A method and device for iterative reception of covert communication waveforms based on orthogonal time-frequency space

Abstract

The present invention relates to the fields of communications and digital signal processing, and discloses a method and apparatus for iterative reception of covert communication waveforms based on orthogonal time-frequency space. For the multi-domain joint covert communication waveforms generated by the transmitter, a semi-independent parallel iterative reception detection algorithm based on the Turbo iterative reception architecture is employed. Specifically, the conventional signal and the covert signal share the same linear multi-carrier equalization module during reception detection, and are divided into two parallel iterative loops for simultaneous and parallel iterative processing. This enables iterative reception detection of the transmitter's covert communication waveform after it passes through a highly dynamic channel. Compared to existing technologies, this invention further reduces the probability of the signal waveform being detected and analyzed by non-cooperative parties while ensuring normal cooperative reception, thereby improving the security of the communication waveform.

Description

Technical Field

[0001] The present invention relates to the field of communications, and in particular to a method and device for iteratively receiving covert communication waveforms based on orthogonal time-frequency-space. Background Art

[0002] Orthogonal Time-Frequency-Space (OTFS) technology is a multi-carrier modulation and demodulation technology that enables signals or data symbols to be converted between the delay-Doppler domain and the time-frequency domain. Its most significant feature is that it utilizes the relative stability of the delay-Doppler domain of the communication physical channel, fundamentally improving the problem of traditional time-frequency domain-based waveforms facing rapid changes in time-frequency dual-selective channels in highly dynamic and fast-changing communication scenarios, making it difficult to perform effective channel estimation and equalization. In addition, OTFS technology has strong compatibility and can be applied to the general signal processing framework under current communication systems. Its precoding unit can be added after the modulator of a single / multi-carrier waveform, and its corresponding decoding processing unit is added before the corresponding receiving-end demodulator, giving it a strong ability to be cascaded with other signal waveforms.

[0003] Covert communication technology based on integrated spread spectrum technology still dominates current physical layer signal concealment techniques. Existing spread spectrum covert communication technologies primarily focus on the time, frequency, code, and power domains. In covert communication scenarios, the use of iterative reception algorithms can further reduce the signal-to-noise ratio required for normal cooperative reception and detection, further reducing the likelihood of covert communication signals being intercepted and analyzed by non-cooperative interceptors, and improving the security of communication waveforms. Summary of the Invention

[0004] The purpose of the present invention is to further improve the anti-interception and anti-detection performance of the covert communication waveform, or to further improve the normal cooperative reception and detection performance of the signal, and to provide a covert communication waveform iterative reception method and device based on orthogonal time-frequency space.

[0005] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0006] A method for iterative reception of covert communication waveforms based on orthogonal time-frequency space effectively improves the reception and detection performance of communication waveforms without affecting the waveform concealment performance and dynamic adaptability. The method includes iterative reception and detection of baseband signals at the receiving end. The specific process is as follows:

[0007] S1. Perform timing synchronization and channel estimation on the baseband signal, perform matched filtering and orthogonal time-frequency-space demodulation on the synchronized baseband signal to obtain a multi-carrier demodulated signal, and obtain an initial priori mean vector and an initial priori covariance matrix;

[0008] S2, perform OTFS channel equalization according to the LMMSE criterion to obtain the external information mean vector and variance of the multi-carrier equalized signal, and transfer them to S3 and S6 respectively;

[0009] S3, performing fixed sequence soft demodulation on the external information mean vector and variance of the multi-carrier equalized signal to obtain the mean vector and variance of the single-carrier demodulated conventional signal, performing discrete signal estimation based on the prior Gaussian noise assumption and the constellation mapping modulation mode of the conventional signal, obtaining the posterior mean vector and variance of the conventional signal reception information, and calculating the absolute difference in variance between the result of the previous iteration, if the absolute difference in variance is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S4, otherwise go to S5;

[0010] S4, the conventional signal iterative reception detection is terminated, hard demodulation or hard decision is performed on the posterior mean vector of the current conventional signal reception information, and an estimation result of the conventional signal reception information and an estimation result of the corresponding random code hopping control information are output;

[0011] S5. Based on the a posteriori mean vector and variance of the conventional signal reception information and the mean vector and variance of the single-carrier demodulated conventional signal, the mean vector and variance of the discrete signal external information are calculated and fixed sequence soft spreading is performed, the mean vector and variance after soft spreading are output, and the a priori mean vector and a priori covariance matrix of the multi-carrier equalized conventional signal in S2 are updated, and then the process is transferred to S2.

[0012] S6, performing random code hopping soft despreading on the external information mean vector and variance of the multi-carrier equalized signal to obtain the mean vector and variance of the single-carrier demodulated concealed signal, calculating the a priori log-likelihood ratio of the concealed signal reception information based on the concealed signal interleaving coding mode and the prior Gaussian noise assumption, obtaining the posterior log-likelihood ratio vector and the corresponding variance through LDPC iterative soft decoding, and calculating the absolute difference in variance between the variance of the posterior log-likelihood ratio vector and the previous iteration result. If the absolute difference in variance is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S7; otherwise, go to S8;

[0013] S7, the iterative reception detection of the concealed signal is terminated, LDPC decoding is performed on the posterior log-likelihood ratio vector of the current concealed signal reception information, and an estimation result of the concealed signal reception information is output;

[0014] S8. Based on the posterior log-likelihood ratio vector and variance, calculate the external information mean vector and variance of LDPC soft decoding, perform row-column interleaving on the external information mean vector, perform random code hopping soft spread based on the random code hopping control information estimation result, output the mean vector and variance after soft spread, and update the prior mean vector and prior covariance matrix of the multi-carrier equalized concealed signal in S2, and go to S2.

[0015] As a preferred solution of the present invention, the orthogonal time-frequency-space demodulation in step S1 includes vector-matrix transformation, matched filtering and Wigner transformation, symplectic-finite Fourier transformation and matrix-vector transformation performed in sequence.

[0016] As a preferred embodiment of the present invention, step S2 specifically includes the following process:

[0017] S21, initializing or updating the prior mean and covariance matrix of the multi-carrier equalized signal;

[0018] S22. Perform OTFS channel equalization according to the prior mean and covariance matrix of the multi-carrier equalized signal and the LMMSE criterion to obtain the posterior mean, covariance matrix and variance of the multi-carrier equalized signal;

[0019] S23. Calculate and output the extrinsic information mean and variance of the multi-carrier equalized signal based on the priori mean, the posterior mean, the covariance matrix, and the variance of the multi-carrier equalized signal.

[0020] As a preferred embodiment of the present invention, step S3 specifically includes the following process:

[0021] S31, performing fixed sequence soft despreading on the external information of the multi-carrier equalized signal to obtain the mean vector and variance of the single-carrier demodulated conventional signal;

[0022] S32. Perform discrete signal estimation based on the prior Gaussian noise hypothesis and the constellation mapping modulation mode of the conventional signal, and calculate the posterior mean vector and variance of the conventional signal reception information;

[0023] S33. Calculate the posterior mean vector and variance of the conventional signal reception information and the absolute difference in variance between the results of the previous iteration. If the absolute difference in variance is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S4; otherwise, go to S5.

[0024] The regular signal reception information is specifically regular signal reception symbols or bits.

[0025] As a preferred embodiment of the present invention, step S5 specifically includes the following process:

[0026] S51, using the mean vector and variance of the single-carrier demodulated conventional signal as the equivalent priori mean vector and variance of the conventional signal reception information, and calculating the mean vector and variance of the discrete signal estimated extrinsic information based on the posterior mean vector and variance of the conventional signal reception information;

[0027] S52, performing fixed sequence soft spreading on the estimated external information of the discrete signal to obtain a mean vector and variance after the soft spreading;

[0028] S53 . Update the a priori mean vector and a priori covariance matrix of the multi-carrier equalized conventional signal in S21 according to the mean vector and variance after soft spreading, and go to S2 .

[0029] As a preferred embodiment of the present invention, step S6 specifically includes the following process:

[0030] S61. Based on the random code hopping control information estimation result obtained by iterative reception detection of the conventional signal, the extrinsic information mean vector and variance of the multi-carrier equalized signal are subjected to random code hopping soft despreading to obtain the mean vector and variance of the single-carrier demodulated concealed signal;

[0031] S62. Deinterleave the single-carrier demodulated concealed signal based on the concealed signal interleaving method to obtain a mean vector and variance after deinterleaving, and obtain a priori log-likelihood ratio of received coded bits of the concealed signal based on the prior Gaussian noise hypothesis data and the constellation mapping method of the concealed signal;

[0032] S63, performing LDPC iterative soft decoding on the a priori log-likelihood ratio of the concealed signal received coded bits, obtaining an LDPC decoded soft bit vector of equal length as a posterior log-likelihood ratio vector, and calculating the corresponding variance;

[0033] S64. Calculate the absolute difference between the variance of the posterior log-likelihood ratio vector and the variance of the last iteration result. If the absolute difference is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S7; otherwise, go to S8.

[0034] As a preferred solution of the present invention, step S8 specifically includes the following process:

[0035] S81, calculating the mean vector and variance of the extrinsic information of the LDPC iterative soft decoding based on the a priori log-likelihood ratio and the a posteriori log-likelihood ratio vector of the concealed signal reception information;

[0036] S82, interleave the LDPC iterative soft decoding external information to obtain an interleaved mean vector and variance;

[0037] S83, performing random code hopping soft despreading on the interleaved mean vector and variance to obtain the mean vector and variance after soft spectrum spreading;

[0038] S84. Update the priori mean vector and priori covariance matrix of the multi-carrier equalized concealed signal in S21 according to the mean vector and variance after soft spreading, and go to S2.

[0039] The concealed signal reception information is specifically concealed signal reception decoding bits.

[0040] A device for iteratively receiving a covert communication waveform based on orthogonal time-frequency-space is used at a receiving end and, when executed, implements the method for iteratively receiving a covert communication waveform based on orthogonal time-frequency-space. The device comprises: a radio frequency processing module, an analog-to-digital converter, a timing synchronization module, a channel estimation module, an orthogonal time-frequency-space demodulation module, an iterative linear channel equalization module, a conventional signal soft despreading module, a conventional signal discrete estimation module, a conventional signal soft spread spectrum module, a conventional signal hard demodulation / hard decision module, a covert signal code hopping despreading domain deinterleaving module, an LDPC iterative soft decoding module, a covert signal re-interleaving and code hopping spread spectrum module, and an LDPC decoding module.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention adopts a simultaneous and co-frequency differentiated power combining design for conventional signals and concealed signals, performs iterative channel equalization according to the LMMSE criterion, and obtains the external information mean vector and variance of the multi-carrier equalized signal. Since the single-carrier modulation modes of the conventional signal and the concealed signal are different, the external information mean vector and variance of the multi-carrier equalized signal are divided into two paths for further reception processing. One path is used for iterative detection of the conventional signal, and the conventional signal bits thereon are obtained through steady despreading and discrete signal estimation. It is also the control information of the random code hopping direct spread of the concealed signal to assist the iterative reception of the concealed signal. The other path is used for iterative detection of the concealed signal, and the concealed data bits are obtained through iterative information updating of random code hopping despreading, deinterleaving and LDPC decoding, thereby realizing signal reception detection at a lower signal-to-noise ratio. The present invention fully considers the needs of low signal-to-noise ratio signal reception and detection in covert communication scenarios. By iteratively detecting conventional signals and covert signals respectively, the probability of the signal being detected and analyzed by non-cooperative intercepting parties is further reduced while ensuring that cooperative reception can be received normally, effectively improving the reception and detection performance of the communication waveform and further enhancing the security of the communication signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a schematic diagram of the process of Example 1 of the present invention;

[0044] Figure 2 This is a signal processing diagram of Example 1 of the present invention;

[0045] Figure 3 This is a simulation result diagram of Example 1 of the present invention;

[0046] Figure 4 This is a partially enlarged schematic diagram of the simulation results of Example 1 of the present invention;

[0047] Figure 5 This is a schematic diagram of a process for generating a baseband signal at a transmitting end in Embodiment 1 of the present invention;

[0048] Figure 6 This is a schematic diagram of signal processing for generating a transmitting end baseband signal in Embodiment 1 of the present invention;

[0049] Figure 7 This is a structural diagram of a receiving detection device according to embodiment 2 of the present invention. DETAILED DESCRIPTION

[0050] The present invention will be further described in detail below in conjunction with test examples and specific embodiments. However, this should not be understood as limiting the scope of the present invention to the following embodiments, and all technologies implemented based on the present invention fall within the scope of the present invention.

[0051] Before describing the present invention in detail, the application prerequisites of the present invention are first given as follows:

[0052] (1) The present invention is limited to the reception and detection of physical layer baseband signals. The mode adaptation and stream adaptation of data and the radio frequency processing of signals are not considered within the scope of the present invention. At the transmitting end, it is assumed that the input data has been adapted, and the output signal will be sent after appropriate radio frequency processing; at the receiving end, it is assumed that the input data has been appropriately radio frequency processed, and the output signal will undergo necessary de-adaptation processing; (2) The power of the non-stationary concealed signal cannot exceed the power of the covert signal used to mask its signal characteristics; (3) The spreading factor of the conventional signal is smaller than the spreading factor of the covert signal to facilitate the transmission of non-stationary control information; (4) The conventional signal used to mask its signal characteristics is a cooperative signal, which can be jointly iteratively detected with the covert signal.

[0053] Example 1

[0054] like Figure 1 and Figure 2 As shown in FIG, a covert communication waveform iterative reception method based on orthogonal time-frequency space further improves the concealment performance of the communication waveform without affecting the waveform reception detection performance and dynamic adaptability. The method includes iterative reception and detection of the baseband signal at the receiving end. The specific process is as follows:

[0055] S1. Perform timing synchronization and channel estimation on the baseband signal, perform matched filtering and orthogonal time-frequency-space demodulation on the synchronized baseband signal to obtain a multi-carrier demodulated signal, and obtain an initial priori mean vector and an initial priori covariance matrix;

[0056] S2, perform OTFS channel equalization according to the LMMSE criterion to obtain the external information mean vector and variance of the multi-carrier equalized signal, and transfer them to S3 and S6 respectively;

[0057] S3. Perform fixed sequence soft demodulation on the external information mean vector and variance of the multi-carrier equalized signal to obtain the mean vector and variance of the single-carrier demodulated conventional signal. Based on the prior Gaussian noise assumption and the constellation mapping modulation mode of the conventional signal, perform discrete signal estimation to obtain the posterior mean vector and variance of the conventional signal reception information, and calculate the absolute difference in variance between the result of the previous iteration. If the absolute difference in variance is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S4; otherwise, go to S5.

[0058] S4, the conventional signal iterative reception detection is terminated, hard demodulation or hard decision is performed on the posterior mean vector of the current conventional signal reception information, and an estimation result of the conventional signal reception information and an estimation result of the corresponding random code hopping control information are output;

[0059] S5. Based on the a posteriori mean vector and variance of the conventional signal reception information and the mean vector and variance of the single-carrier demodulated conventional signal, the mean vector and variance of the discrete signal external information are calculated and fixed sequence soft spreading is performed, the mean vector and variance after soft spreading are output, and the a priori mean vector and a priori covariance matrix of the multi-carrier equalized conventional signal in S2 are updated, and then the process is transferred to S2.

[0060] S6. Perform random code hopping soft despreading on the external information mean vector and variance of the multi-carrier equalized signal to obtain the mean vector and variance of the single-carrier demodulated concealed signal. Based on the concealed signal interleaving coding method and the prior Gaussian noise assumption, calculate the prior log-likelihood ratio of the concealed signal reception information. Perform LDPC iterative soft decoding to obtain the posterior log-likelihood ratio vector and the corresponding variance. Calculate the absolute difference between the variance of the posterior log-likelihood ratio vector and the variance of the previous iteration result. If the absolute difference in variance is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S7; otherwise, go to S8.

[0061] S7, the iterative reception detection of the concealed signal is terminated, LDPC decoding is performed on the posterior log-likelihood ratio vector of the current concealed signal reception information, and an estimation result of the concealed signal reception information is output;

[0062] S8. Based on the posterior log-likelihood ratio vector and variance, calculate the external information mean vector and variance of LDPC soft decoding, perform row-column interleaving on the external information mean vector, perform random code hopping soft spread based on the random code hopping control information estimation result, and output the mean vector and variance after soft spread, and update the prior mean vector and prior covariance matrix of the multi-carrier equalized concealed signal in S2, and go to S2.

[0063] like Figure 5 and Figure 6 As shown in FIG, the generation of the transmitting end baseband signal includes the following processes:

[0064] S01, LDPC encoding and row-column interleaving are performed on the hidden data bits to be transmitted to obtain a hidden signal coding bit sequence with a coding rate of R y ;

[0065] S02. Perform random code hopping spread spectrum on the concealed coded bit sequence to obtain a concealed signal code hopping spread spectrum sequence; and perform fixed sequence spread spectrum on the bit sequence of the conventional signal corresponding to the concealed coded bit to be transmitted to obtain a conventional signal fixed spread spectrum sequence;

[0066] S03, performing BPSK modulation on the concealed signal code hopping spread spectrum sequence and the conventional signal fixed spread spectrum sequence respectively, and then performing unequal power combination to generate a combined symbol stream;

[0067] S04. Perform orthogonal time-frequency-space transformation and shaping filtering on the combined symbol stream to generate a baseband signal.

[0068] The expression of random code hopping spread spectrum in step S02 is:

[0069]

[0070] Among them, x y [n] represents the hidden signal code hopping spread spectrum symbol sequence, d y [n] represents the hidden signal encoding bit sequence, p y [n] represents the pseudo-random sequence of random code hopping spread spectrum, p y The bit rate of [n] is d y [n]k y times, k y is the spreading factor of the concealed signal, bpskmod{·} is the BPSK modulation operator;

[0071] The expression of fixed sequence spread spectrum is:

[0072]

[0073] Among them, x d [n] represents the fixed spread spectrum symbol sequence of the conventional signal, d d [n] represents the bit sequence of the conventional signal corresponding to the coded bits of the concealed signal to be transmitted, p d [n] represents the pseudo-random sequence of fixed sequence spread spectrum, p d The bit rate of [n] is d d [n]k d times, k d is the spreading factor of the conventional signal.

[0074] In step S02, the principle of conventional signal control random code hopping spread spectrum is as follows: Assume that the system transmits N y hidden data encoding bits and Nd Normal data bits. Since the present embodiment needs to consider the concealment of the signal, the code rate of the concealed signal must be equal to the code rate of the normal signal, that is, N y ·k y =N d ·k d , and let k y =k·k d (k>1, and k is an integer). When one hidden data coding bit is transmitted, k regular data bits are transmitted. These k regular data bits are called a regular bit group in this embodiment. There are 2 k Different situations, and 2 in the code hopping spread spectrum sequence library / set k The hidden signal selects which spreading sequence to use to spread the hidden coded bits at the corresponding position based on the one-to-one correspondence between the different spreading sequences.

[0075] The expression for unequal power combining in step S03 is:

[0076] x B [n]=A d x d [n]+A y x y [n]

[0077] Among them, x B [n] represents the combined symbol stream, and its combined power ratio is A d represents the amplitude of the regular symbol sequence, A y Represents the amplitude of the hidden symbol sequence.

[0078] Step S04 includes the following process:

[0079] S041, the combined symbol stream x B [n], split into a set of vectors of length MN, one of which is a vector To illustrate, specifically, a complex column vector containing MN symbols is placed on an M×N dimensional delay-Doppler domain complex matrix through vector & matrix transformation, that is, X dd ==vec -1 (x dd ),

[0080] S042. Convert the combined symbol stream from the delay-Doppler domain to the time-frequency domain through the inverse symplectic-finite Fourier transform (ISFFT). Its matrix expression is as follows:

[0081]

[0082] in, F is the matrix that transforms the signal from the delay-Doppler domain to the time-frequency domain. M and F N are the normalized FFT matrices of M and N dimensions respectively, F N Conjugate transpose of a matrix;

[0083] S043. The two-dimensional signal in the time-frequency domain is transformed into a one-dimensional signal in the time domain through Heisenberg transform, and then the transmitter shaping filter is performed to obtain a time domain matrix signal. The matrix expression is as follows:

[0084]

[0085] in, G is the matrix that transforms the signal from the time-frequency domain to the time domain. tx is the equivalent matrix of the shaping filter at the transmitting end, F M Conjugate transpose of a matrix;

[0086] S044. Perform matrix-vector transformation on the time domain matrix signal to generate a baseband transmit signal. The specific expression is as follows:

[0087]

[0088] in, is the vector obtained by discretizing the time domain baseband transmission signal, and Correspondingly, vec(·) is the matrix-to-vector conversion function. The covert communication waveform after OTFS modulation passes through the baseband equivalent high dynamic channel. The high dynamic channel is modeled in the delay-Doppler domain, and the expression of the channel h(τ,v) is:

[0089]

[0090] Where K represents the total number of multipaths, τ i is the delay value of the i-th path, v i is the Doppler frequency value of the i-th path, h i is the coefficient of the i-th multipath, δ(τ) and δ(v) are impulse functions.

[0091] The baseband received signal r(t) is obtained by superimposing the baseband transmitted signal on the channel with Gaussian white noise, and its expression is:

[0092]

[0093] Where w(t) is Gaussian white noise. After discretizing r(t), we can get:

[0094] r=Hs+w

[0095] in, is the vector obtained by discretizing the received signal in the time domain. The expression of the equivalent channel matrix of the channel in the time domain is as follows:

[0096]

[0097] Among them, l i and k i are the time delay and Doppler discretization grid position of the i-th path respectively.

[0098] The orthogonal time-frequency-space demodulation in step S1 includes vector-matrix transformation, matched filtering and Wigner transformation, symplectic-finite Fourier transformation and matrix-vector transformation performed in sequence, wherein:

[0099] Vector-matrix transformation: Convert the baseband signal into a time domain matrix signal. The expression is as follows:

[0100] R=vec -1 (r)

[0101] in, is the time domain received signal matrix;

[0102] Wigner transform: Performs matched filtering and Wigner transform on the time-domain matrix signal to generate a time-frequency domain matrix. The Wigner transform is the inverse transform of the Heisenberg transform. The principle expression is as follows:

[0103]

[0104] Where t is time, f is frequency, Δf is the bandwidth of the frequency domain sampling interval, T is the time domain sampling interval, n is the time / Doppler domain index of OTFS modulation (n=0,1,...,N-1), and m is the frequency / delay domain index of OTFS modulation (m=0,1,...,M-1). is the cross fuzzy function, and its expression is as follows:

[0105]

[0106] Among them, g rx is the time domain function of the receiving shaping filter at baseband, t' is the process parameter;

[0107] After the Wigner transformation is matrixed, its expression is as follows:

[0108] Y tf =F M G rx R

[0109] in, G is the matrix that transforms the signal from the time domain to the time-frequency domain. rx is the equivalent matrix of the shaping filter at the receiving end, and R is the time domain received signal matrix;

[0110] Symplectic Fourier Transform (SFFT): Perform a symplectic Fourier transform on the time-frequency domain matrix to obtain the delay-Doppler domain matrix. The principle formula of SFFT is as follows:

[0111]

[0112] Where y[l,k] is the (l,k)th element of the signal in the delay-Doppler domain matrix, n = 0, 1, ..., N-1, m = 0, 1, ..., M-1;

[0113] After SFFT is matrixed, the expression of SFFT algorithm is:

[0114]

[0115] in, is the matrix that transforms the signal from the time-frequency domain to the delay-Doppler domain;

[0116] Matrix-vector transformation: Perform matrix-vector transformation on the delay-Doppler domain matrix to generate the OTFS demodulated signal. Finally, matrix and vector transformation are performed at the receiving end to obtain the OTFS demodulated signal:

[0117] y dd =vec(Y dd )

[0118] Among them, y dd is the demodulated signal of OTFS.

[0119] Due to the characteristics of OTFS modulation technology and the quasi-time-invariant characteristics of highly dynamic time-varying channels in the delay-Doppler domain, based on the discretization of the baseband signal and according to the properties of the vec(·) operator and the Kronecker basis, the transformation relationship between the transmit and receive signals in the delay-Doppler domain can be approximated as follows:

[0120] y dd =H eff x dd +w eff

[0121]

[0122] in, is the equivalent delay-Doppler domain channel state matrix of the channel, is the equivalent delay-Doppler domain noise vector of the channel. The core of OTFS channel estimation at the receiving end is to obtain the channel state matrix estimation result in the delay-Doppler domain, that is, E is the channel estimation error matrix.

[0123] Step S2 specifically includes the following process:

[0124] S21, initializing or updating the prior mean and covariance matrix of the multi-carrier equalized signal;

[0125] S22. Perform OTFS channel equalization according to the prior mean and covariance matrix of the multi-carrier equalized signal and the LMMSE criterion to obtain the posterior mean, covariance matrix and variance of the multi-carrier equalized signal;

[0126] S23. Calculate and output the extrinsic information mean and variance of the multi-carrier equalized signal based on the priori mean, the posterior mean, the covariance matrix, and the variance of the multi-carrier equalized signal.

[0127] In step S21, before iterative reception detection, an initial value is given to the priori mean vector and covariance matrix of the iterative linear signal equalization module as the priori statistical characteristics of the received combined symbol; the conventional signal and the hidden signal symbol of the transmitting end can be regarded as zero-mean random signals that obey approximately independent and identical distribution and the channel noise is Gaussian white noise, so the prior assumption is that the multi-carrier equalization signal obeys the mean The covariance matrix is (I represents the unit matrix) complex Gaussian random process, where It should not be too small to reduce the negative impact on the iterative detection convergence time caused by the large difference between the prior distribution characteristics and the actual characteristics;

[0128] During the iterative reception detection operation, update the and The expression is as follows:

[0129]

[0130] in, and It is the multi-carrier equalization prior update information from the regular signal iterative reception feedback, and The multi-carrier equalization prior information is updated from the iterative reception feedback of the concealed signal.

[0131] In step S22, the multi-carrier demodulated signal vector y dd , the equivalent delay-Doppler domain channel state matrix obtained by channel estimation Receive combined symbol prior distribution and receiver noise power σ2 As parameters, based on the LMMSE channel equalization criterion, iterative linear signal equalization is performed, and the posterior mean of the combined symbol vector and covariance matrix The expression is as follows:

[0132]

[0133] Since the symbols of the conventional signal and the hidden signal at the transmitter can be regarded as approximately independent and identically distributed, the non-diagonal elements of the posterior covariance matrix can be ignored, and the posterior variance is obtained as follows:

[0134]

[0135] Where tr{·} is the matrix trace operation.

[0136] In step S23, the extrinsic information of the multi-carrier equalized signal is additional information obtained during the channel equalization process that is a priori independent of the multi-carrier demodulated signal and the multi-carrier equalized signal, and satisfies the following formula:

[0137]

[0138] Among them, CN(·) represents the complex Gaussian distribution, and the variance and mean of the output external information are:

[0139]

[0140] in, Considered as x dd Additive complex Gaussian noise measurement of, is x dd The observed quantity is:

[0141]

[0142] Step S3 specifically includes the following process:

[0143] S31, performing fixed sequence soft despreading on the external information of the multi-carrier equalized signal to obtain the mean vector and variance of the single-carrier demodulated conventional signal;

[0144] S32. Based on the prior Gaussian noise assumption and the constellation mapping modulation mode of the conventional signal, discrete signal estimation is performed to calculate the posterior mean vector and variance of the conventional signal reception information;

[0145] S33. Calculate the posterior mean vector and variance of the conventional signal reception information and the absolute difference in variance between the results of the previous iteration. If the absolute difference in variance is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S4; otherwise, go to S5.

[0146] The regular signal reception information is specifically regular signal reception symbols or bits.

[0147] In step S31, fixed sequence soft despreading is performed on the external information of the multi-carrier equalized signal. According to the properties of the independent random process and the soft despreading operation, the expressions for the mean and variance of the single-carrier demodulated conventional signal after despreading are:

[0148]

[0149] in, is a bipolar fixed despread pseudo-random sequence, By Q d indivual Vector transformed by rearranging elements MNQ d Should be MN and k y The least common multiple of L, where L is the length of the LDPC code block, is the mean vector of the single-carrier demodulated conventional signal, is the variance of the elements of this vector.

[0150] In step S32, based on the prior Gaussian noise assumption, discrete signal estimation is performed to obtain the a posteriori mean vector and variance of the conventional signal reception symbol / bit. dp ) i ∈A, so its prior distribution is expressed as:

[0151] P[(x dp ) i ]=∑ a∈A P[(x dp ) i =a]δ[(x dp ) i -a],a∈A

[0152] Among them, (x dp ) i is x dp The i-th element in the vector, x dp For The corresponding actual noise-free value at the transmitting end, A is the constellation point set of the transmitting end baseband modulation, taking standard BPSK modulation as an example: A = {-1 + 0j, 1 + 0j}, P[·] is the probability operation;

[0153] Combined with the external information output by S31, the following formula is obtained:

[0154]

[0155] Where c is a constant related to the baseband modulation mode of the transmitter. Taking the standard BPSK modulation as an example, c = 1.dp ) i =a] into P[(x dp ) i ], and obtain the posterior mean of each element of the conventional signal received symbol / bit vector and variance Expressed as:

[0156]

[0157] Furthermore, the conventional signal received symbol / bit mean vector is variance is a vector The mean of each element in .

[0158] In step S33, according to the parameters of the transmitted waveform, the conventional signal iterative detection variance convergence threshold of the receiving algorithm is pre-set and conventional signal iterative detection threshold And in this embodiment, Th d-iter ≥2, used to determine whether the iterative algorithm should make convergence and iteration termination decisions.

[0159] Only when the regular signal receives the posterior variance of the symbol / bit and the posterior variance of the previous iteration The absolute difference is less than That is, the conventional signal iterative reception detection algorithm converges approximately to the current conventional signal reception symbol / bit mean When, or the number of iterations Num of the conventional signal iterative reception detection algorithm d Equal to the threshold Th d-iter Time (Num d =Th d-iter ), as in step S4, the conventional signal iterative detection is terminated, and the latest conventional signal reception symbol / bit mean vector result is used as the final conventional signal iterative reception result, and hard demodulation / hard decision is performed based on the result, and the estimation result of the conventional signal reception symbol / bit and the random hopping control information estimation result constituted by it are output; if the conventional signal iterative reception detection algorithm does not meet the above threshold constraint, then go to S5, and output the posterior mean vector and variance of the conventional signal reception symbol / bit.

[0160] Step S5 specifically includes the following process:

[0161] S51, using the mean vector and variance of the single-carrier demodulated conventional signal as the equivalent priori mean vector and variance of the conventional signal reception information, and calculating the mean vector and variance of the discrete signal estimated extrinsic information based on the posterior mean vector and variance of the conventional signal reception information;

[0162] S52, performing fixed sequence soft spreading on the estimated external information of the discrete signal to obtain a mean vector and variance after the soft spreading;

[0163] S53 . Update the a priori mean vector and a priori covariance matrix of the multi-carrier equalized conventional signal in S21 according to the mean vector and variance after soft spreading, and go to S2 .

[0164] In step S51, the posterior mean vector and variance of the received symbols / bits based on the conventional signal are and And the mean vector and variance of the single carrier demodulated conventional signal and As the priori mean vector and variance of the equivalent conventional signal received symbol / bit, analogously to step S23, the mean vector and variance expressions of the discrete signal estimated extrinsic information are obtained as follows:

[0165]

[0166] In step S52, fixed sequence soft spreading is performed on the estimated external information of the discrete signal. According to the properties of the independent random process and the soft spreading operation, the expressions of the mean vector and variance after the soft spreading are calculated as follows:

[0167]

[0168] in, is the Kronecker product operator, truc (MN×1) {·} is a vector / matrix interception operation, which means intercepting a MN×1-dimensional vector from a MNQ×1-dimensional vector. Vector split into Q d MN×1 dimensional With Q in S31 d indivual Vectors have one-to-one correspondence.

[0169] Step S6 specifically includes the following process:

[0170] S61. Based on the random code hopping control information estimation result obtained by iterative reception detection of the conventional signal, the external information mean vector and variance of the multi-carrier equalized signal are subjected to random code hopping soft despreading to obtain the mean vector and variance of the single-carrier demodulated concealed signal;

[0171] S62. Deinterleave the single-carrier demodulated concealed signal based on the concealed signal interleaving method to obtain a mean vector and variance after deinterleaving, and obtain a priori log-likelihood ratio of received coded bits of the concealed signal based on the prior Gaussian noise hypothesis data and the constellation mapping method of the concealed signal;

[0172] S63, performing LDPC iterative soft decoding on the a priori log-likelihood ratio of the concealed signal received coded bits, obtaining an LDPC decoded soft bit vector of equal length as a posterior log-likelihood ratio vector, and calculating the corresponding variance;

[0173] S64. Calculate the absolute difference between the variance of the posterior log-likelihood ratio vector and the variance of the last iteration result. If the absolute difference is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S7; otherwise, go to S8.

[0174] In step S61, random code hopping soft despreading is performed on the external information of the multi-carrier equalized signal. Based on the properties of the independent random process and the soft despreading operation, the mean and variance of the single-carrier demodulated concealed signal after despreading are calculated as follows:

[0175]

[0176] in, It means taking the diagonal elements of the matrix and rearranging it into (MNQ y / k y )×1 column vector, is the bipolar random code hopping despreading pseudo-random matrix, By Q y indivual The vector is transformed by rearranging the elements. MNQ y For MN and k y The least common multiple of L, where L is the length of the LDPC code block, is the mean vector of the single-carrier demodulated concealed signal, is the variance of the elements of this vector.

[0177] In step S62, the mean vector and variance after deinterleaving are:

[0178]

[0179] Among them, Π -1 is the deinterleaving operator, is the mean vector after deinterleaving, is the variance after deinterleaving (the deinterleaving operation does not change the variance).

[0180] In step S63, according to the requirements of LDPC iterative soft decoding, the deinterleaved mean vector is first converted into the form of log-likelihood ratio (LLR). The corresponding prior distribution expression is:

[0181]

[0182] in, and Respectively and The elements in For The corresponding LDPC iterative soft decoding a priori LLR vector is soft-decoded using a typical LDPC soft decoding module adapted to the encoding method of the hidden data bits to be transmitted to perform soft decoding on the a priori LLR sequence and obtain the posterior LLR vector The sum of the variances is:

[0183]

[0184] Among them, LDPC S-de (·) represents the LDPC soft decoding operation, For The corresponding posterior mean vector, the conversion relationship between the two is:

[0185]

[0186] In step S64, the concealed signal iterative detection variance convergence threshold of the receiving algorithm is pre-set according to the parameters of the transmission waveform. and the threshold value of the number of iterative detection times of the hidden signal Th y-iter , used to determine whether the iterative algorithm should converge and terminate the iteration only when the posterior variance of the hidden signal decoding soft bit is and the posterior variance of the previous iteration The absolute interpolation value is less than That is, the concealed signal iterative reception detection algorithm converges approximately to the current concealed signal soft decoding bit mean When, or the number of iterations of the hidden signal iterative reception detection algorithm Num y Equal to the threshold Th y-iter Time (Num y =Th y-iter ), the concealed signal iterative detection is terminated, and the latest concealed signal soft-decoded bit mean vector result is used as the final concealed signal iterative reception result. LDPC decoding is then performed based on this result, and an estimated result of the received concealed data bits is output. If the concealed signal iterative reception detection algorithm does not meet the above threshold constraint, the concealed signal iterative reception detection continues. The LDPC decoding process can be expressed as:

[0187]

[0188] in, is the estimated result vector of the received hidden data bits after LDPC decoding,

[0189] Step S8 specifically includes the following process:

[0190] S81, calculating the mean vector and variance of the extrinsic information of the LDPC iterative soft decoding based on the a priori log-likelihood ratio and the a posteriori log-likelihood ratio vector of the concealed signal reception information;

[0191] S82, interleave the LDPC iterative soft decoding external information to obtain an interleaved mean vector and variance;

[0192] S83, performing random code hopping soft despreading on the interleaved mean vector and variance to obtain the mean vector and variance after soft spectrum spreading;

[0193] S84. Update the priori mean vector and priori covariance matrix of the multi-carrier equalized concealed signal in S21 according to the mean vector and variance after soft spreading, and go to S2.

[0194] In step S81, based on the a priori and a posteriori LLR vectors of the coded bits received by the concealed signal, the LLR vector expression of the LDPC iterative soft decoding external information is obtained according to the properties of the LLR:

[0195]

[0196] And transform the LLR back to the mean:

[0197]

[0198] At the same time, calculate the variance of the external information:

[0199]

[0200] Among them, the variance is a vector The mean of each element in , C = {-1, 1}.

[0201] In step S82, the LDPC iterative soft decoding external information is interleaved, and the mean vector and variance after interleaving are:

[0202]

[0203] Among them, π is the interleaving operator, is the mean vector after interleaving, is the variance after interleaving.

[0204] In step S83, random code hopping soft despreading is performed on the interleaved signal. According to the properties of the independent random process and the soft spreading operation, the expressions of the mean vector and variance after soft spreading are as follows:

[0205]

[0206] in, is a generalized diagonal extraction and rearrangement operation, which means This MNQ y ×Q y The matrix of Q y ×Q y The block diagonal matrix composed of MN×1 dimensional sub-matrices is extracted from the Q y MN×1 dimensional sub-matrices and arrange them sequentially into an MNQ y ×1-dimensional vector, truc (MN×1) {·} is a vector / matrix interception operation, which means intercepting a MN×1-dimensional vector from a MNQ×1-dimensional vector, which divides the generalized diagonal extraction and rearranged vector into Q y MN×1 dimensional With Q in S61 y indivual Vectors have one-to-one correspondence.

[0207] In specific implementation, the main parameters of the simulation are as follows:

[0208] (1) Covert signal coding: 1 / 3-LDPC code; (2) Covert signal interleaving: interleaver depth 48 bits; (3) Conventional signal spreading factor: k d =512; (4) Covert signal spreading factor: k y =4096; (5) Covert signal random code hopping spread spectrum code group size: 2 k =16; (6) Baseband spread spectrum code rate: 1.024Mchip / s; (7) Baseband constellation mapping method for conventional signal: BPSK; (8) Baseband constellation mapping method for covert signal: BPSK; (9) Baseband sampling frequency: 8.192MHz (8x upsampling); (10) Baseband shaping filter: square root raised cosine filter with a roll-off factor of 0.25; (11) Combined power ratio of conventional signal and covert signal: [A d 2 / A y 2 ] dB =7dB; (12) OTFS modulation matrix dimension: M×N=8×64; (13) Maximum Doppler frequency deviation: 460kHz; (14) Maximum multipath delay: 1us; (15) Channel multipath model: Strong Rice channel; (16) Multipath amplitude model: Exponential fading; (17) Channel Doppler model: Jakes model; (18) The energy leakage factor caused by the introduction of fractional delay Doppler was considered in the simulation, and both delay and Doppler showed quasi-sparse characteristics; (19) Simulation SNR variation range: [-40dB, 0dB]; (20) Conventional signal iterative detection number threshold Th d-vand the threshold value of the number of iterative detection times of the hidden signal Th y-v Both are 10 -4 ; (21) Conventional signal iterative detection times threshold Th d-iter and the threshold value of the number of iterative detection times of the hidden signal Th y-iter Both are 6.

[0209] The simulation results are as follows Figure 3 and 4 Under the above simulation parameters, compared with the typical non-iterative LMMSE algorithm, the iterative reception detection algorithm proposed in the present invention can improve the detection performance of conventional signals by about 1 dB and the detection performance of covert signals by about 0.6 dB.

[0210] Example 2

[0211] like Figure 7 As shown, a device for iteratively receiving a covert communication waveform based on orthogonal time-frequency space is used at a receiving end, and when executed, implements a method for iteratively receiving a covert communication waveform based on orthogonal time-frequency space, including:

[0212] (1) RF processing module, which receives the transmission signal and obtains the analog baseband signal; (2) analog-to-digital converter, which converts the analog baseband received signal into a digital baseband signal; (3) timing synchronization module, which is used for time synchronization of the received signal and obtains the synchronized baseband signal; (4) channel estimation module, which estimates the channel state information of the synchronized signal in the time-frequency domain or delay-Doppler domain and outputs it to the iterative linear channel equalization module; (5) orthogonal time-frequency-space demodulation module, which performs shaping matching filtering and orthogonal time-frequency-space demodulation on the synchronized baseband signal and outputs a multi-carrier demodulated signal; (6) channel equalization module, which performs channel state information estimation results and LMMSE criterion provided by the channel estimation module on the multi-carrier demodulated signal. The signal is linearly equalized based on the delay-Doppler domain equivalent channel state information, and the mean and variance of the external information of the multi-carrier equalized signal are output; (7) the conventional signal soft demodulation module performs fixed sequence soft demodulation on the mean of the external information of the multi-carrier equalized signal, and outputs the mean and variance of the single-carrier demodulated conventional signal after soft demodulation; (8) the conventional signal discrete estimation module performs discrete symbol estimation on the single-carrier demodulated conventional signal based on the prior Gaussian noise assumption, and outputs the posterior mean and variance of the conventional signal received symbol / bit; (9) the conventional signal soft spread spectrum module is used when the conventional signal is iteratively received, and calculates the conventional signal received symbol / bit external information based on the conventional signal received prior and posterior soft symbols / bits, and then The received symbol / bit external information is soft-spreaded with a fixed sequence, and the mean and variance after soft-spreading are output as the prior information for updating the multi-carrier equalized conventional signal; (10) The conventional signal hard demodulation / hard decision module is used when the conventional signal iterative reception is terminated, and further hard demodulation / hard decision is performed on the conventional signal received a posteriori soft symbols / bits to obtain the conventional signal received detection data sequence, that is, the hopping control data sequence; (11) The concealed signal hopping despreading and deinterleaving module performs hopping soft despreading and deinterleaving on the mean of the multi-carrier equalized signal external information based on the hopping control data sequence, and outputs the mean and variance of the single-carrier demodulated concealed signal after deinterleaving; (12) The LDPC iterative soft decoding module performs the single-carrier demodulation after deinterleaving. The concealed signal is subjected to LDPC iterative soft decoding, and the posterior mean and variance of the LDPC decoding soft bits of the same length as the input are output; (13) The concealed signal re-interleaving and code hopping spreading module is used when the concealed signal iterative reception is in progress, and the concealed signal LDPC decoding external information is calculated based on the concealed signal LDPC decoding prior and posterior soft bits, and then the concealed signal LDPC decoding external information is sequentially re-interleaved and code hopping soft spread, and the mean and variance after soft spreading are output as the updated multi-carrier equalized concealed signal prior information; (14) The LDPC decoding module is used when the concealed signal iterative reception is terminated, and the LDPC decoding posterior soft bits are further LDPC decoded to obtain the concealed signal reception detection data sequence.

[0213] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for iterative reception of covert communication waveforms based on orthogonal time-frequency space, characterized in that: The receiving end performs iterative reception detection on the baseband signal. The specific process is as follows: S1. Perform timing synchronization and channel estimation on the baseband signal, perform matched filtering and orthogonal time-frequency-space demodulation on the synchronized baseband signal to obtain a multi-carrier demodulated signal, and obtain an initial priori mean vector and an initial priori covariance matrix; S2, perform OTFS channel equalization according to the LMMSE criterion to obtain the external information mean vector and variance of the multi-carrier equalized signal, and transfer them to S3 and S6 respectively; S3, performing fixed sequence soft demodulation on the external information mean vector and variance of the multi-carrier equalized signal to obtain the mean vector and variance of the single-carrier demodulated conventional signal, performing discrete signal estimation based on the prior Gaussian noise assumption and the constellation mapping modulation mode of the conventional signal, obtaining the posterior mean vector and variance of the conventional signal reception information, and calculating the absolute difference in variance between the result of the previous iteration, if the absolute difference in variance is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S4, otherwise go to S5; S4, the conventional signal iterative reception detection is terminated, hard demodulation or hard decision is performed on the posterior mean vector of the current conventional signal reception information, and an estimation result of the conventional signal reception information and an estimation result of the corresponding random code hopping control information are output; S5. Based on the a posteriori mean vector and variance of the conventional signal reception information and the mean vector and variance of the single-carrier demodulated conventional signal, the mean vector and variance of the discrete signal external information are calculated and fixed sequence soft spreading is performed, the mean vector and variance after soft spreading are output, and the a priori mean vector and a priori covariance matrix of the multi-carrier equalized conventional signal in S2 are updated, and then the process is transferred to S2. S6, performing random code hopping soft despreading on the external information mean vector and variance of the multi-carrier equalized signal to obtain the mean vector and variance of the single-carrier demodulated concealed signal, calculating the a priori log-likelihood ratio of the concealed signal reception information based on the concealed signal interleaving coding mode and the prior Gaussian noise assumption, obtaining the posterior log-likelihood ratio vector and the corresponding variance through LDPC iterative soft decoding, and calculating the absolute difference in variance between the variance of the posterior log-likelihood ratio vector and the previous iteration result. If the absolute difference in variance is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S7; otherwise, go to S8; S7, the iterative reception detection of the concealed signal is terminated, LDPC decoding is performed on the posterior log-likelihood ratio vector of the current concealed signal reception information, and an estimation result of the concealed signal reception information is output; S8. Based on the posterior log-likelihood ratio vector and variance, calculate the external information mean vector and variance of LDPC soft decoding, perform row-column interleaving on the external information mean vector, perform random code hopping soft spread based on the random code hopping control information estimation result, output the mean vector and variance after soft spread, and update the prior mean vector and prior covariance matrix of the multi-carrier equalized concealed signal in S2, and go to S2.

2. The method for iterative reception of covert communication waveforms based on orthogonal time-frequency-space according to claim 1, characterized in that: The orthogonal time-frequency-space demodulation in step S1 includes vector-matrix transformation, matched filtering and Wigner transformation, symplectic-finite Fourier transformation and matrix-vector transformation performed in sequence.

3. The method for iterative reception of covert communication waveforms based on orthogonal time-frequency-space according to claim 1, characterized in that: Step S2 specifically includes the following process: S21, initializing or updating the prior mean and covariance matrix of the multi-carrier equalized signal; S22. Perform OTFS channel equalization according to the prior mean and covariance matrix of the multi-carrier equalized signal and the LMMSE criterion to obtain the posterior mean, covariance matrix and variance of the multi-carrier equalized signal; S23. Calculate and output the extrinsic information mean and variance of the multi-carrier equalized signal based on the priori mean, the posterior mean, the covariance matrix, and the variance of the multi-carrier equalized signal.

4. The method for iterative reception of covert communication waveforms based on orthogonal time-frequency-space according to claim 1, characterized in that: Step S3 specifically The following processes are included: S31, performing fixed sequence soft despreading on the external information of the multi-carrier equalized signal to obtain the mean vector and variance of the single-carrier demodulated conventional signal; S32. Perform discrete signal estimation based on the prior Gaussian noise hypothesis and the constellation mapping modulation mode of the conventional signal, and calculate the posterior mean vector and variance of the conventional signal reception information; S33. Calculate the posterior mean vector and variance of the conventional signal reception information and the absolute difference in variance between the results of the previous iteration. If the absolute difference in variance is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S4; otherwise, go to S5.

5. The method for iterative reception of covert communication waveforms based on orthogonal time-frequency-space according to claim 2, characterized in that: Step S5 specifically The following processes are included: S51, using the mean vector and variance of the single-carrier demodulated conventional signal as the equivalent priori mean vector and variance of the conventional signal reception information, and calculating the mean vector and variance of the discrete signal estimated extrinsic information based on the posterior mean vector and variance of the conventional signal reception information; S52, performing fixed sequence soft spreading on the estimated external information of the discrete signal to obtain a mean vector and variance after the soft spreading; S53 , updating the a priori mean vector and a priori covariance matrix of the multi-carrier equalized normal signal in S21 according to the mean vector and variance after soft spreading, and then going to S2 .

6. The method for iterative reception of covert communication waveforms based on orthogonal time-frequency-space according to claim 1, characterized in that: Step S6 specifically includes the following process: S61. Based on the random code hopping control information estimation result obtained by iterative reception detection of the conventional signal, the extrinsic information mean vector and variance of the multi-carrier equalized signal are subjected to random code hopping soft despreading to obtain the mean vector and variance of the single-carrier demodulated concealed signal; S62. Deinterleave the single-carrier demodulated concealed signal based on the concealed signal interleaving method to obtain a mean vector and variance after deinterleaving, and obtain a priori log-likelihood ratio of received coded bits of the concealed signal based on the prior Gaussian noise hypothesis data and the constellation mapping method of the concealed signal; S63, performing LDPC iterative soft decoding on the a priori log-likelihood ratio of the concealed signal received coded bits, obtaining an LDPC decoded soft bit vector of equal length as a posterior log-likelihood ratio vector, and calculating the corresponding variance; S64. Calculate the absolute difference between the variance of the posterior log-likelihood ratio vector and the variance of the last iteration result. If the absolute difference is less than the iteration convergence threshold or the number of iterations is equal to the iteration number threshold, go to S7; otherwise, go to S8.

7. The method for iterative reception of covert communication waveforms based on orthogonal time-frequency-space according to claim 2, characterized in that: Step S8 specifically The following processes are included: S81, calculating the mean vector and variance of the extrinsic information of the LDPC iterative soft decoding based on the a priori log-likelihood ratio and the a posteriori log-likelihood ratio vector of the concealed signal reception information; S82, interleave the LDPC iterative soft decoding external information to obtain an interleaved mean vector and variance; S83, performing random code hopping soft despreading on the interleaved mean vector and variance to obtain the mean vector and variance after soft spectrum spreading; S84. Update the priori mean vector and priori covariance matrix of the multi-carrier equalized concealed signal in S21 according to the mean vector and variance after soft spreading, and go to S2.

8. The method for iterative reception of covert communication waveforms based on orthogonal time-frequency-space according to claim 1, characterized in that: The regular signal reception information is specifically regular signal reception symbols or bits.

9. The method for iterative reception of covert communication waveforms based on orthogonal time-frequency-space according to claim 1, characterized in that: The concealed signal reception information is specifically concealed signal reception decoding bits.

10. A covert communication waveform iterative receiving device based on orthogonal time-frequency space, characterized in that: Used at the receiving end, when executed, it implements the iterative reception method for covert communication waveform based on orthogonal time-frequency-space as described in any one of claims 1-9, including: a radio frequency processing module, an analog-to-digital converter, a timing synchronization module, a channel estimation module, an orthogonal time-frequency-space demodulation module, an iterative linear channel equalization module, a conventional signal soft despreading module, a conventional signal discrete estimation module, a conventional signal soft spread spectrum module, a conventional signal hard demodulation / hard decision module, a concealed signal code hopping despreading domain deinterleaving module, an LDPC iterative soft decoding module, a concealed signal re-interleaving and code hopping spread spectrum module, and an LDPC decoding module.

Images