RU2356167C1 - Method for adaptive transfer of data in radio link with pseudo-random tuning of working frequency - Google Patents

Abstract

FIELD: physics, radio.

SUBSTANCE: invention is related to the field of radio engineering and may find application in adaptive systems of special radio communication for data transfer along radio channel under conditions of purposive noise complex effect. Method is characterised by the fact that on transmitting end of radio link information speed is transformed into technical speed with separation of information packets comprising frames and slots, values of variables for generation of which are defined in process of radio communication, modulation of information and service slots at appropriate frequencies of transmission in compliance with the first pseudo-random sequence (PRS); on receiving end - separation of accepted signal is done into service and information parts, for service parts on appropriate frequencies in compliance with the second PRS transmission channel is selected, generation of clock signal, peaks of mutual-correlation function for accurate synchronisation during demodulation, quality of received clock sequence is analysed, and this analysis is used to define error probability per service bit, which is transferred in service frames to transmitting side, for information part demodulation of service frames is carried out at appropriate frequencies of reception in compliance with the first PRS, variable values received during demodulation are used for generation of the following packet, at that received information frame is used to analyse quality of information with separation of error per information bit, which is sent in service frames to transmitting side, and also used for calculation of probability of correct reception of possible methods of noise-immune coding.

EFFECT: introduction of clock signal processing adaptation to changes of noise environment and reduction of information losses during reception.

2 cl, 10 dwg

Description

The proposed method relates to the field of radio engineering and may find application in adaptive systems of special radio communications for transmitting data over a radio channel under the influence of a complex of intentional interference.

A large number of inventions are known that use such a method of transmitting information as pseudo-random frequency tuning (frequency hopping).

There is a method of transmitting information for use in communication systems operating in conditions of uncertain interference, described in [1], in which the increase in noise immunity of a radio line under the influence of unsteady interference is achieved in the presence of long-term heterogeneous quality of individual frequency channels.

A known method of transmitting information according to the patent [2], which considers the method of frequency hopping, involving the use of pseudo-random sequences (PSP), in order to change the value of the operating frequency generated by the pseudo-random law at pseudo-random times, guaranteed synchronously for both sides of the radio line.

The disadvantage of the above methods is the lack of adaptation of the clock signal to changes in the noise environment.

The closest in technical essence to the proposed method is the method according to the patent [3], adopted as a prototype.

The prototype method is as follows.

At the transmitting end, the input signal is divided into blocks with an equal number of elements. The resulting sequence of blocks is recoded by changing the order of the pulses between the blocks. The received signal is changed block by block, increasing the pulse repetition rate, after which the signal is modulated at the appropriate frequencies and radiated into space.

At the receiving end, the received signal is converted to an intermediate frequency, then demodulation, inverse transformation of the information rate, decoding, combining the blocks and supplying information to the receiver.

In this case, transcoding is performed so that 1 block element 1 becomes 1 block element 1, 2 block 1 element becomes 1 block 2 element, L block 1 element becomes 1 block L element, 1 block 2 element becomes 2 block 1 element, 2 element 2 block becomes 2 block 2 element, L block 2 element becomes 2 block L element, 1 block N element becomes N block 1 element, 2 block N element becomes N block 2 element, L block N element becomes N block L element. Using error correction codes for transmission, it is possible to correctly receive information. That is, depending on the correcting ability of the code, it is possible to correct errors when the number of frequencies is equal to the correcting ability of the code.

In the prototype method, the conversion of information rate is aimed at interleaving bits for subsequent application of error-correcting code. However, this method is only effective under unintentional interference. In special radio communications, the central direction is to increase noise immunity in the context of continuous improvement of radio countermeasures (SRS), in particular when exposed to a radio frequency transmission frequency hopper in order to suppress one of the delayed interference effective from the point of view of energy capabilities in combination with a barrage noise in the part of the band (RFP ) The application of frequency hopping leads to the allocation of time intervals for synchronization of the radio link and for transferring it from frequency to frequency without loss of message elements at these intervals. The last requirement leads to the conversion of the block into a slot: the information bits are “framed” by the frequency tuning bits (the time spent by the equipment on tuning) and the protection bits (taking into account the desynchronization of generators). The lack of clear bitwise synchronization can negate all the advantages of the frequency hopping system. With this in mind, PSAs act not only on slots with information (delayed interference, tracking over frequencies), but also on slots with clock signals (using STDs). To protect against the latter, it is advisable to introduce a change in the type of clock signals in the radio line.

However, a change in the form of the clock signal cannot but affect the overall structure of the time diagram of the radio line. Modern data transmission equipment should be designed to operate at different transmission speeds (corresponding to GOST 17422), as well as to protect information from intentional interference by changing the frequency hopping frequency, the type of noise-resistant code and the use of the feedback channel between the receiver and transmitter to analyze the interference situation. Thus, the adaptation of the clock signal should entail the adaptation of the entire timing diagram of the radio link.

The last condition indicates that the unchanged radio line presented in the prototype method is not able to provide a given quality of communication throughout the communication session, since the input conditions may change from time to time. In this case, it becomes necessary to use an adaptive radio link, which, using the regular search process, constantly searches for the optimum within the limits of an acceptable class of capabilities. Thus, the disadvantage of the prototype method is the low quality of information transfer under the influence of intentional interference.

The aim of the invention is the introduction of adaptation of the above method of processing the clock signal to changes in the noise environment and reduce information loss during reception.

секунд, состоящего из g служебных фреймов для обмена информации о размере формируемых пакетов и о качестве принимаемых сигналов на приеме и передаче и

информационных фреймов размером Ψ бит, учитывая априорно известные: минимально необходимое число бит для служебных фреймов e, время, затрачиваемое аппаратурой на перестройку частоты T_n, и время рассинхронизации частот на приема и передаче T_р, причем g является функцией от T, е и Ψ, а каждый информационный фрейм имеет размер d+δ слотов, где d - количество служебных слотов радиолинии, равное априорно известному числу повторов служебных слотов, a δ - количество информационных слотов, каждый служебный слот имеет размер y+L+z бит, а каждый информационный слот - размер y+x+q бит, где

- биты для перестройки частоты ППРЧ, L - биты синхропоследовательности,

- биты информационного блока, z=[L·λ₁] и q=[x·λ₂] - защитные биты, λ₁ и λ₂ - коэффициенты, зависящий от количества частот ППРЧ для канала синхронизации и информационного канала, соответственно, и априорно известных условий распространения сигнала, причем L соответствует той длине синхропоследовательности из имеющихся, которая соответствует минимальному среднему риску, рассчитываемому на основе определения вероятности ошибки на служебный бит в принимаемой синхропоследовательности, получаемой на основе определения качества данной синхропоследовательности, причем значение вероятности ошибки на служебный бит принимается в служебных фреймах, значения L, z, x, q, d, δ, g, h, λ₁, λ₂ для формирования следующего пакета передают приемной стороне в служебных фреймах, модуляцию информационных слотов на соответствующих частотах передачи, в соответствии с первой псевдослучайной последовательностью (ПСП), модуляцию служебных слотов на соответствующих частотах передачи, в соответствии со второй ПСП, также в служебных фреймах принимается значение вероятности ошибки на информационный бит, получаемое на приемной стороне на основе определения качества информационного фрейма; на приемном конце радиолинии проводят разделение принятого сигнала на служебную и информационную части, где для служебной части с помощью частотной фильтрации на соответствующих частотах приема, в соответствии со второй ПСП, осуществляют выбор канала, по которому производилась передача, на основании служебной информации, принятой в служебных фреймах, формируют синхросигнал размера L и по нему формируют пики взаимно корреляционной функции для точной тактовой синхронизации при последующей демодуляции сигнала, причем по принятой синхропоследовательности производят ее анализ качества с определением вероятности ошибки на служебный бит, которую передают в служебных фреймах передающей стороне, для информационной части осуществляют демодуляцию служебных фреймов и информационных блоков на соответствующих частотах приема, в соответствии с первой ПСП, с учетом полученной точности тактовой синхронизации и известных значений z, x, q, d, δ, g, h, λ₁, λ₂, далее полученные после демодуляции служебных фреймов значения z, x, q, d, δ, g, h, λ₁, λ₂ используют для демодуляции следующего пакета, причем по принятому информационному фрейму производят анализ качества информации с определением вероятности ошибки на информационный бит, которую передают в служебных фреймах передающей стороне.To achieve this goal in a method for adaptive data transmission in a radio line with a pseudo-random tuning of the operating frequency, including at the transmitting end dividing the input digital signal at speed C into information blocks, emitting a modulated signal into space, at the receiving end, combining the demodulated blocks into the original signal and feeding to the recipient of the information according to the invention, at the transmitting end of the radio link, the information speed C is converted to the technical speed V of the digital signal selection of information packets of duration

seconds, consisting of g service frames for exchanging information about the size of the generated packets and the quality of the received signals at the reception and transmission, and

information frames of size Ψ bits, taking into account a priori known: the minimum required number of bits for service frames e, the time taken by the equipment to tune the frequency T _n , and the time of the frequency desynchronization in reception and transmission T _p , and g is a function of T, e and Ψ , and each information frame has a size of d + δ slots, where d is the number of service slots of the radio line equal to the a priori known number of repetitions of service slots, a δ is the number of information slots, each service slot has a size of y + L + z bits, and each information n slot - size y + x + q bits, where

- bits for frequency hopping, L - bits of the sync sequence,

are the bits of the information block, z = [L · λ ₁ ] and q = [x · λ ₂ ] are the protective bits, λ ₁ and λ ₂ are the coefficients depending on the number of frequency hopping frequencies for the synchronization channel and the information channel, respectively, and a priori known propagation conditions of the signal, and L corresponds to the length of the available sync sequence that corresponds to the minimum average risk calculated on the basis of determining the probability of an error on the service bit in the received sync sequence, obtained on the basis of determining the quality of a given sync been consistent, the value of the error probability to control bit is received in the service frames, the values of L, z, x, q, d, δ, g, h, λ _1, λ ₂ for forming the next packet is transmitted the receiving side in service frames, modulation information slots at the appropriate transmission frequencies, in accordance with the first pseudorandom sequence (SRP), the modulation of the service slots at the corresponding transmission frequencies, in accordance with the second SRP, the error probability per information bit is also received in the service frames, according to radiated at the receiving side based on determining the quality of the information frame; at the receiving end of the radio line, the received signal is divided into service and information parts, where for the service part, using the frequency filtering at the corresponding reception frequencies, in accordance with the second SRP, a channel is selected for transmission on the basis of service information received in the service frames, form a clock signal of size L and the peaks of the cross-correlation function are formed on it for accurate clock synchronization during subsequent demodulation of the signal, and according to the received clock of the sequence perform its quality analysis with the determination of the probability of an error on the service bit, which is transmitted in the service frames to the transmitting side, for the information part, the service frames and information blocks are demodulated at the corresponding reception frequencies, in accordance with the first SRP, taking into account the obtained accuracy of clock synchronization and known values of z, x, q, d, δ, g, h, λ ₁ , λ ₂ , then obtained after demodulating service frames the values z, x, q, d, δ, g, h, λ ₁ , λ ₂ are used for demodulation next package, p When in use, from the received information frame analysis produce quality information with the determination error probability on the information bits, which are transmitted in the transmitting side service frames.

где n_i - длина кода, есть целое число, причем информационные блоки информационного фрейма кодируют найденным кодом, если уменьшение скорости передачи информации, вызванное применением данного кода в

раз, где s_i (i∈{1, 2,…,M}) - число информационных символов кода, приведет к увеличению качества приема информации не менее чем в µ раз и

где отношение

- априорно известная величина, определяемая исходя из задач теории информации, номер i∈{0, 1, 2, …, M} используемого кода передают приемной стороне в служебных фреймах, где i=0 соответствует случаю некодированных информационных блоков, на приемном конце радиолинии производят декодирование декодером с номером i∈{0, 1, 2, …, М} информационных блоков информационного фрейма, по принятым и декодированным информационным блокам производят анализ качества принимаемой информации, на основе которого определяют вероятность ошибки на информационный бит, которую передают в служебных фреймах передающей стороне, причем полученное после демодуляции служебных фреймов значение i∈{0, 1, 2, …, М} используют для декодирования следующего пакета.In addition, at the transmitting end of the radio link, based on the probability of an error per information bit, obtained on the basis of determining the quality of the information frame, the probabilities of the correct reception of W ₁ , ..., W _{M of} each of the M possible error-correcting coding methods used in this radio device are calculated, and the probability W for the correct reception of an uncoded information frame, where the error probability per information bit is taken in service frames, and if W is less than the necessary a priori known level search is conducted of the error-correcting code, in which W _i (i∈ {1, 2, ..., M}) is greater than W and obtained, based on the known x, δ, the number of its codeword

where n _i is the length of the code, there is an integer, and the information blocks of the information frame are encoded by the found code if the decrease in the information transfer rate caused by the application of this code in

times, where s _i (i∈ {1, 2, ..., M}) is the number of information symbols of the code, will lead to an increase in the quality of information reception by at least µ times and

where is the relation

- an a priori known value, determined based on the tasks of information theory, the number i∈ {0, 1, 2, ..., M} of the used code is transmitted to the receiving side in service frames, where i = 0 corresponds to the case of non-encoded information blocks, at the receiving end of the radio line decoding by the decoder with the number i∈ {0, 1, 2, ..., M} of the information blocks of the information frame, the received and decoded information blocks analyze the quality of the received information, based on which the probability of error per information bit is determined, otorrhea service frames transmitted in the transmitting side, wherein the received frames after demodulation service value i∈ {0, 1, 2, ..., M} are used for decoding the next packet.

The proposed method for adaptive data transmission in a radio frequency hopping system is as follows.

At the transmitting end, the information speed C bit / s is converted to the technical speed of the used radio means V Bod. An information package consisting of frames and slots is formed. To do this, first, based on the available characteristics of the frequency generation devices, the time spent by the equipment for frequency tuning, T _n s and the frequency desynchronization time at the reception and transmission T _p s are determined. In addition, the minimum required number of bits for service frames c and the number of service slots d, equal to the number of repetitions of one synchronization slot, are determined.

, где [] - целая часть.Based on the value of T _n , the required number of frequency tuning bits is calculated

where [] is the integer part.

In addition, the number of information bits in the slot is calculated.

,

,

where q = [x · λ ₂ ] is the number of protective bits of the information slot;

δ is the number of information slots of the frame;

λ ₂ is a coefficient depending on the number of frequency hopping frequencies and a priori known signal propagation conditions (for example, for a fixed frequency, you can take λ ₂ ≈3%; for frequency hopping of two frequencies - λ ₂ ≈6%, etc.).

Moreover, the number of information slots of the frame δ can sequentially take values from 1 to infinity, and the first value δ, which gives an integer value x, will be the number of information slots of the frame δ, and the corresponding value x will be the number of information bits of the slot.

An analysis is made of the reception quality of the sync sequence from the previous received packet. Based on this analysis, the probability u _x error per bit in the synchronization channel is determined. The value u _{cx is} used to determine the probabilities of the correct reception of P _pr and false alarm P _lt for each of the possible values of L _j where j∈ {1, ..., K}, i.e. those lengths of sync sequences that are used in a given radio. For each of the pairs received

R _CR and R _LT calculated average risks R _j , where j∈ {1, ..., K}, according to the formula

where R _s.sign - the a priori probability of the absence of a signal;

P _nal.sign - the a priori probability of the presence of a signal;

a and b are weights, depending on how dangerous P _lt is compared to 1-P, _etc.

The minimum of the calculated average risks R ₁ , ..., R _K corresponds to the length of the sync sequence L, which should be used in the service slots of the generated package. In addition, the number of protective bits of the service slot z = [L · λ ₁ ] is calculated, where λ ₁ is a coefficient depending on the number of frequency hopping frequencies and a priori known signal propagation conditions, similar to that described above.

Figure 1 shows the structure of the information frame, which contains d service slots and δ information slots.

At the same time, each service slot contains (y + L + z) bits, and each information slot contains (y + x + q) bits.

Then the total packet transmission time is calculated as

where Ψ = (y + L + z) · d + (y + x + q) · δ is the number of bits in the frame;

g is the number of service frames, which is determined as follows:

;1) if e≤Ψ, then g ₀ = 1; if e> Ψ, then

;

2) g can sequentially take values from g ₀ to infinity, and the first value of g, which will give an integer value of T, will be the number of service frames, and the corresponding value of T will be the transmission time of the packet.

информационных фреймов содержится (h·Ψ) бит.Figure 1 shows the structure of the information package, which contains g service frames and h information frames. Moreover, g service frames contain (g · Ψ) bits, and in

information frames contain (h · Ψ) bits.

Given the above, the input information signal is divided into blocks of x bits each. Next, d service slots are formed at the transmitting end for synchronization of the radio line with a size of (y + L + z) bits each, and each service slot contains (Fig. 1):

y frequency tuning bit;

L bit of the synchronization sequence, which provides the minimum value of the average risk R in the formula (1);

z guard bits.

Also, at the transmitting end, δ information slots are formed with a size of (y + x + q) bits each, and each information slot contains (Fig. 1):

y frequency tuning bit,

x bit of information;

q guard bits.

The resulting d service and δ information slots are combined into an information information frame. The generated g service frames intended for the exchange of service information about the size of the generated packets and the interference situation between the receiver and the transmitter are combined with h information frames in an information packet of duration T s (see Fig. 1). During the transmission of the packet to the receiver, T · C bits of the input information signal are transmitted from the transmitter.

скач/с. Модулированные слоты пакета излучаются в пространство. Величины х, L, λ₁, λ₂, d, δ, g, h для следующего пакета передаются в служебном фрейме.Further, each of the received information slots of the packet is modulated by the corresponding transmission frequency determined by the first SRP, and each of the received service slots of the packet is modulated by the corresponding transmission frequency determined by the second SRP, and the number of frequency hopping frequencies for service slots and information slots is different, which affects the value of the coefficients λ ₁ and λ ₂ as described above. The frequency hopping rate is defined as

jump / s. Modulated packet slots are emitted into space. The values x, L, λ ₁ , λ ₂ , d, δ, g, h for the next packet are transmitted in the service frame.

Тот код, у которого W_i(i∈{1, 2, …, M})=max и Ω_i - целое число, является кодом (n, s), который может осуществлять кодирование информационного фрейма.In addition to the analysis of the reception quality of the sync sequence, the radio analysis of the reception quality of information slots is also carried out. Based on this analysis, the signal-to-noise and signal-to-noise ratios are determined. These relations are used to determine the probability of the correct reception W of the information frame, as well as the probabilities of the correct reception of W ₁ , ..., W _M for each of the possible (n ₁ , s ₁ ), ..., (n _M , s _M ), i.e. those error-correcting codes that are used in a given radio medium (where n is the total number of code symbols and s is the number of information symbols of the code). If W is less than the necessary a priori known level, a search is made for those error-correcting codes for which W _i (i∈ {1, 2, ..., M}) exceeds W. For each of these codes, the number of code words in the information frame is calculated

The code for which W _i (i∈ {1, 2, ..., M}) = max and Ω _i is an integer is the code (n, s) that can encode the information frame.

Figure 2 presents the structural diagram of the package for the case of the use of error-correcting coding.

. В результате кодовое слово может занять не один, а

информационных слотов, содержащих только s битов информации. Остальные биты в f информационных слотах - проверочные биты кода ε=n-s. В результате в каждом слоте могут быть:However, the use of the code (n, s) leads to a decrease in the information rate C to the information rate

. As a result, the code word may take more than one

information slots containing only s bits of information. The remaining bits in f information slots are the verification bits of the code ε = ns. As a result, in each slot there can be:

1) only bits of information;

2) information bits and check bits;

3) only test bits.

раз приведет к улучшению качества приема сигнала в

и более раз (где отношение

определяется априорно исходя из задач теории информации [5, стр.607-658]) данный код используется. Таким образом, определяется вид помехоустойчивого кода, который соответствует номеру М из известной на передающей и приемной стороне второй кодовой книги, и сообщается в служебных фреймах на приемную сторону для следующего пакета.The criterion for applying the code (n, s) is as follows: if the loss of information transfer rate in

times will lead to better signal reception in

and more times (where is the ratio

is determined a priori based on the tasks of information theory [5, p. 607-658]) this code is used. Thus, the type of error-correcting code is determined, which corresponds to the number M from the second codebook known on the transmitting and receiving sides, and is reported in the service frames to the receiving side for the next packet.

At the receiving end of the radio line using frequency filtering, the channel through which the transmission was carried out is determined. Based on the information received in the service frames, the values of y, x, L, λ ₁ , λ ₂ , d, δ, g, h are determined for the next packet.

According to the synchronization sequence of size L, peaks of the cross-correlation function (FCF) are formed to synchronize the received signal with its subsequent demodulation and decoding. After demodulation, the decoding takes into account the type of error-correcting code used by the number M, adopted in the service frame. Decoded blocks are combined into the original signal, which is fed to the recipient of information. At the same time, a quality analysis is carried out: 1) the received sync sequence and 2) information frames. The results of the analysis are transmitted in service frames to the transmitting end.

As an example (only the speeds determined by GOST 17422 are used), we assume that through a radio station having a technical speed of V = 19.2 kBaud, data is transmitted from data transmission equipment with a speed of C = 12 kbit / s. An example of a timing diagram satisfying this case is shown in FIG. 3, from which it follows that the number of information bits in the block is x = 45, and the frequency hopping rate is γ = 300 jump / s. Three binary M-sequences with L ₁ = 31, L ₂ = 63 and L ₃ = 127 are used as a sync sequence, and a discrete matched filter (DSF) is used as a device for their determination at reception [4, p. 275]. It is assumed that the used DSF detects the synchronized sequence in real time [4, pp. 279-282] by uniformly gating each conditional bit not at one but at u = 8 points, which uses signal information more fully: if at least 3 matches consecutive sequences (from the obtained u = 8) with a copy of the signal (with a tolerance of up to u ₀ errors for each sequence), a decision is made on the reception of the clock signal. The appearance of each of the M-sequences is set by arranging the sequence of trigger cells in the DSF [4, p. 279].

If in formula (1) we take R _ssign = 0.8, P _cash.sign = 0.2 and the ratio a / b = 0.01, then with the probability of error per bit p∈ {0.15, ..., 0.45}, determined by the signal quality analyzer, formula (1) you can calculate the average risks R ₁ , R ₂ , R ₃ .

Figure 4 presents the average risks when taking M-sequences of various lengths. Analyzing FIG. 4, it can be determined that the M-sequence whose schedule lies “below” the others is used as the synchronization for the next packet.

After determining the type of the clock signal, the following problem is solved: determining the need for the use of an error-correcting code under the influence of delayed interference.

To do this, the error probabilities per bit are calculated:

Where

Where

ϑ is the signal power;

σ ₀ ² is the spectral power density of white noise;

σ _c ² is the spectral power density of the delayed interference.

Figure 5 presents a timing diagram of the information bits of the slot when exposed to delayed interference.

For a given timeline

x _n = [x · θ] is the number of uninfected information bits in the slot;

x _n = xx _n is the number of affected information bits in the slot;

- коэффициент запаздывания помехи;Where

- delay coefficient of interference;

T _SL - time of delayed interference;

T _s - the delay time of the interference.

Finally, the average probability of error per bit is determined

If the conditions of signal propagation require a higher guarantee from the user to bring information to the recipient, the timing diagram allows you to "sacrifice" the speed of information transfer to increase the quality of the latter. An example of a timing diagram satisfying this case is shown in FIG. 6. In the case of using majority processing as a noise-tolerant code (selection rule 3 of 5), the timing diagram takes the form shown in Fig. 6, and the average probability of error per bit, according to [6], is determined

where ξ is the summation index.

Since in case of using a majority the information rate will decrease from C = 12 ^kbit / _s to β = 2.4 ^kbit / _s , the number of bits of information in the slot will be 396. Therefore, it is possible to calculate the coefficient of improvement in the quality of bit transmission as a ratio of the probability of correct reception of 396 bits of information after and before applying the major

while the coefficient of use of the speed of information is

The dependence of µ _Kach on τ for the values of the delay coefficient θ = 0.8 and θ = 0.2 are shown in Figs. 7 and 8, respectively. Thus, provided that the user selects a criterion for changing speeds

(that is, the user “sacrifices” the transmission rate only if the decrease in the speed by a factor of µ will lead to an improvement in the transmission quality by more than 2µ times) the quality of the information signal is analyzed, which determines the values of τ, θ and φ, from which the values of µ _ck and μ _kach , which are compared according to criterion (10) and the decision is made based on the result of the comparison: whether to change the transmission speed of information or not (that is, whether to apply the noise-immunity protection method or not).

In this case, at θ = 0.8, the majority is used at τ> 1.7, and at θ = 0.2 - at τ> 4.8.

Figure 9 shows the dependence of _μqu on τ and θ for various cases: 1) with φ = 5; 2) at φ = 20,000; 3) the boundaries of the criterion. If as a result of the analysis of the signal quality it is determined that the operating point lies on any of the planes (1 or 2) above plane 3, majority processing is applied.

Functional diagram of a device for implementing the proposed method of transmitting data with frequency hopping using noise-correcting coding is shown in figure 10, where the following notation:

1 - source of information (AI);

2 - speed switch (CS);

3 - packer;

4 - the first demultiplexer (D1);

5 ₁ -5 _M - encoders;

6 - modulator;

7 - the first antenna-feeder unit (AFB1);

8 - clock generator (GTI);

9 - code sequence generator (GKP);

10 - control unit (CU);

11 - the first signal quality analyzer (ACS 1);

12 - the first frequency synthesizer (MF1);

13 - pseudo-random sequence generator (GPSP);

14 - frequency converter (IF);

15 - intermediate frequency amplifier (UPCH);

16 - demodulator;

17 - the second demultiplexer (D2);

18 ₁ -18 _M - decoders;

19 - depacketizer;

20 - information receiver (PI).

21 - the second antenna-feeder unit (AFB2);

22 - signal element identification unit (BOES);

23 ₁ -23 _K - comparison schemes (SS);

24 ₁ -24 _K- counters;

25 - the third multiplexer (MOH);

26 - the first multiplexer (Ml);

27 - second multiplexer (M2);

28 - second signal quality analyzer (AKS2);

29 - the second frequency synthesizer (MF2);

A device for implementing the proposed method contains a series-connected information source (AI) 1, the input of which is the first input of the device, and a speed switch (CS) 2, the output of which is connected to the first signal input of the packetizer 3, the output of which is connected to the signal input of the first demultiplexer (D1 ) 4, M outputs are connected to respective signal inputs of encoders May ₁ M 5 _M, whose outputs are connected respectively with the M signal inputs of the first multiplexer (M1) 26, whose output is connected to the first the signal input of the modulator 6, the output of which is connected to the input of the first antenna-feeder block (AFB1) 7 whose output is the first output device.

In addition, the device comprises a second antenna-feeder unit (AFB2) 21 connected in series, the input of which is the second input of the device, a frequency converter (IF) 14 and an intermediate frequency amplifier (IF) 15, the output of which is connected to the first signal input of the demodulator 16, the output which is connected to the signal input of the second demultiplexer (D2) 17, the M outputs of which are connected to the signal inputs of the corresponding M decoders 18 ₁ -18 _M , the outputs of which are connected respectively to the M signal inputs of the second multiplexer (M2) 27, the output of which is connected to the signal input of the depacketator 19, the output of which is connected to the signal input of the information receiver (PI) 20, the output of which is the second output of the device.

Also, the outputs of the M decoders 18 ₁ -18 _{M are} connected respectively to the M signal inputs of the first signal quality analyzer (AKS1) 11, the (M + 1) -th signal input of which is connected to the output of the demodulator 16.

The second output of AFB2 21 is connected to the first signal input of the signal element identification unit (BOES) 22, the output of which is connected to the signal inputs K of the comparison circuits (CC) 23 ₁ -23 _K , the first outputs of which are connected to the signal inputs K of the corresponding counters 24 ₁ -24 _K , the outputs of which are connected respectively to the K signal inputs of the second signal quality analyzer (AKS2) 28, the (K + 1) -th signal input of which is connected to the output of the BOES 22.

The second outputs K of the comparison circuits (CC) 23 ₁ -23 _{K are} connected respectively to the K signal inputs of the third multiplexer M3 25, the output of which is connected to the second signal input of the demodulator 16.

From the first to the tenth outputs of the control unit (BU) 10 are connected respectively to the control inputs of the packetizer 3, D1 4, M1 26, modulator 6, depacketator 19, D2 17, M2 27, demodulator 16, GKP 9 and M3 25, and the output of GKP 9 connected to the second signal input of the packetizer 3. The output of the AKS1 11 is connected to the first input of the control unit 10, and the output of the AKS2 28 is connected to the second input of the control unit 10.

The first output of the pseudo-random sequence generator (GPS) 13 is connected to the signal input of the first frequency synthesizer (MF1) 12, the first output of which is connected to the second signal input of the modulator 6, and the second output to the second input of the inverter 14.

The second output of the GPSP 13 is connected to the signal input of the second frequency synthesizer (MF2) 29, the first output of which is connected to the third signal input of the modulator 6, and the second output to the second signal input of the BOES 22.

The output of the clock pulse generator (GTI) 8 is connected to the clock inputs of AI 1, KS 2, packetizer 3, D1 4, M encoders 5 ₁ -5 _M , M1 26, modulator 6, GKP 9, BU 10, AKS1 11, AKC2 28, СЧ1 12, СЧ2 29, ГПСП 13, BOES 22, demodulator 16, Д2 17, М decoders 18 ₁ -18 _М , М2 27, depacketator 19, ПИ 20, М3 25, as well as К counters 24 ₁ -24 _К and SS 23 ₁ -23 _K.

The device operates as follows. At the transmitting end, from the first input of the device, the signal enters AI 1. From AI 1, data at a speed of C bits / s intended for transmission over a communication channel is sent to KS 2, where the information speed C is converted to the technical speed V Baud. From KS 2, information is fed to the first signal input of the packetizer 3, where it is divided into information blocks with the organization of slots, frames, and data packets based on a control signal coming from the first output of control unit 10. GKP 9 based on a control signal coming from the ninth output BU 10 generates bits of the sync sequence and issues them to form the synchronization slots in the packetizer 3.

From the output of the packer 3, the information blocks are sent to the signal input D1 4, to the control input of which a signal is supplied from the second output of the control unit 10, based on which the information block in D1 4 is divided into M channels, the signals of each of which are supplied from the corresponding M outputs of the block D1 4 to the signal inputs of the corresponding M encoders 5 ₁ -5 _M , where there is noise-tolerant coding of the information bits of the slot. From the outputs of blocks 5 ₁ -5 _M, coded information slots are fed to the corresponding signal inputs M1 26, in which, based on the control signal coming from the third output of the control unit 10, M encoded signals are multiplexed into one encoded signal.

Thus, from the output of M1 26 to the first, signal input of the modulator 6, an information packet containing service slots and coded information slots is supplied.

In block 6, the synchronization sequence of the service slots is modulated using the transmission frequency obtained at the first output of SCH2 29 based on the second SRP code supplied to its signal input from the second output of the GPSSP 13, and the encoded information bits of the information slots are modulated using the transmission frequency obtained at the first the output of SCH1 12 based on the first SRP code supplied to its signal input from the first output of the GPSP 13. Moreover, the frequency tuning speed in the modulator 6 is determined based on the control signal, coming output of the fourth unit BU 10. The modulated information is emitted into the space 7 via AFB1.

At the receiving end, the input signal, which is a mixture of the useful signal and interference, is received by AFB2 21, where it is divided into service and information slots.

A part of the input signal containing service slots from the second output of AFB2 21 is fed to the first signal input of the BOES 22. The sequence of random elements of the mixture is converted into BOES 22 into a random sequence of decisions about the symbols corresponding to the received elements of the signal using the receive frequency received at the second output SCH 29 on the basis of the second SRP code supplied from the second output of the GPSS 13. This sequence of decisions on symbols in the general case due to the action of interference does not coincide with any of the sequences of the symbol s of elements that were used in the formation of sync sequences in HCP 9. In BOES 22, the mixture of the signal element and the noise is optimally processed, and a sequence of decisions is issued on its output about which sync sequence element (logical “0” or logical “1”) worked, which arrives at the (K + 1) -th signal input of AKS2 28. In addition, the obtained sequence of decisions on symbols is compared element-wise with the codes of the expected sync sequences in the corresponding CC 23 ₁ -23 _K.

The number of logical "1" at the first outputs of the SS 23 ₁ -23 _K , equal to the number of matches in each channel, is fixed by the corresponding counters 24 ₁ -24 _K and fed to the corresponding inputs of AKS2 28, where the quality of the synchronization channel is analyzed, the result of which is fed to the second input BU 10. In addition, as a result of comparison in SS 23 ₁ -23 _K , the initial clock cycles of information slots in K channels are determined by cross-correlation functions, which are fed from the second outputs of SS 23 ₁ -23 _K to the corresponding signal inputs M3 25, in the cat based on the signal coming from the tenth output of the control unit 10, they are multiplexed into one signal supplied to the second signal input of the demodulator 16.

Another part of the input signal containing information slots is fed to the first input of the inverter 14, in which it is converted to a reception frequency, the value of which is determined based on the first code of the inverter, is fed to the second input of the inverter 14 from the second output of the MF1 12.

Then, the intermediate frequency signal is amplified in the IFA 15 and supplied to the first signal input of the demodulator 16, where it is demodulated, and the frequency tuning speed is determined based on the control signal coming from the eighth output of the control unit 10. Moreover, the accuracy of the demodulation is determined by the accuracy of synchronization of clock pulses from the GTI 8 and the signal from the UPCH 15, determined on the basis of the signal from M3 25.

From the output of the demodulator 16, information slots are fed to the signal input D2 17, to the control input of which a signal is supplied from the sixth output of the control unit 10, based on which the demodulated signal in D2 17 is divided into M channels, the signals of each of which from the corresponding M outputs are fed to the signal inputs corresponding M decoders 18 ₁ -18 _M , where the error-correcting code of the slot information bits is decoded.

From the outputs of blocks 18 ₁ -18 _{M, the} decoded information slots are fed to the corresponding signal inputs M2 27, in which, based on the control signal coming from the seventh output of the control unit 10, M decoded signals are multiplexed into one decoded signal, which is output from the output of M2 27 to depacketator 19, where the binary information sequence is combined, which is then fed to PI 20, where it is converted into the output signal form.

In addition, the decoding results determined by the decoders 18 ₁ -18 _M are supplied respectively to the M signal inputs of AKS1 11, to the (M + 1) th signal input of which demodulated information slots from the output of the demodulator 16 are supplied. In the AKS 11, the quality of information channel, the result of which is fed to the first input of the control unit 10.

BU 10 generates control signals based on information supplied to its first and second inputs from the corresponding ACS 1 11 and ACS2 28.

GTI 8 delivers clock pulses to the corresponding blocks, which determine the beginning of each microoperation, and synchronizes the operation of the device as a whole.

The implementation of the blocks of this device can be carried out using digital devices known from the technical literature.

Thus, a device for implementing the proposed method for adaptive data transmission in radio links with pseudo-random tuning of the operating frequency allows a significant increase in the reliability of information reception under the influence of intentional interference.

Information sources

1. RF patent for the invention No. 2185029 "Radio line with pseudo-random tuning of the operating frequency", Odoevsky S. M., Eryshov V. G., 2001.

2. RF application for invention No. 2001102653 "Method and device for pseudo-random tuning of the operating frequency", Postnikov VA, Shubenkin VV, 2001.

3. RF patent for invention No. 2097923 "A method for transmitting discrete information in a radio line with pseudo-random tuning of operating frequencies and a device implementing it" Bulychev O.A .; Ignatov V.V .; Schukin A.N., 1997.

4. Noise-like signals in information transmission systems. Ed. prof. V. B. Pestryakova. M., "Sov. Radio", 1973.

5. V.I. Tikhonov. Statistical radio engineering. M., "Sov. Radio", 1966.

6. Feer. K. Wireless digital communications. Modulation and spreading methods: Transl. From English. / ed. V.I. Zhuravleva. - M., Radio and Communications, 2000.

Claims (2)

1. A method for adaptive data transmission in a radio line with pseudo-random tuning of the operating frequency, including dividing the input digital signal at speed C at information blocks, emitting a modulated signal into space, at the receiving end, combining the demodulated blocks into the original signal and supplying the receiver with information, characterized in that at the transmitting end of the radio line, the information speed C is converted to the technical speed V of the digital signal with the allocation of information packets duration

seconds, consisting of g service frames for exchanging information about the size of the generated packets and the quality of the received signals at the reception and transmission, and

information frames of size Ψ bits, taking into account a priori known: the minimum required number of bits for service frames e, the time spent by the equipment on the frequency tuning, T _n and the frequency desynchronization time for receiving and transmitting T _p , and g is a function of T, e and Ψ , and each information frame has the size d + δ slots, where d is the number of service slots of the radio line equal to the a priori known number of repetitions of service slots, and δ is the number of information slots, each service slot has a size of y + L + z bits, and each information one slot - size y + x + q bits, where

- bits for frequency hopping, L - bits of the sync sequence,

- bits of the information block,
z = [L · λ ₁ ] and q = [x · λ ₂ ] are the protection bits, λ ₁ and λ ₂ are the coefficients depending on the number of frequency hopping frequencies for the synchronization channel and the information channel, respectively, and a priori known signal propagation conditions, moreover, L corresponds to the length of the available synchronization sequence that corresponds to the minimum average risk calculated on the basis of determining the probability of an error on the service bit in the received synchronization sequence obtained on the basis of determining the quality of a given synchronization sequence, and ix error probability control bit is received in the service frames, the values of L, z, x, q, d, δ, g , h, λ _1, λ ₂ for forming the next packet is transmitted the receiving side in service frames, modulation information slots at the respective frequencies transmission in accordance with the first pseudorandom sequence (SRP), modulation of the service slots at the corresponding transmission frequencies in accordance with the second SRP, the error probability per information bit received at the receiving side is also received in the service frames based on determining the quality of the information frame; at the receiving end of the radio line, the received signal is divided into service and information parts, where for the service part, using the frequency filtering at the corresponding reception frequencies, in accordance with the second SRP, a channel is selected for transmission on the basis of service information received in the service frames, form a clock signal of size L and the peaks of the cross-correlation function are formed on it for accurate clock synchronization during subsequent demodulation of the signal, and according to the received sync the sequence performs its quality analysis with the determination of the probability of an error on the service bit, which is transmitted in the service frames to the transmitting side, for the information part, the service frames and information blocks are demodulated at the corresponding reception frequencies, in accordance with the first SRP, taking into account the obtained accuracy of clock synchronization and known values of z, x, q, d, δ, g, h, λ ₁ , λ ₂ , then obtained after demodulating service frames the values z, x, q, d, δ, g, h, λ ₁ , λ ₂ are used for demodulation next package, pr than from the received information frame analysis produce quality information with the determination error probability on the information bits, which are transmitted in the transmitting side service frames. 2. The method according to claim 1, characterized in that at the transmitting end of the radio line, based on the probability of error per information bit, obtained on the basis of determining the quality of the information frame, the probabilities of the correct reception of W ₁ , ..., W _{M of} each of the M possible methods of error-correcting coding are calculated , which are used in this radio, and the probability W of the correct reception of an uncoded information frame, where the value of the error probability per information bit is taken in service frames, and if W is less than necessary imogo priori known level search is conducted of the error-correcting code, in which W ₁ (i∈ {1, 2, ..., M}) is greater than W and obtained based on the known x, δ number of its codeword

where n _i is the length of the code, there is an integer, and the information blocks of the information frame are encoded by the found code if the decrease in the information transfer rate caused by the application of this code in

time,
where s _i (i∈ {1, 2, ..., M}) is the number of information symbols of the code, will lead to an increase in the quality of information reception by at least µ times and

where is the relation

- an a priori known value, determined based on the tasks of information theory, the number i∈ {0, 1, 2, ..., M} of the used code is transmitted to the receiving side in service frames, where i = 0 corresponds to the case of non-encoded information blocks, at the receiving end of the radio line decode the decoder with the number i∈ {0, 1, 2, ..., M} of the information blocks of the information frame, decode the received information from the received and decoded information blocks, and analyze the quality of the received information, based on which the probability of error per information bit is determined, which is transmitted in the service frames to the transmitting side, and the value i∈ {0, 1, 2, ..., M} obtained after demodulation of the service frames is used to decode the next packet.

Images