A method for enhancing the concealment of shortwave speech signals and the corresponding frequency offset estimation algorithm

Abstract

The present invention relates to the field of shortwave communication technology, and more specifically to a shortwave covert communication method based on an embedded synchronization sequence. The present invention embeds the synchronization sequence into colored noise, eliminating the need for a specific modulation scheme. This avoids the high complexity and difficulty of modulation and demodulation caused by such a specific modulation scheme, thereby achieving strong concealment of the synchronization sequence. Furthermore, a corresponding frequency offset estimation algorithm is provided, which has the advantages of high accuracy and ease of implementation.

Description

Technical Field

[0001] The present invention relates to the technical field of shortwave communication, and in particular to a shortwave covert communication method based on an embedded synchronization sequence. Background Art

[0002] With the rapid development of communications technology, a growing number of communication technologies and systems are constantly emerging. However, shortwave communication, an ancient and traditional communication method, remains widely valued worldwide. Far from being eliminated, it is experiencing rapid growth. This is because it offers advantages that other communication systems lack. First, shortwave is the only long-distance communication method unconstrained by networks and relays. For example, in the event of war, disasters, or satellite attacks, shortwave's survivability and autonomous communication capabilities are unmatched by other communication equipment. Second, shortwave is a primary means of communication in remote areas such as mountainous areas, deserts, and oceans. Finally, its low cost also ensures a broad market for shortwave. In summary, shortwave communication remains a vital means of information transmission in the communications sector.

[0003] Any communication system needs to send a synchronization sequence, followed by user information. The purpose of the synchronization sequence is to inform the receiver of the synchronization starting point and calculate the precise starting position of the user information. Only after the synchronization information is accurately understood can the user information be processed. Therefore, synchronization capture is the initial step for any communication system to communicate. When sending the synchronization sequence, corresponding modulation (such as PSK, CPM, etc.) needs to be performed to adapt it to the transmission characteristics of the channel. Taking into account the special application scenarios of shortwave communication, when both parties transmit information, factors such as signal concealment and interception need to be fully considered. However, due to the use of a modulation method well known in the field of communications, the signal has a special structure and does not have concealment characteristics, making the signal easy to observe and intercept.

[0004] To extract useful information, the receiver must downconvert the received high-frequency signal to a low-frequency signal. This process requires a high-frequency carrier with the same frequency as the transmitter. However, due to factors such as component manufacturing processes, materials, and electrical characteristics, the carrier frequencies generated by the transmitter and receiver are not always identical, and there will always be an error. This error affects the back-end demodulation. When the error is small, the impact on the demodulator is almost negligible. However, when the error is large, phase rotation occurs, resulting in uncorrectable errors and significantly degrading communication system performance. To address the performance degradation caused by frequency error between the transmitter and receiver, a common approach is to estimate the frequency offset of the signal after downconversion but before demodulation. The frequency offset estimate is artificially removed from the downconverted signal, so that the data entering the demodulator is treated as if it were frequency-offset-free, thus achieving correct results.

[0005] In summary, it is necessary to design a special synchronization sequence that is concealed and difficult to observe or intercept. Furthermore, due to the different carrier frequencies between the transmitter and receiver, a frequency offset estimation algorithm corresponding to this special synchronization sequence is also required. This invention embeds synchronization information in a distributed manner within colored noise, making the synchronization sequence invisible in both the time and frequency domains, thereby significantly reducing the probability of interception and eavesdropping. Summary of the Invention

[0006] In view of the problems existing in the prior art, the object of the present invention is to provide a shortwave covert communication method based on an embedded synchronization sequence.

[0007] In order to achieve the above objectives, the present invention adopts the following technical solutions to achieve them.

[0008] A method for enhancing the concealment of a shortwave voice signal comprises the following steps:

[0009] Step 1: Generate colored noise with the same bandwidth as the shortwave voice signal, power P, and length L at the transmitting end;

[0010] Step 2: amplify the amplitude of the original synchronization sequence of length K in the shortwave speech signal to P; the length L of the colored noise is much larger than the length K of the original synchronization sequence;

[0011] Step 3, embedding the original synchronization sequence into colored noise;

[0012] Step 4, extracting the colored noise embedded in the original synchronization sequence from the first element to the last element of the original synchronization sequence as the embedded synchronization sequence;

[0013] Step 5: The transmitter replaces the original synchronization sequence in the shortwave voice signal with the embedded synchronization sequence, and sends the replaced shortwave voice signal.

[0014] A frequency deviation estimation algorithm, based on the above-mentioned method for enhancing the concealment of shortwave speech signals, comprises the following steps:

[0015] Step 1: Construct a sinusoidal signal based on the local sequence and synchronization segment

[0016] Step 2: Non-uniformly sampled sinusoidal signal Perform fast Fourier transform to obtain the frequency offset estimate

[0017] Step 3: Based on the frequency offset estimate De-frequency offset the synchronization segment;

[0018] Step 4: Reconstruct the sinusoidal signal based on the local sequence and the synchronization segment after frequency offset removal Return to step 1 until the maximum number of iterations;

[0019] Step 5: The frequency offset estimate obtained in each iteration is Accumulate and get the total frequency deviation estimate

[0020] Compared with the prior art, the beneficial effects of the present invention are: the synchronization sequence is embedded in the colored noise, and there is no need to adopt a specific modulation method, thus avoiding the high complexity and difficulty of modulation and demodulation caused by the specific modulation method, and the strong concealment of the synchronization sequence can be achieved; and a corresponding frequency deviation estimation algorithm is provided, which has the advantages of high accuracy and easy implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] Figure 1 A schematic diagram of the structure of the embedded synchronization sequence of the present invention;

[0023] Figure 2 This is a synchronization curve diagram of the embedded synchronization head in the simulation result 1 of the present invention under the Gaussian channel signal-to-noise ratio of -5dB;

[0024] Figure 3 This is a waveform diagram of the embedded synchronization header and the voice signal in the time domain in the simulation result 1 of the present invention;

[0025] Figure 4 This is a curve diagram of the change of the total frequency offset estimation value after each iteration of 30 iterations when the actual frequency offset is 100 Hz and the Gaussian channel signal-to-noise ratio SNR=-5 dB in simulation result 2 of the present invention. DETAILED DESCRIPTION

[0026] The embodiments of the present invention will be described in detail below with reference to examples. However, those skilled in the art will understand that the following examples are only used to illustrate the present invention and should not be construed as limiting the scope of the present invention.

[0027] A method for enhancing the concealment of a shortwave voice signal comprises the following steps:

[0028] Step 1: Generate colored noise noise = [n0,n1,…,n L-1 ];

[0029] Step 2: convert the original synchronization sequence X of length K in the shortwave speech signal into [x0, x1, ..., x K-1] is amplified to P; the length L of the colored noise is much larger than the length K of the original synchronization sequence;

[0030] Step 3: embed the original synchronization sequence X into the colored noise;

[0031] Specifically, for the first element x0 of the original synchronization sequence X:

[0032] In the colored noise, starting from the first element, find the first element that belongs to the neighborhood range [x0-r,x0+r] of x0 and replace it with x0;

[0033] For the kth element x of the original synchronization sequence X k-1 :

[0034] In the colored noise, starting from the element after the last replaced element, find the first element belonging to x k-1 The neighborhood range [x k-1 -r,x k-1 +r] and replace that element with x k-1 ;

[0035] Among them, r is the fluctuation range.

[0036] The specific process is as follows:

[0037] For the first element x0 of the original synchronization sequence X: In the colored noise, starting from the first element n0, find the first i-th element n in the neighborhood range [x0-r,x0+r] of the element x0 i-1 , and n i-1 Replace with x0;

[0038] For the second element x1 of the original synchronization sequence X: in the colored noise noise, from the i+1th element n i First, find the first j-th element n in the neighborhood range [x1-r,x1+r] of element x1 j-1 , and n j-1 Replace with x1;

[0039] For the third element x2 of the original synchronization sequence X: in the colored noise noise, from the j+1th element n j First, find the first qth element n in the neighborhood range [x2-r,x2+r] of element x2 q-1 , n q-1 Replace with x2;

[0040] Similarly, the other elements x3, x4, ..., x K-1 Replace with colored noise.

[0041] Step 4, extracting the colored noise embedded in the original synchronization sequence from the first element to the last element of the original synchronization sequence as the embedded synchronization sequence;

[0042] Step 5: The transmitter replaces the original synchronization sequence in the shortwave voice signal with the embedded synchronization sequence, and sends the replaced shortwave voice signal.

[0043] Furthermore, the embedded synchronization sequence obtained in step 4 has a random length. In order to obtain an embedded synchronization sequence of a specified length, a step is further included between step 4 and step 5: in the embedded synchronization sequence, any part of the elements between any two adjacent original synchronization sequence elements is removed to obtain an embedded synchronization sequence of the specified length.

[0044] In order to understand the construction process of embedded synchronization sequence more intuitively, Figure 1 A schematic diagram of the construction is given. From the figure, we can see that in the colored noise, some sampling points are selected by comparison, and the sampling points that meet the conditions are marked. At the same time, these sampling points are embedded into the colored noise as actual synchronization points.

[0045] Simulation results 1

[0046] refer to Figure 2 , Figure 2 The figure shows the synchronization curve of the embedded synchronization header under a Gaussian channel signal-to-noise ratio of -5dB. It can be seen that there is a very obvious peak in the synchronization curve, which corresponds to the received synchronization header, so that the starting position of the data segment can be accurately determined.

[0047] refer to Figure 3 , Figure 3 Figure 2 is the waveform diagram of the embedded synchronization header and the voice signal in the time domain. It can be seen from the figure that the waveform of the embedded synchronization header is basically similar to that of the voice signal, which can achieve a good concealment effect.

[0048] It can be seen from the above simulation results that the method for enhancing the concealment of shortwave speech signals of the present invention has strong concealment while ensuring correct synchronous capture.

[0049] During synchronization capture, the receiver intercepts a stream of received baseband signals and identifies them based on the local sequence, recording the identification results. The received signals are then traversed and a recognition result curve is plotted. The peak of the recognition result curve corresponds to the starting point of the synchronization segment.

[0050] From the above principle, it can be seen that the process of synchronization head capture is the process of identifying the input baseband signals one by one, and the commonly used identification method is Fast Fourier Transform (FFT).

[0051] Since the local sequence X=(x0,Λ,x n ,Λ,x K-1 ) is known, so the phase change can be extracted from the intercepted received signal to obtain the sequence ,in So the sequence It can be equivalent to a sinusoidal signal superimposed with noise, and the frequency of the signal is f. Perform a fast Fourier transform to obtain the recognition result. Traverse the received signal and plot the recognition result curve. The peak of the recognition result curve corresponds to the starting point of the synchronization segment. The frequency corresponding to the peak of the recognition result curve is the frequency deviation.

[0052] The embedded synchronization sequence is obtained by embedding the original synchronization sequence into colored noise in a random manner. Therefore, the interval between any two elements in the embedded synchronization sequence is uncertain. The resulting ~R can be regarded as a non-uniformly sampled noisy sinusoidal signal.

[0053] For a sinusoidal signal with frequency deviation and noise obtained by non-uniform sampling, the frequency deviation value cannot be accurately obtained by directly using FFT transformation. This is determined by the randomness of the embedding process (equivalent to non-uniform sampling).

[0054] From the perspective of information theory, non-uniformly sampled signals still contain information, but the information is non-uniformly distributed. Although we cannot obtain all the information from them, we can still obtain some information from non-uniformly sampled signals.

[0055] A frequency deviation estimation algorithm, based on the above-mentioned method for enhancing the concealment of shortwave speech signals, comprises the following steps:

[0056] Step 1: Construct a sinusoidal signal based on the local sequence and synchronization segment

[0057] Among them, the sinusoidal signal

[0058] Synchronous segment R=(r0,…,r n ,…,r K-1 ),in Indicates the frequency deviation of the transmitted signal x k The influence of is a multiplicative interference and is the main factor causing the phase rotation of the symbol; represents the magnitude of the additional phase of the kth symbol, Rsym represents the symbol transmission rate; w k represents the noise on x k The influence of is an additive interference. k The mean is 0 and the variance is σ 2 Based on the above mathematical model, after estimating the frequency error f at both ends of the transmitter and receiver, the frequency deviation of the received signal can be removed by performing the corresponding "reverse" rotation on the received signal, that is,

[0059] Step 2: Non-uniformly sampled sinusoidal signal Perform fast Fourier transform to obtain the frequency offset estimate

[0060] Step 3: Based on the frequency offset estimate De-frequency offset the synchronization segment;

[0061] Step 4: Reconstruct the sinusoidal signal based on the local sequence and the synchronization segment after frequency offset removal Return to step 1 until the maximum number of iterations;

[0062] Step 5: The frequency offset estimate obtained in each iteration is Accumulate and get the total frequency deviation estimate

[0063] Simulation results 2

[0064] refer to Figure 4 , Figure 4 The following is a graph showing the variation of the total frequency deviation estimate after each iteration for 30 iterations with a true frequency deviation of 100Hz and a Gaussian channel signal-to-noise ratio (SNR) of -5dB. It can be seen that as the number of iterations increases, the total frequency deviation estimate gradually approaches the true value. The partial information (single frequency deviation value) estimated in the first iteration is the largest, and then the information gradually decreases. In the later stages of the iteration, the total estimated value oscillates around the true frequency deviation value. This indicates that the frequency deviation estimation iterative algorithm of the present invention gradually approaches the true frequency deviation value as the number of iterations increases, thus having the ability to estimate frequency deviation, and the estimated value has the advantages of high accuracy and ease of implementation.

[0065] Although this specification has provided a detailed description of the present invention using general descriptions and specific embodiments, it will be apparent to those skilled in the art that modifications and improvements may be made based on the present invention. Therefore, such modifications and improvements, which do not depart from the spirit of the present invention, are intended to be within the scope of protection claimed herein.

Claims

1. A method for enhancing the concealment of shortwave voice signals, characterized in that: The following steps are involved: Step 1: Generate a shortwave voice signal with the same bandwidth and power as the shortwave voice signal at the transmitter. , length is Colored noise; Step 2: convert the shortwave speech signal to The amplitude of the original synchronization sequence is amplified to ; Length of colored noise Much longer than the length of the original synchronization sequence ; Step 3, embedding the original synchronization sequence into colored noise; Specifically, for the first element of the original synchronization sequence : In the colored noise, starting from the first element, find the first Neighborhood range and replace it with ; For the original synchronization sequence Elements : In colored noise, starting from the element after the last replaced element, find the first element that belongs to Neighborhood range and replace it with ; in, is the fluctuation range; Step 4, extracting the colored noise embedded in the original synchronization sequence from the first element to the last element of the original synchronization sequence as the embedded synchronization sequence; Step 5: The transmitter replaces the original synchronization sequence in the shortwave voice signal with the embedded synchronization sequence, and sends the replaced shortwave voice signal.

2. The method for enhancing the concealment of shortwave voice signals according to claim 1, characterized in that: The method further includes the following steps between step 4 and step 5: removing any part of the elements between any two adjacent original synchronization sequence elements in the embedded synchronization sequence to obtain an embedded synchronization sequence of a specified length.

3. A frequency offset estimation algorithm, based on the method for enhancing the concealment of shortwave speech signals according to any one of claims 1 or 2, comprising the following steps: Step 1: Construct a sinusoidal signal based on the local sequence and synchronization segment ; Step 2: Non-uniformly sampled sinusoidal signal Perform fast Fourier transform to obtain the frequency offset estimate ; Step 3: Based on the frequency offset estimate De-frequency offset the synchronization segment; Step 4: Reconstruct the sinusoidal signal based on the local sequence and the synchronization segment after frequency offset removal , return to step 1 until the maximum number of iterations; Step 5: The frequency offset estimate obtained in each iteration is Accumulate and get the total frequency deviation estimate .

Images