the adder and the first input of the ratio generator, the output of the ranking node, which is the output of the ordered addresses of the memory cells, is connected to the first input of the control node, the second input of the ranking node is connected to the fourth control output control unit, the fifth output of the control unit, which is the output of the recording signal, is connected to the second input of the second memory node, the first input of which is connected to the output of the adder, output q of the second memory node, which is the output of the total powers of all the previous ranked components, is connected to the second input of the adder and the second input of the ratio former, the output of the ratio former is connected to the first input of the second comparator, the second input of which is a threshold input, the output of the second comparator, which is the output the threshold signal is connected to the input of the circuit NOT, to the counter input, and to the first key input, the second key input is connected to the output of the first memory node, the key output, which is the time delay output of the multipath component, connected to the second input of the memory register, the output of the memory register, which is the output of the set of time delays of the selected components, is connected to the third input of the multi-channel data receiver, the circuit output is NOT an inverse output and connected to the first input of the OR circuit, the counter output is a signal output equal to the number of threshold exceeded, and is connected to the second input of the first comparator, the output of the first comparator is connected to the second input of the IL circuit And, the output of which is the output of the signal indicating the completion of the selection procedure for components for forming soft decisions and is connected to the second input of the control unit.

Images