Weak signal noise reduction and capture method and device

Abstract

The invention discloses a weak signal noise reduction and capture method and device, and the method comprises the steps: carrying out the carrier separation of a collected to-be-processed digital signal, obtaining an initial signal, grouping the initial signal, and obtaining a data block with a preset length; and performing coherent accumulation on the data blocks with the preset lengths to generate a coherent accumulation result. And dividing the data blocks with the preset lengths into odd-numbered groups of data blocks and even-numbered groups of data blocks, respectively performing incoherent accumulation on the odd-numbered groups of data blocks and the even-numbered groups of data blocks according to coherent accumulation results to generate corresponding incoherent accumulation results, and determining a larger value from the obtained incoherent accumulation results as a judgment threshold. And comparing the judgment threshold value with a preset capture threshold value, and when the judgment threshold value is greater than or equal to the preset capture threshold value, determining that the signal capture is successful. By adopting the method, the coherent accumulation frequency can be reduced, the operand and complexity are reduced, the weak signal is accurately and quickly captured, and the weak signal capturing efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a method and device for reducing noise and capturing weak signals. Background Art

[0002] With the development of communication technology, the technology of using the Global Positioning System (GPS) or the Beidou Satellite Navigation System (BDS) to provide users with accurate location, speed and precise time information has been widely used. In order to provide navigation services to users in different locations, it is necessary to capture the associated signals and conduct the next step of analysis, processing and feedback. Since the actual location of the user is usually in a complex environment, the signal will be attenuated to a weaker state. In order to ensure that the weak signal can be successfully captured and provide users with accurate navigation services, it is necessary to apply the weak signal capture method.

[0003] Traditional accumulation algorithms for weak signal capture include coherent accumulation algorithm, incoherent accumulation algorithm and differential coherent accumulation algorithm. Among them, the coherent accumulation algorithm can effectively remove the high-frequency components and noise of the signal, but when the signal power is low, it is necessary to increase the coherent accumulation time to effectively capture the signal. Incoherent accumulation requires incoherent accumulation of the square value after modulus, which can solve the problem that coherent accumulation cannot be accumulated for a long time, but it will cause loss and the effect is not ideal. Differential coherent accumulation is only applicable to the case where the coherent accumulation time is short. When the coherent accumulation time is long, differential coherent accumulation will be more sensitive to bit flipping, which will affect the capture of effective information.

[0004] Therefore, the currently commonly used weak signal capture method is not ideal for successfully capturing effective weak signals, and the work efficiency of weak signal capture is relatively low. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a weak signal denoising and capturing method and device which can improve the weak signal capturing efficiency.

[0006] A weak signal capture method, the method comprising:

[0007] A segmented hybrid stochastic resonance mathematical model is constructed, and a circuit design is performed on the segmented hybrid stochastic resonance mathematical model. Then, a signal to be processed is collected, and the signal to be processed is subjected to noise reduction processing using the segmented hybrid stochastic resonance mathematical model, and the analog signal is converted into a digital signal.

[0008] Collecting a digital signal to be processed, firstly performing carrier separation on the digital signal to be processed to obtain an initial signal after stripping the carrier, grouping the initial signal to obtain data blocks of a preset length after grouping; then coherently accumulating the data blocks of each preset length to generate a corresponding coherent accumulation result, and dividing the data blocks of each preset length into an odd-group data block and an even-group data block, and then performing incoherent accumulation on the odd-group data block and the even-group data block respectively according to the coherent accumulation result to generate a corresponding incoherent accumulation result; then determining a larger value from the incoherent accumulation result corresponding to the odd-group data block and the incoherent accumulation result corresponding to the even-group data block as a determination threshold;

[0009] The determination threshold is compared with the preset capture threshold. When the determination threshold is less than the preset capture threshold, the step of performing carrier separation on the digital signal to be processed to obtain the initial signal after carrier stripping is returned until the determination threshold is greater than or equal to the preset capture threshold and the signal is captured successfully.

[0010] In the present invention, a new segmented hybrid stochastic resonance weak signal detection model is proposed, comprising:

[0011] Construct a mathematical model of segmented hybrid stochastic resonance.

[0012] Conduct circuit design for the mathematical model of segmented hybrid stochastic resonance;

[0013] Collect the signal to be processed, and use the segmented hybrid stochastic resonance mathematical model to perform noise reduction on the signal to be processed;

[0014] And convert the analog signal into a digital signal;

[0015] In the present invention, the digital signal to be processed is subjected to carrier separation to obtain an initial signal after the carrier is stripped, including:

[0016] Obtaining a preset Doppler frequency shift carrier point;

[0017] Obtain a carrier signal sent by a local carrier generator;

[0018] The digital signal to be processed is multiplied with the carrier signal at the location of the preset Doppler frequency shift carrier point to obtain an initial signal after stripping the carrier.

[0019] In the present invention, the step of performing coherent accumulation on data blocks of each preset length to generate corresponding coherent accumulation results includes:

[0020] Performing fast Fourier transform processing on each data block of preset length to obtain a corresponding first transform result;

[0021] Get the local pseudo-random code;

[0022] Performing fast Fourier transform processing on the local pseudo-random code and taking conjugate, to obtain a transformed local pseudo-random code;

[0023] multiplying the first transformation result and the transformed pseudo-random code to obtain a second transformation result;

[0024] Perform inverse fast Fourier transform processing on the second transform result to generate a corresponding coherent accumulation result.

[0025] In the present invention, performing incoherent accumulation on the odd-numbered data blocks and the even-numbered data blocks respectively according to the coherent accumulation result to generate corresponding incoherent accumulation results includes:

[0026] The coherent accumulation results of each data block in the odd-numbered data blocks are squared and summed up to obtain the incoherent accumulation results of the odd-numbered data blocks; the coherent accumulation results of each data block in the even-numbered data blocks are squared and summed up to obtain the incoherent accumulation results of the even-numbered data blocks.

[0027] In the present invention, the step of comparing the determination threshold with the preset capture threshold, and returning to perform carrier separation on the digital signal to be processed to obtain the initial signal after stripping the carrier when the determination threshold is less than the preset capture threshold, includes:

[0028] Comparing the determination threshold with a preset capture threshold;

[0029] When the determination threshold is less than the preset capture threshold, changing the Doppler frequency shift carrier point by frequency difference correction;

[0030] The digital signal to be processed is multiplied by the carrier signal at the location of the modified Doppler frequency shift carrier point to obtain an initial signal after the carrier is stripped.

[0031] In the present invention, the initial signal is grouped to obtain a data block of a preset length after grouping, including:

[0032] Dividing the initial signal into a plurality of first data blocks having a first preset length;

[0033] Acquire sampling points with the same phase in each of the first data blocks;

[0034] According to the sampling points having the same phase, the first data blocks are superimposed to form a data block having a second preset length; the first preset length is greater than the second preset length.

[0035] In the present invention, the method of generating the preset capture threshold includes:

[0036] Get the preset capture accuracy;

[0037] Determine the corresponding false alarm probability according to the preset capture accuracy;

[0038] The corresponding capture threshold is calculated according to the false alarm probability and the variance of Gaussian white noise.

[0039] A weak signal capture device, comprising:

[0040] The signal noise reduction and amplification module collects weak signals and processes them through the segmented chaotic stochastic resonance mathematical model to obtain the noise-reduced amplified signal;

[0041] The analog-to-digital conversion module is used to convert the processed analog signal into the required digital signal;

[0042] A digital signal acquisition module, used for acquiring digital signals to be processed;

[0043] A carrier separation module, used for performing carrier separation on the digital signal to be processed to obtain an initial signal after stripping the carrier;

[0044] An initial signal grouping module, used for grouping the initial signal to obtain data blocks of a preset length after grouping;

[0045] A coherent accumulation module, used to perform coherent accumulation on data blocks of each preset length and generate corresponding coherent accumulation results;

[0046] A data block grouping module, used for dividing each data block of the preset length into an odd-numbered group of data blocks and an even-numbered group of data blocks;

[0047] A non-coherent accumulation module, used for performing non-coherent accumulation on the odd-numbered data blocks and the even-numbered data blocks respectively according to the coherent accumulation result, to generate corresponding non-coherent accumulation results;

[0048] A determination threshold determination module, used to determine a larger value from the non-coherent accumulation results corresponding to the odd-numbered data blocks and the non-coherent accumulation results corresponding to the even-numbered data blocks as a determination threshold;

[0049] The comparison module is used to compare the judgment threshold with the preset capture threshold. When the judgment threshold is less than the preset capture threshold, the comparison module returns to the step of performing carrier separation on the digital signal to be processed to obtain the initial signal after stripping the carrier, until the judgment threshold is greater than or equal to the preset capture threshold, and the signal is captured successfully.

[0050] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0051] Construct a mathematical model of segmented hybrid stochastic resonance.

[0052] Conduct circuit design for the mathematical model of segmented hybrid stochastic resonance;

[0053] Collect the signal to be processed, and use the segmented hybrid stochastic resonance mathematical model to perform noise reduction on the signal to be processed;

[0054] And convert the analog signal into a digital signal;

[0055] Collecting digital signals to be processed;

[0056] Carrier separation is performed on the digital signal to be processed to obtain an initial signal after the carrier is stripped off;

[0057] Grouping the initial signal to obtain grouped data blocks of a preset length;

[0058] Performing coherent accumulation on data blocks of each preset length to generate corresponding coherent accumulation results;

[0059] Dividing each data block of the preset length into an odd-numbered data block and an even-numbered data block;

[0060] According to the coherent accumulation result, respectively performing incoherent accumulation on the odd-numbered data blocks and the even-numbered data blocks to generate corresponding incoherent accumulation results;

[0061] Determine a larger value from the non-coherent accumulation results corresponding to the odd-numbered data blocks and the non-coherent accumulation results corresponding to the even-numbered data blocks as a determination threshold;

[0062] The determination threshold is compared with the preset capture threshold. When the determination threshold is less than the preset capture threshold, the step of performing carrier separation on the digital signal to be processed to obtain the initial signal after carrier stripping is returned until the determination threshold is greater than or equal to the preset capture threshold and the signal is captured successfully.

[0063] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0064] Construct a mathematical model of segmented hybrid stochastic resonance.

[0065] Conduct circuit design for the mathematical model of segmented hybrid stochastic resonance;

[0066] Collect the signal to be processed, and use the segmented hybrid stochastic resonance mathematical model to perform noise reduction on the signal to be processed;

[0067] And convert the analog signal into a digital signal;

[0068] Collecting digital signals to be processed;

[0069] Carrier separation is performed on the digital signal to be processed to obtain an initial signal after the carrier is stripped off;

[0070] Grouping the initial signal to obtain grouped data blocks of a preset length;

[0071] Performing coherent accumulation on data blocks of each preset length to generate corresponding coherent accumulation results;

[0072] Dividing each data block of the preset length into an odd-numbered data block and an even-numbered data block;

[0073] According to the coherent accumulation result, respectively performing incoherent accumulation on the odd-numbered data blocks and the even-numbered data blocks to generate corresponding incoherent accumulation results;

[0074] Determine a larger value from the non-coherent accumulation results corresponding to the odd-numbered data blocks and the non-coherent accumulation results corresponding to the even-numbered data blocks as a determination threshold;

[0075] The determination threshold is compared with the preset capture threshold. When the determination threshold is less than the preset capture threshold, the step of performing carrier separation on the digital signal to be processed to obtain the initial signal after carrier stripping is returned until the determination threshold is greater than or equal to the preset capture threshold and the signal is captured successfully.

[0076] The present invention realizes a weak signal noise reduction and capture method and device, the method comprising: constructing a segmented hybrid stochastic resonance mathematical model, and designing a circuit for the segmented hybrid stochastic resonance mathematical model, then collecting a signal to be processed, using the segmented hybrid stochastic resonance mathematical model to perform noise reduction processing on the signal to be processed, and converting the analog signal into a digital signal; performing carrier separation on the collected digital signal to be processed to obtain an initial signal after stripping the carrier, and grouping the initial signal to obtain a data block of a preset length after grouping. By coherently accumulating the data blocks of each preset length after grouping to generate a corresponding coherent accumulation result, the number of coherent accumulation operations can be reduced, the amount of calculation and complexity of the algorithm can be reduced, and the signal capture speed can be improved. By dividing the data blocks of each preset length into odd-numbered data blocks and even-numbered data blocks, and according to the coherent accumulation result, the odd-numbered data blocks and the even-numbered data blocks are respectively incoherently accumulated to generate a corresponding incoherent accumulation result, the capture result can be reduced from being affected by bit flipping, and the accuracy of the capture result can be improved. Then, a larger value is determined from the incoherent accumulation results as a decision threshold, and the decision threshold is compared with a preset capture threshold. When the decision threshold is greater than or equal to the preset capture threshold, it indicates that the signal is captured successfully.

[0077] Carrier separation is performed on the collected digital signal to be processed to obtain an initial signal, and the initial signal is grouped to obtain a data block of a preset length. Coherent accumulation is performed on the data blocks of each preset length to generate a coherent accumulation result. The data blocks of each preset length are divided into an odd-numbered data block and an even-numbered data block. According to the coherent accumulation result, incoherent accumulation is performed on the odd-numbered data block and the even-numbered data block respectively to generate corresponding incoherent accumulation results, and a larger value is determined from the obtained incoherent accumulation results as a determination threshold. The determination threshold is compared with the preset capture threshold. When the determination threshold is greater than or equal to the preset capture threshold, the signal is captured successfully. The use of this method can reduce the number of coherent accumulation times, reduce the amount of calculation and complexity, accurately and quickly capture weak signals, and improve the efficiency of weak signal capture. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 This is an application environment diagram of the weak signal capture method of the present invention;

[0079] Figure 2 It is a signal noise reduction circuit of the segmented hybrid stochastic resonance mathematical system of the present invention;

[0080] Figure 3 A chaotic circuit for generating noise for the present invention;

[0081] Figure 4 An amplifier circuit for adjusting the noise level of the present invention;

[0082] Figure 5The input is a sine wave with an amplitude of 300mv and a frequency of 1Hz, and the system noise reduction performance diagram after noise ((a) is a sine wave with an input amplitude of 300mv and a frequency of 1Hz and noise that is sufficient to cover the sine wave, (b) is a comparison between the waveform after system noise reduction and the input waveform);

[0083] Figure 6 The input is a sine wave with an amplitude of 5V and a frequency of 10Hz, and the system noise reduction performance diagram after noise ((a) is a comparison diagram of the waveform after system noise reduction and the input waveform when the input amplitude is 5V, the frequency is 10Hz, and the noise is sufficient to cover the sine wave);

[0084] Figure 7 It is a schematic diagram of the process of the weak signal capture method of the present invention;

[0085] Figure 8 A schematic diagram of a digital signal processing flow of a weak signal capture method of the present invention;

[0086] Fig. 9 A schematic diagram of a process of obtaining data blocks of a preset length after grouping according to the present invention;

[0087] Fig.10 A schematic diagram of comparing data block groups of the weak signal capture method of the present invention; Fig.10 (a) is a schematic diagram of a processing flow of data blocks that have not been superimposed; Fig.10 (b) Schematic diagram of the processing flow of data block grouping after superposition by the block processing method;

[0088] Fig.11 It is a schematic diagram of the overall architecture of the weak signal capture method of the present invention;

[0089] Fig.12 It is a structural block diagram of the weak signal capture device of the present invention;

[0090] Fig.13 1 is a diagram showing the internal structure of the computer device of the present invention. DETAILED DESCRIPTION

[0091] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0092] The weak signal capture method provided by the present invention can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The server 104 collects the digital signal to be processed, and performs carrier separation on the digital signal to be processed to obtain the initial signal after stripping the carrier. Among them, the digital signal to be processed can be obtained from the cloud storage corresponding to the server 104, or from the local storage of the terminal 102. The server 104 groups the initial signal to obtain data blocks of a preset length after grouping, and performs coherent accumulation on the data blocks of each preset length to generate corresponding coherent accumulation results. The server 104 divides the data blocks of each preset length into odd-array data blocks and even-array data blocks, and according to the coherent accumulation results, performs incoherent accumulation on the odd-array data blocks and the even-array data blocks respectively, generates corresponding incoherent accumulation results, and then determines the larger value from the incoherent accumulation results corresponding to the odd-array data blocks and the incoherent accumulation results corresponding to the even-array data blocks as the judgment threshold. The server 104 compares the determination threshold with the preset capture threshold. When the determination threshold is less than the preset capture threshold, the server 104 returns to the step of performing carrier separation on the digital signal to be processed to obtain the initial signal after the carrier is stripped, until the determination threshold is greater than or equal to the preset capture threshold, and the signal is successfully captured. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices, and the server 104 can be implemented as an independent server or a server cluster consisting of multiple servers.

[0093] In the present invention, a new segmented hybrid stochastic resonance weak signal detection model is constructed, comprising the following steps:

[0094] Step S102, constructing a segmented hybrid stochastic resonance mathematical model.

[0095] The one-dimensional continuous nonlinear bistable system model described by the Langevin equation is shown in formula (1):

[0096]

[0097] in, H(t) is the input signal, η(t) is the noise signal, a L and b L is a real number greater than 0. L (x) is the potential function of the system:

[0098]

[0099] The Langevin equation of the piecewise mixed model is shown in formula (3):

[0100]

[0101] Among them, U(x) is the potential function of the piecewise hybrid model, H(t) is the input signal, η(t) is the noise signal, K is the amplification factor, and the expression of U(x) is shown in formula (4):

[0102]

[0103] Among them, a, b, c are all real numbers, and a>0, b>0,

[0104] By comparing equation (2) and equation (4), we can see that the piecewise hybrid model is composed of two linear equations and one nonlinear equation. The nonlinear equation in the middle segment is a bistable system equation. The two symmetrical linear equations of the piecewise hybrid model are precisely to overcome the saturation characteristics of the original bistable system. The potential function of the piecewise hybrid model also has two potential wells, with the potential bottom located at The left and right outer curves intersect the horizontal axis at Pick

[0105] Step S104, performing circuit design on the segmented hybrid stochastic resonance mathematical model;

[0106] Substituting formula (4) into formula (3), we can get:

[0107]

[0108] In formula (5), s(t)=H(t)+η(t), and U′(x) represents dU(x) / dx.

[0109] According to formula (5), the system can be composed of amplification, summation, integration, and comparison feedback. Select appropriate resistors and capacitors, adjust the resistance of the variable potentiometer, so that a = 2, b = 1, c = 2, K = 10, and get formula 6. According to formula 6, the circuit Figure 2 .

[0110]

[0111] The hardware circuit of the segmented hybrid stochastic resonance system is as follows Figure 2 The noise is given by Figure 3 The chaotic circuit can be generated by Figure 4 The amplifier circuit adjusts the noise level, and the weak signal to be detected H(t) and the noise signal η(t) are input through the same-direction input terminals 1 and 2 of the pre-summing amplifier U1A. The amplification factor K is changed by adjusting R1, and Ks(t) is obtained through U1A. U2A is an integrator, which adjusts the input signal The integral is x(t), which is the processed signal and can be displayed in real time by an oscilloscope.

[0112] Figure 2Among them, v1, v2, v3, and v4 are comparison voltages, which are provided to the comparator as reference comparison voltages. The three selection feedback paths are mainly composed of comparators, multipliers, and amplifiers. At any time, only one path is selected, and whether it is selected is determined by comparing x(t) with the reference comparison voltage.

[0113] Feedback path 1 is selected and controlled by comparator U5. Comparator U5 is powered by positive and negative dual power supplies and is used to detect signals with x(t)<-2. When x(t)<-2, comparator U5 outputs a positive level of 5v, otherwise it is 0. After adjusting by resistors R8 and R10, the input voltage value of the reverse operational amplifier U4A is -4v or 0. Feedback path 2 is selected and controlled by comparator U8. Comparator U8 is powered by positive and negative dual power supplies and is used to detect signals with x(t)>2. When x(t)>2, comparator U8 outputs a positive level of 5v, otherwise it is 0. After adjusting by resistors R16 and R11, the output voltage value of the reverse operational amplifier U7A is -4v or 0, and then outputs 4v or 0 through inverter U6A. Feedback path 3 is selected and controlled by comparators U10, U14, amplifiers U13A, U15A, U19A and multipliers A2, A4, A5, A6. Comparators U10 and U14 are used to detect signals of -2≤x(t)≤2. When -2≤x(t), the output of comparator U10 is a positive level 5v, and when x(t)>-2, the output of comparator U10 is 0. When x(t)≤2, the output of comparator U14 is a positive level 5v, and when x(t)>2, the output of comparator U14 is 0. The output signals of U10 and U14 are multiplied by A2, so only when -2≤x(t)≤2 can the positive level 25v be output, otherwise it is 0. After adjustment by resistors R22 and R23, the input voltage value of the reverse operational amplifier FU3 is -1 or 0. At the same time, x(t) is output by multipliers A5 and A6. 3 , and outputs -x(t) through inverter U15A 3 , and then adjust the output -2x(t)+x(t) through the reverse summer U19A 3 When -2≤x(t)≤2, the output level of U13A is -1 and the output of U19A is -2x(t)+x(t) 3 , are input to multiplier A4 together to obtain the output 2x(t)-x(t) of the third selection feedback path 3 The final outputs of the three feedback channels are input into the same direction adder U1A via R3.

[0114] exist Figure 5 In the figure, (a) is a sine wave with an input amplitude of 300mv and a frequency of 1Hz and noise that is sufficient to cover the sine wave, and (b) is a comparison between the waveform after system noise reduction and the input waveform, proving that the system has a significant noise reduction effect, but the signal strength will be attenuated. Figure 6In the figure, (a) is a sine wave with an input amplitude of 5V and a frequency of 10Hz, and the noise is sufficient to cover the sine wave. The comparison between the waveform after system noise reduction and the input waveform is shown in Figure (b). The results show that the noise reduction effect of the system is obvious and better than that at 300mV.

[0115] like Figure 7 As shown, a weak signal capture method is provided, which is applied to Figure 1 The server in the example is used to illustrate the following steps:

[0116] Step S202, collecting the digital signal to be processed.

[0117] Specifically, the collected digital signal to be processed is a weak signal, wherein the weak signal refers to a digital signal with a signal-to-noise ratio below -22dB, wherein the source of the digital signal to be processed is the BeiDou Navigation Satellite System (BDS).

[0118] Furthermore, the collected weak signal is an intermediate frequency digital signal, and the collected digital signal to be processed is expressed by the following formula (1):

[0119]

[0120] Where r(n) represents the intermediate frequency digital signal, A is the signal amplitude, d(n) is the navigation data, c(n) is the local C / A code, ω0 is the intermediate frequency carrier angular frequency, ω d is the angular frequency of the Doppler shift, is the initial phase of the carrier, f(n) is Gaussian white noise, n=0, 1, 2…N-1, represents the number of sampling points of the digital signal to be processed.

[0121] Step S204, performing carrier separation on the digital signal to be processed to obtain an initial signal after the carrier is stripped.

[0122] Specifically, by obtaining a preset Doppler frequency shift carrier point and a carrier signal emitted by a local carrier generator, the digital signal to be processed is multiplied with the carrier signal at the location of the preset Doppler frequency shift carrier point to obtain an initial signal after stripping the carrier.

[0123] Furthermore, the following formula (2) is used to represent the initial signal after stripping the carrier:

[0124] x(n)=I(n)+jQ(n); (2)

[0125] Among them, x(n) represents the initial signal after stripping the carrier, I(n) represents the local carrier signal of the first branch, Q(n) represents the local carrier signal of the second branch, and j represents that the local carrier signal of the first branch is orthogonal to the local carrier signal of the second branch.

[0126] Specifically, the local carrier is stripped off by multiplying the digital signal to be processed with the carrier signal at the location of the preset Doppler frequency shift carrier point, and two branch signals I(n) and Q(n) are obtained. I(n) and Q(n) are used as the real part and imaginary part of the complex number respectively, and the initial signal after stripping the carrier is constructed.

[0127] In one embodiment, the local carrier signal of the first branch and the local carrier signal of the second branch are expressed by the following formula (3) and formula (4):

[0128]

[0129] in, represents an estimate of the angular frequency of the Doppler shift, Represents an estimate of the initial phase of the wave.

[0130] Step S206, grouping the initial signal to obtain grouped data blocks of a preset length.

[0131] Specifically, the initial signal after stripping the carrier is grouped into a plurality of first data blocks of a first preset length, and then superimposed into a plurality of second data blocks of a second preset length. The first preset length is greater than the second preset length. In this embodiment, the first preset length can be 10 ms and the second preset length can be 1 ms.

[0132] Step S208, performing coherent accumulation on the data blocks of each preset length to generate corresponding coherent accumulation results.

[0133] Specifically, by performing fast Fourier transform processing on each data block of a preset length, a corresponding first transform result is obtained. By obtaining a local pseudo-random code, performing fast Fourier transform processing on the local pseudo-random code and taking a conjugate, a transformed local pseudo-random code is obtained. Then, the first transform result and the transformed pseudo-random code are multiplied to obtain a second transform result, and the second transform result is subjected to inverse fast Fourier transform processing to generate a corresponding coherent accumulation result.

[0134] Further, each data block of the preset length is subjected to a fast Fourier transform process, and the first transform result obtained is expressed by the following formula (5):

[0135]

[0136] Where k represents the number of data blocks, k = 1, 2, L, 2N, M is the number of digital sampling points in a 1ms data block, X k (M) is the result of fast Fourier transform of the signal after stripping the carrier.

[0137] Furthermore, after the local pseudo-random code is fast Fourier transformed and conjugated, it is expressed by the following formula (6):

[0138]

[0139] Among them, C k (M) represents the result of fast Fourier transform of the local pseudo-random code.

[0140] Further, formula (5) and formula (6) are multiplied to obtain a second transformation result, and the second transformation result is subjected to inverse fast Fourier transform processing, and the obtained coherent accumulation result is expressed by the following formula (7):

[0141]

[0142] Among them, Y k (M) represents the result of performing inverse fast Fourier transform processing on the second transform result, that is, the coherent accumulation result. k is the coherent integration result of the kth data block, Y k (1) is the result of the first sampling point in the k-th data block. Finally, the coherent integration result of the k-th data block is composed of M sampling points.

[0143] Furthermore, the gain obtained by coherent accumulation is expressed by the following formula (8):

[0144] G c (f) = 10ilg(f); (8)

[0145] Among them, G c (f) represents the gain obtained by coherent accumulation, and f is the data length relative to 1 ms.

[0146] Step S210: divide each data block of a preset length into an odd-numbered group of data blocks and an even-numbered group of data blocks.

[0147] Specifically, the data blocks of each preset length corresponding to the coherent accumulation result are divided into odd-numbered data blocks and even-numbered data blocks by using a half-bit method. Since the received satellite intermediate frequency signal has at most one bit flip every 20ms, if the signal flip occurs within the first 10ms, the data in the next 10ms will not be affected by the bit flip. Therefore, the method of grouping data blocks by using a half-bit method can prevent the capture result from being affected by the bit flip.

[0148] Step S212: performing incoherent accumulation on the odd-numbered data blocks and the even-numbered data blocks respectively according to the coherent accumulation result to generate corresponding incoherent accumulation results.

[0149] Specifically, the coherent accumulation results of each data block in the odd-numbered data blocks are squared and summed to obtain the incoherent accumulation results of the odd-numbered data blocks. Similarly, the coherent accumulation results of each data block in the even-numbered data blocks are squared and summed to obtain the incoherent accumulation results of the even-numbered data blocks.

[0150] Furthermore, the following formula (9) can be used to express the accumulation result obtained by non-coherently accumulating the odd-numbered data blocks and the even-numbered data blocks respectively using the half-bit method:

[0151]

[0152] Among them, Y A Indicates the incoherent accumulation result corresponding to the odd-numbered data blocks, Y B represents the non-coherent accumulation result corresponding to an even number of data blocks, and k represents the number of data blocks.

[0153] The gain obtained by performing incoherent accumulation is expressed by the following formula (10):

[0154] G fc (f) = G c (f)-L(f)=10lg(f)-L(f); (10)

[0155] Among them, f is the data length relative to 1ms, G fc (f) represents the gain obtained by incoherent accumulation, and L(f) represents the loss of incoherent accumulation, which is expressed by the following formula (11):

[0156]

[0157] Among them, D c It represents the ideal detection factor under a specific false alarm probability and a specific detection probability. -7 , when the detection probability is 0.9, D c ≈21. In order to ensure that the capture sensitivity and signal-to-noise ratio are not reduced, the accumulated coherent accumulation length needs to be strictly controlled.

[0158] Step S214: determine a larger value from the non-coherent accumulation results corresponding to the odd-numbered data blocks and the non-coherent accumulation results corresponding to the even-numbered data blocks as a decision threshold.

[0159] Specifically, take Y=max(Y A ,YB ), that is, determine the larger value from the incoherent accumulation results corresponding to the odd-numbered data blocks and the incoherent accumulation results corresponding to the even-numbered data blocks, and use the obtained maximum value as the decision threshold. The decision threshold is used to be compared with the preset capture threshold to determine the final weak signal capture result.

[0160] In the present invention, Figure 8 As shown, a digital signal processing flow of a weak signal capture method is provided. Figure 8 , by superimposing 2N data blocks with a length of 10ms into corresponding 2N data blocks with a length of 1ms, and coherently accumulating each data block with a length of 1ms respectively, according to the coherent accumulation result, it is divided into odd-group data blocks and even-group data blocks, and further incoherently accumulating the grouped odd-group data blocks and even-group data blocks respectively to obtain corresponding coherent accumulation results, and determining the larger value from the incoherent accumulation results corresponding to the odd-group data blocks and the incoherent accumulation results corresponding to the even-group data blocks as the judgment threshold, and comparing the judgment threshold with the capture threshold.

[0161] Step S216, compare the judgment threshold with the preset capture threshold. When the judgment threshold is less than the preset capture threshold, return to the step of carrier separation of the digital signal to be processed to obtain the initial signal after stripping the carrier, until the judgment threshold is greater than or equal to the preset capture threshold, and the signal is captured successfully.

[0162] Specifically, by comparing the determination threshold with the preset capture threshold, when the determination threshold is greater than or equal to the preset capture threshold, it indicates that the signal capture is successful.

[0163] Furthermore, when the judgment threshold is less than the preset capture threshold, the Doppler frequency shift carrier point is changed by frequency difference correction, the digital signal to be processed and the carrier signal are multiplied at the position of the changed Doppler frequency shift carrier point, and the digital signal to be processed is carrier-separated again to obtain the initial signal after the carrier is stripped.

[0164] It can be understood that after obtaining the initial signal after stripping the carrier, steps S206 to S214 are repeatedly performed until the threshold is determined to be greater than or equal to the preset capture threshold, and the signal capture operation is completed.

[0165] The methods for generating preset capture thresholds include:

[0166] Obtaining a preset capture accuracy; determining a corresponding false alarm probability according to the preset capture accuracy;

[0167] According to the false alarm probability and the variance of Gaussian white noise, the corresponding capture threshold is calculated.

[0168] Specifically, the performance of weak signal capture is affected by the capture threshold value. When the capture threshold value is too large, it will cause missed alarms, and when the capture threshold value is too small, it will cause false alarms. The capture threshold is set according to the Newman-Pearson criterion and is expressed by the following formula (12):

[0169]

[0170] Where V is the capture threshold, σ n is the variance of Gaussian white noise, P fa is the false alarm probability. The false alarm probability can be preset or modified according to the capture accuracy requirements of the system. The smaller the false alarm probability value, the better the capture performance. Generally, the false alarm probability P is taken as fa =10 -5 or 10 -7 , or a smaller value. After the false alarm probability is defined, the corresponding capture threshold value can be calculated according to formula (12) as the preset capture threshold.

[0171] In the present invention, Fig. 9 As shown, the step of obtaining the data blocks of the preset length after grouping, that is, the step of grouping the initial signal to obtain the data blocks of the preset length after grouping, specifically includes:

[0172] Step S402: divide the initial signal into a plurality of first data blocks with a first preset length.

[0173] Specifically, in this embodiment, the first preset length is 10 ms, and the initial signal after stripping the carrier is divided into multiple data blocks with a length of 10 ms.

[0174] Step S404: Acquire sampling points with the same phase in each first data block.

[0175] Specifically, sampling points with the same phase in each first data block with a length of 10 ms are obtained, and a superposition operation is implemented based on the sampling points with the same phase to reduce the number of subsequent fast Fourier transform and inverse fast Fourier transform executions, thereby improving the capture speed of weak signals.

[0176] Step S406: According to the sampling points having the same phase, the first data blocks are superimposed to form a data block having a second preset length.

[0177] The first preset length is greater than the second preset length. In this embodiment, the first preset length is 10 ms, and the second preset length is 1 ms.

[0178] Specifically, since the signal is periodic, the same position in different data blocks has the same phase. Therefore, each 1ms data block has sampling points with the same phase. For example, in the first data block, there is a sampling point with a phase of If the signal sampling point is , then there are also phases of in the 2nd to 10th blocks of data. It is known that there are M sampling points in a 1ms data block, and there are 10 1ms in total, so each phase sampling point can be superimposed 10 times.

[0179] For example, if Fig.10 As shown, Fig.10 A schematic diagram of data block grouping and comparison of a weak signal capture method is provided. Fig.10 b provides a schematic diagram of the processing flow of data block grouping after superposition by the block processing method, referring to Fig.10 b. A 10ms data can be divided into 10 1ms data. There are M sampling points in a 1ms data block. The sampling points with the same phase are superimposed, so that each phase sampling point can be superimposed 10 times. Then, the M phase sampling points can be superimposed 10M times in total. The length after superposition is the second preset length, that is, the sampling points of a 1ms data block are M, and the amplitude is 10 times that of a 10ms length data block. The corresponding coherent accumulation time is reduced from the original 10ms to 1ms, and only M fast Fourier transforms and inverse fast Fourier transforms need to be performed subsequently.

[0180] Further, Fig.10 a provides a schematic diagram of the processing flow of unsuperimposed data blocks, referring to Fig.10 a. In the existing capture scheme, the data blocks are not grouped and superimposed, and the 10ms data is directly sampled, where M points are sampled per millisecond, and there are 10M sampling points in total. The corresponding coherent accumulation time is 10ms, and 10M fast Fourier transforms and inverse fast Fourier transforms need to be performed. It can be seen that the existing capture scheme consumes more time and resources to perform fast Fourier transform and inverse fast Fourier transform operations, resulting in a low capture speed for weak signals.

[0181] By dividing the initial signal into a plurality of first data blocks with a first preset length, acquiring each sampling point with the same phase in each first data block, and then superimposing the first data blocks into data blocks with a second preset length according to each sampling point with the same phase, it is possible to reduce the corresponding coherent accumulation time, reduce the number of fast Fourier transform and inverse fast Fourier transform operations, and thus improve the capture speed of weak signals.

[0182] It should be understood that although Figure 7 and Fig. 9The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 7 and Fig. 9 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0183] like Fig.11 As shown, an overall architecture of a weak signal capture method is provided. Fig.11 ,The overall architecture of the weak signal capture method includes five parts: the first part : carrier separation, the second part : signal grouping, the third part : FFT / IFFT ,transformation and coherent accumulation, the fourth part : incoherent accumulation, and the fifth part : signal ,capture judgment.

[0184] Specifically, the distribution of each part is as follows Fig.11 As shown, refer to Fig.11 :

[0185] Part 1: Carrier separation. The digital signal to be processed is multiplied with the carrier signal at the location of the preset Doppler frequency shift carrier point to obtain the initial signal after the carrier is stripped, and the two branch signals I(n) and Q(n) are output. The two branch signals are orthogonal. The two branches I(n) and Q(n) are the real part and imaginary part of the complex number, respectively, to construct a new signal after the carrier is stripped.

[0186] Part 2: Signal grouping: The initial signal is divided into a number of first data blocks with a length of 10ms, and each sampling point with the same phase in each first data block is obtained, and then the first data blocks are superimposed into a data block with a length of 1ms according to each sampling point with the same phase.

[0187] Part 3: FFT / IFFT transformation and coherent accumulation. Perform fast Fourier transform processing on each data block of preset length, perform fast Fourier transform processing on the local pseudo-random code and take the conjugate to obtain the transformed local pseudo-random code, that is, the local C / A code. The corresponding coherent accumulation result is generated by multiplying the first transformation result and the transformed pseudo-random code and then performing inverse fast Fourier transform processing.

[0188] Part 4: Incoherent accumulation. Using a half-bit method, the odd-array data blocks and the even-array data blocks are separated for incoherent accumulation. The coherent accumulation results of each data block in the odd-array data blocks are squared and summed to obtain the incoherent accumulation results of the odd-array data blocks. Similarly, the coherent accumulation results of each data block in the even-array data blocks are squared and summed to obtain the incoherent accumulation results of the even-array data blocks. Finally, the larger value is taken from the incoherent accumulation results of the odd-array data blocks and the incoherent accumulation results of the even-array data blocks as the decision variable, that is, the decision threshold.

[0189] Part 5: Signal capture determination. Get the preset capture threshold, that is Fig.11 The capture threshold value shown in , by comparing the decision variable with the capture threshold value, if the decision variable is greater than or equal to the capture threshold value, the signal capture is successful; if the decision variable is less than the capture threshold value, the signal capture fails, and the Doppler frequency shift is changed according to the execution steps to re-capture the signal.

[0190] like Fig.12 As shown, a weak signal capture device is provided, including: a digital signal acquisition module 702, a carrier separation module 704, an initial signal grouping module 706, a coherent accumulation module 708, a data block grouping module 710, a non-coherent accumulation module 712, a judgment threshold determination module 714 and a comparison module 716, wherein:

[0191] The signal noise reduction and amplification module collects weak signals and processes them through the segmented chaotic stochastic resonance mathematical model to obtain the noise-reduced amplified signal;

[0192] The analog-to-digital conversion module is used to convert the processed analog signal into the required digital signal;

[0193] The digital signal acquisition module 702 is used to acquire the digital signal to be processed.

[0194] The carrier separation module 704 is used to perform carrier separation on the digital signal to be processed to obtain an initial signal after the carrier is stripped.

[0195] The initial signal grouping module 706 is used to group the initial signal to obtain data blocks of a preset length after grouping.

[0196] The coherent accumulation module 708 is used to perform coherent accumulation on data blocks of each preset length to generate corresponding coherent accumulation results.

[0197] The data block grouping module 710 is used to divide each data block of a preset length into an odd-numbered group of data blocks and an even-numbered group of data blocks.

[0198] The non-coherent accumulation module 712 is used to perform non-coherent accumulation on the odd-numbered data blocks and the even-numbered data blocks respectively according to the coherent accumulation results to generate corresponding non-coherent accumulation results.

[0199] The decision threshold determination module 714 is used to determine a larger value from the non-coherent accumulation results corresponding to the odd-numbered data blocks and the non-coherent accumulation results corresponding to the even-numbered data blocks as a decision threshold.

[0200] The comparison module 716 is used to compare the judgment threshold with the preset capture threshold. When the judgment threshold is less than the preset capture threshold, it returns to the step of performing carrier separation on the digital signal to be processed to obtain the initial signal after stripping the carrier, until the judgment threshold is greater than or equal to the preset capture threshold, and the signal is captured successfully.

[0201] In specific implementation, as a preferred embodiment of the present invention, the initial signal grouping module is also used for:

[0202] Dividing the initial signal into a plurality of first data blocks having a first preset length;

[0203] Acquire each sampling point having the same phase in each first data block;

[0204] According to each sampling point having the same phase, the first data block is superimposed to form a data block having a second preset length; the first preset length is greater than the second preset length.

[0205] In specific implementation, as a preferred embodiment of the present invention, by dividing the initial signal into multiple first data blocks with a first preset length, and obtaining each sampling point with the same phase in each first data block, and then superimposing the first data blocks into data blocks with a second preset length according to each sampling point with the same phase, it is possible to reduce the corresponding coherent accumulation time, reduce the number of fast Fourier transform and inverse fast Fourier transform operations, and thus improve the capture speed of weak signals.

[0206] In specific implementation, as a preferred embodiment of the present invention, the carrier separation module is also used for:

[0207] Obtaining a preset Doppler frequency shift carrier point;

[0208] Obtain a carrier signal sent by a local carrier generator;

[0209] The digital signal to be processed is multiplied with the carrier signal at the location of the preset Doppler frequency shift carrier point to obtain the initial signal after stripping the carrier.

[0210] In specific implementation, as a preferred embodiment of the present invention, the coherent accumulation module is also used for:

[0211] Performing fast Fourier transform processing on each data block of preset length to obtain a corresponding first transform result;

[0212] Get the local pseudo-random code;

[0213] Performing fast Fourier transform processing on the local pseudo-random code and taking conjugate, to obtain a transformed local pseudo-random code;

[0214] Multiplying the first transformation result and the transformed pseudo-random code to obtain a second transformation result;

[0215] Perform inverse fast Fourier transform processing on the second transform result to generate a corresponding coherent accumulation result.

[0216] In specific implementation, as a preferred embodiment of the present invention, the non-coherent accumulation module is also used for:

[0217] The coherent accumulation results of each data block in the odd-group data block are squared and summed to obtain the incoherent accumulation results of the odd-group data block; the coherent accumulation results of each data block in the even-group data block are squared and summed to obtain the incoherent accumulation results of the even-group data block.

[0218] In specific implementation, as a preferred embodiment of the present invention, the comparison module is also used for:

[0219] Compare the decision threshold with the preset capture threshold;

[0220] When the judgment threshold is less than the preset capture threshold, the Doppler frequency shift carrier point is changed by frequency difference correction;

[0221] The digital signal to be processed is multiplied with the carrier signal at the location of the changed Doppler frequency shift carrier point to obtain the initial signal after the carrier is stripped.

[0222] In specific implementation, as a preferred embodiment of the present invention, a weak signal capture device is provided, which also includes a preset capture threshold generation module for:

[0223] Get the preset capture accuracy;

[0224] Determine the corresponding false alarm probability according to the preset capture accuracy;

[0225] According to the false alarm probability and the variance of Gaussian white noise, the corresponding capture threshold is calculated.

[0226] For the specific definition of the weak signal capture device, please refer to the definition of the weak signal capture method above, which will not be repeated here. Each module in the above-mentioned weak signal capture device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0227] In specific implementation, as a preferred embodiment of the present invention, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Fig.13 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store digital signal data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a weak signal capture method is implemented.

[0228] Those skilled in the art will understand that Fig.13 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0229] In specific implementation, as a preferred embodiment of the present invention, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0230] Construct a mathematical model of segmented hybrid stochastic resonance.

[0231] Conduct circuit design for the mathematical model of segmented hybrid stochastic resonance;

[0232] Collect the signal to be processed, and use the segmented hybrid stochastic resonance mathematical model to perform noise reduction on the signal to be processed;

[0233] And convert the analog signal into a digital signal;

[0234] Collecting digital signals to be processed;

[0235] Carrier separation is performed on the digital signal to be processed to obtain the initial signal after the carrier is stripped off;

[0236] Grouping the initial signal to obtain grouped data blocks of a preset length;

[0237] Performing coherent accumulation on data blocks of each preset length to generate corresponding coherent accumulation results;

[0238] Dividing each data block of a preset length into an odd-numbered data block and an even-numbered data block;

[0239] According to the coherent accumulation results, non-coherent accumulation is performed on the odd-numbered data blocks and the even-numbered data blocks respectively to generate corresponding non-coherent accumulation results;

[0240] Determine a larger value from the non-coherent accumulation results corresponding to the odd-numbered data blocks and the non-coherent accumulation results corresponding to the even-numbered data blocks as a determination threshold;

[0241] The judgment threshold is compared with the preset capture threshold. When the judgment threshold is less than the preset capture threshold, the process returns to the step of performing carrier separation on the digital signal to be processed to obtain the initial signal after stripping the carrier, until the judgment threshold is greater than or equal to the preset capture threshold and the signal is captured successfully.

[0242] In specific implementation, as a preferred embodiment of the present invention, the processor further implements the following steps when executing the computer program:

[0243] Obtaining a preset Doppler frequency shift carrier point;

[0244] Obtain a carrier signal sent by a local carrier generator;

[0245] The digital signal to be processed is multiplied with the carrier signal at the location of the preset Doppler frequency shift carrier point to obtain the initial signal after stripping the carrier.

[0246] In specific implementation, as a preferred embodiment of the present invention, the processor further implements the following steps when executing the computer program:

[0247] Performing fast Fourier transform processing on each data block of preset length to obtain a corresponding first transform result;

[0248] Get the local pseudo-random code;

[0249] Performing fast Fourier transform processing on the local pseudo-random code and taking conjugate, to obtain a transformed local pseudo-random code;

[0250] Multiplying the first transformation result and the transformed pseudo-random code to obtain a second transformation result;

[0251] Perform inverse fast Fourier transform processing on the second transform result to generate a corresponding coherent accumulation result.

[0252] In specific implementation, as a preferred embodiment of the present invention, the processor further implements the following steps when executing the computer program:

[0253] The coherent accumulation results of each data block in the odd-group data block are squared and summed to obtain the incoherent accumulation results of the odd-group data block; the coherent accumulation results of each data block in the even-group data block are squared and summed to obtain the incoherent accumulation results of the even-group data block.

[0254] In specific implementation, as a preferred embodiment of the present invention, the processor further implements the following steps when executing the computer program:

[0255] Compare the decision threshold with the preset capture threshold;

[0256] When the judgment threshold is less than the preset capture threshold, the Doppler frequency shift carrier point is changed by frequency difference correction;

[0257] The digital signal to be processed is multiplied with the carrier signal at the location of the changed Doppler frequency shift carrier point to obtain the initial signal after the carrier is stripped.

[0258] In specific implementation, as a preferred embodiment of the present invention, the processor further implements the following steps when executing the computer program:

[0259] Dividing the initial signal into a plurality of first data blocks having a first preset length;

[0260] Acquire each sampling point having the same phase in each first data block;

[0261] According to each sampling point having the same phase, the first data block is superimposed to form a data block having a second preset length; the first preset length is greater than the second preset length.

[0262] In specific implementation, as a preferred embodiment of the present invention, the processor further implements the following steps when executing the computer program:

[0263] Get the preset capture accuracy;

[0264] Determine the corresponding false alarm probability according to the preset capture accuracy;

[0265] According to the false alarm probability and the variance of Gaussian white noise, the corresponding capture threshold is calculated.

[0266] In specific implementation, as a preferred embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0267] Construct a mathematical model of segmented hybrid stochastic resonance.

[0268] Conduct circuit design for the mathematical model of segmented hybrid stochastic resonance;

[0269] Collect the signal to be processed, and use the segmented hybrid stochastic resonance mathematical model to perform noise reduction on the signal to be processed;

[0270] And convert the analog signal into a digital signal;

[0271] Collecting digital signals to be processed;

[0272] Carrier separation is performed on the digital signal to be processed to obtain the initial signal after the carrier is stripped off;

[0273] Grouping the initial signal to obtain grouped data blocks of a preset length;

[0274] Performing coherent accumulation on data blocks of each preset length to generate corresponding coherent accumulation results;

[0275] Dividing each data block of a preset length into an odd-numbered data block and an even-numbered data block;

[0276] According to the coherent accumulation results, non-coherent accumulation is performed on the odd-numbered data blocks and the even-numbered data blocks respectively to generate corresponding non-coherent accumulation results;

[0277] Determine a larger value from the non-coherent accumulation results corresponding to the odd-numbered data blocks and the non-coherent accumulation results corresponding to the even-numbered data blocks as a determination threshold;

[0278] The judgment threshold is compared with the preset capture threshold. When the judgment threshold is less than the preset capture threshold, the process returns to the step of performing carrier separation on the digital signal to be processed to obtain the initial signal after stripping the carrier, until the judgment threshold is greater than or equal to the preset capture threshold and the signal is captured successfully.

[0279] In specific implementation, as a preferred embodiment of the present invention, when the computer program is executed by the processor, the following steps are also implemented:

[0280] Obtaining a preset Doppler frequency shift carrier point;

[0281] Obtain a carrier signal sent by a local carrier generator;

[0282] The digital signal to be processed is multiplied with the carrier signal at the location of the preset Doppler frequency shift carrier point to obtain the initial signal after stripping the carrier.

[0283] In specific implementation, as a preferred embodiment of the present invention, when the computer program is executed by the processor, the following steps are also implemented:

[0284] Performing fast Fourier transform processing on each data block of preset length to obtain a corresponding first transform result;

[0285] Get the local pseudo-random code;

[0286] Performing fast Fourier transform processing on the local pseudo-random code and taking conjugate, to obtain a transformed local pseudo-random code;

[0287] Multiplying the first transformation result and the transformed pseudo-random code to obtain a second transformation result;

[0288] Perform inverse fast Fourier transform processing on the second transform result to generate a corresponding coherent accumulation result.

[0289] In specific implementation, as a preferred embodiment of the present invention, when the computer program is executed by the processor, the following steps are also implemented:

[0290] The coherent accumulation results of each data block in the odd-group data block are squared and summed to obtain the incoherent accumulation results of the odd-group data block; the coherent accumulation results of each data block in the even-group data block are squared and summed to obtain the incoherent accumulation results of the even-group data block.

[0291] In specific implementation, as a preferred embodiment of the present invention, when the computer program is executed by the processor, the following steps are also implemented:

[0292] Compare the decision threshold with the preset capture threshold;

[0293] When the judgment threshold is less than the preset capture threshold, the Doppler frequency shift carrier point is changed by frequency difference correction;

[0294] The digital signal to be processed is multiplied with the carrier signal at the location of the changed Doppler frequency shift carrier point to obtain the initial signal after the carrier is stripped.

[0295] In specific implementation, as a preferred embodiment of the present invention, when the computer program is executed by the processor, the following steps are also implemented:

[0296] Dividing the initial signal into a plurality of first data blocks having a first preset length;

[0297] Acquire each sampling point having the same phase in each first data block;

[0298] According to each sampling point having the same phase, the first data block is superimposed to form a data block having a second preset length; the first preset length is greater than the second preset length.

[0299] In specific implementation, as a preferred embodiment of the present invention, when the computer program is executed by the processor, the following steps are also implemented:

[0300] Get the preset capture accuracy;

[0301] Determine the corresponding false alarm probability according to the preset capture accuracy;

[0302] According to the false alarm probability and the variance of Gaussian white noise, the corresponding capture threshold is calculated.

[0303] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0304] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0305] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A weak signal noise reduction and capture method, characterized in that: Including noise reduction, carrier separation and capture steps: Step 1: Noise reduction steps: Construct a mathematical model of segmented hybrid stochastic resonance and design circuits based on the mathematical model of segmented hybrid stochastic resonance; Collect the signal to be processed, use the segmented hybrid stochastic resonance mathematical model to perform noise reduction on the signal to be processed, and perform analog-to-digital conversion on the analog signal output by the model to obtain a digital signal; Step 2: Carrier separation and capture steps Collecting a digital signal to be processed, first performing carrier separation on the digital signal to be processed to obtain an initial signal after stripping the carrier; Grouping the initial signal to obtain grouped data blocks of a preset length; Then, coherent accumulation is performed on the data blocks of each preset length to generate a corresponding coherent accumulation result, and the data blocks of each preset length are divided into an odd-numbered group of data blocks and an even-numbered group of data blocks, and then, according to the coherent accumulation result, incoherent accumulation is performed on the odd-numbered group of data blocks and the even-numbered group of data blocks respectively to generate a corresponding incoherent accumulation result; Then from the non-coherent accumulation results corresponding to the odd number of data blocks, and Determine a larger value among the non-coherent accumulation results corresponding to the even number of data blocks as a determination threshold; The determination threshold is compared with the preset capture threshold. When the determination threshold is less than the preset capture threshold, the step of performing carrier separation on the digital signal to be processed to obtain the initial signal after carrier stripping is returned until the determination threshold is greater than or equal to the preset capture threshold and the signal is captured successfully.

2. The weak signal processing method according to claim 1, characterized in that: In step 1, the circuit adds two sets of comparison feedback circuits to the original bistable stochastic resonance model, which are composed of amplification and summation, integration and three comparison feedback circuits. The three feedback circuits judge the size of the system output signal to form a segmented hybrid stochastic resonance mathematical model for signal noise reduction.

3. The weak signal processing method according to claim 2, characterized in that: In step 2, a preset Doppler frequency shift carrier point and a carrier signal emitted by a local carrier generator are obtained, and the digital signal to be processed is multiplied by the carrier signal at the location of the preset Doppler frequency shift carrier point to obtain an initial signal after stripping the carrier, and the initial signal is divided according to a first preset length to obtain each sampling point with the same phase, and based on this, the first data block is superimposed to form a data block with a length of a second preset length. Perform fast Fourier transform processing on data blocks of each preset length to obtain a local pseudo-random code, then perform fast Fourier transform processing and take conjugate, multiply the local pseudo-random code and the transformed pseudo-random code, and then perform inverse fast Fourier transform processing on the result to generate a corresponding coherent accumulation result, and then square the coherent accumulation results of each data block in the odd-array data block and the even-array data block respectively, and then sum them to obtain incoherent accumulation results of the odd-array data block and the even-array data block respectively. The preset capture accuracy is obtained to determine the corresponding false alarm probability, and then the corresponding capture threshold is calculated according to the false alarm probability and the variance of Gaussian white noise. When the judgment threshold is less than the preset capture threshold, the Doppler frequency shift carrier point is changed by frequency difference correction, and the digital signal to be processed is multiplied by the carrier signal at the position of this point to obtain the initial signal after stripping the carrier.

4. The weak signal processing method according to claim 3, characterized in that: In step 1: constructing a weak signal detection model of segmented hybrid stochastic resonance, including the following steps: Step S102, constructing a segmented hybrid stochastic resonance mathematical model. The one-dimensional continuous nonlinear bistable system model described by the Langevin equation is shown in formula (1): in, H(t) is the input signal, η(t) is the noise signal, a L and b L is a real number greater than 0; U L (x) is the potential function of the system: The Langevin equation of the piecewise mixed model is shown in formula (3): Among them, U(x) is the potential function of the piecewise hybrid model, H(t) is the input signal, η(t) is the noise signal, K is the amplification factor, and the expression of U(x) is shown in formula (4): Among them, a, b, c are all real numbers, and a>0, b>0, Comparing equations (2) and (4), we can see that the piecewise hybrid model is composed of two linear equations and one nonlinear equation, where the middle nonlinear equation is a bistable system equation; the potential function of the piecewise hybrid model also has two potential wells, with the potential bottom located at The left and right outer curves intersect the horizontal axis at Pick Step S104, performing circuit design on the segmented hybrid stochastic resonance mathematical model; Substituting formula (4) into formula (3), we can get: In formula (5), s(t)=H(t)+η(t), and U′(x) represents dU(x) / dx. According to formula (5), the system consists of amplification, summation, integration and comparison feedback parts; select appropriate resistors and capacitors, adjust the resistance of the variable potentiometer, so that a=2, b=1, c=2, K=10, and get formula 6. According to formula 6, the circuit is obtained; 5. The weak signal processing method according to claim 3, characterized in that: The capture method in step 2 is: Step S202, collecting the digital signal to be processed: The collected digital signal to be processed is a weak signal, wherein the weak signal refers to a digital signal with a signal-to-noise ratio of less than -22 dB, wherein the source of the digital signal to be processed is the BeiDou Navigation Satellite System (BDS); The collected weak signal is an intermediate frequency digital signal. The following formula is used to express the collected digital signal to be processed: Where r(n) represents the intermediate frequency digital signal, A is the signal amplitude, d(n) is the navigation data, c(n) is the local C / A code, ω0 is the intermediate frequency carrier angular frequency, ω d is the angular frequency of the Doppler shift, is the initial phase of the carrier, f(n) is Gaussian white noise, n=0, 1, 2…N-1, represents the number of sampling points of the digital signal to be processed; Step S204, performing carrier separation on the digital signal to be processed to obtain an initial signal after the carrier is stripped; By obtaining a preset Doppler frequency shift carrier point and a carrier signal emitted by a local carrier generator, the digital signal to be processed and the carrier signal are multiplied at the location of the preset Doppler frequency shift carrier point to obtain an initial signal after stripping the carrier; The following formula is used to represent the initial signal after stripping the carrier: x(n)=I(n)+jQ(n); Wherein, x(n) represents the initial signal after carrier stripping, I(n) represents the local carrier signal of the first branch, Q(n) represents the local carrier signal of the second branch, and j represents that the local carrier signal of the first branch is orthogonal to the local carrier signal of the second branch; By multiplying the digital signal to be processed with the carrier signal at the location of the preset Doppler frequency shift carrier point, the local carrier is stripped out, and two branch signals I(n) and Q(n) are obtained. I(n) and Q(n) are respectively used as the real part and imaginary part of the complex number to construct the initial signal after stripping the carrier. The following two formulas are used to represent the local carrier signal of the first branch and the local carrier signal of the second branch respectively: in, represents an estimate of the angular frequency of the Doppler shift, represents an estimate of the initial phase of the wave; Step S206, grouping the initial signal to obtain grouped data blocks of a preset length; The initial signal after stripping the carrier is grouped into several first data blocks with a first preset length, and then superimposed into multiple second data blocks with a second preset length; wherein the first preset length is greater than the second preset length, the first preset length can be 10ms, and the second preset length can be 1ms. Step S208, performing coherent accumulation on the data blocks of each preset length to generate corresponding coherent accumulation results; By performing fast Fourier transform processing on each data block of a preset length, a corresponding first transform result is obtained. By obtaining a local pseudo-random code, performing fast Fourier transform processing on the local pseudo-random code and taking a conjugate, a transformed local pseudo-random code is obtained. Then, the first transform result and the transformed pseudo-random code are multiplied to obtain a second transform result, and the second transform result is subjected to inverse fast Fourier transform processing to generate a corresponding coherent accumulation result; The data blocks of each preset length are further subjected to fast Fourier transform processing, and the first transform result obtained is expressed by the following formula: Where k represents the number of data blocks, k = 1, 2, L, 2N, M is the number of digital sampling points in a 1ms data block, X k (M) is the result of fast Fourier transform of the signal after stripping the carrier; After the local pseudo-random code is fast Fourier transformed and conjugated, it is expressed by the following formula: Among them, C k (M) represents the result of fast Fourier transform of local pseudo-random code; The above two formulas are multiplied to obtain the second transformation result, and the second transformation result is subjected to inverse fast Fourier transform processing, and the obtained coherent accumulation result is expressed by the following formula: Among them, Y k (M) represents the result of performing inverse fast Fourier transform processing on the second transform result, that is, the coherent accumulation result. k is the coherent integration result of the kth data block, Y k (1) is the result of the first sampling point in the k-th data block. The coherent integration result of the k-th data block is composed of M sampling points. The gain obtained by coherent accumulation is expressed by the following formula: G c (f)=10ilg(f); Among them, G c (f) represents the gain obtained by coherent accumulation, and f is relative to the data length of 1 ms; Step S210, dividing each data block of a preset length into an odd-numbered group of data blocks and an even-numbered group of data blocks; The data blocks of each preset length corresponding to the coherent accumulation result are divided into odd-numbered data blocks and even-numbered data blocks by using a half-bit method. Since the received satellite intermediate frequency signal has a bit flip at most once every 20ms, if the signal flip occurs within the first 10ms, the data in the next 10ms will not be affected by the bit flip. Therefore, the half-bit method is used to group the data blocks to avoid the capture result being affected by the bit flip. Step S212, performing incoherent accumulation on the odd-numbered data blocks and the even-numbered data blocks respectively according to the coherent accumulation result, to generate corresponding incoherent accumulation results; The coherent accumulation results of each data block in the odd-numbered data blocks are squared and summed up to obtain the incoherent accumulation results of the odd-numbered data blocks. Similarly, the coherent accumulation results of each data block in the even-numbered data blocks are squared and summed up to obtain the incoherent accumulation results of the even-numbered data blocks; The following formula is used to express the accumulation result obtained by non-coherently accumulating odd-group data blocks and even-group data blocks respectively using the half-bit method: Among them, Y A Indicates the incoherent accumulation result corresponding to the odd-numbered data blocks, Y B represents the non-coherent accumulation result corresponding to an even number of data blocks, and k represents the number of data blocks; The gain obtained by incoherent accumulation is expressed by the following formula: G fc (f)=G c (f)-L(f)=10lg(f)-L(f); Among them, f is the data length relative to 1ms, G fc (f) represents the gain obtained by incoherent accumulation, and L(f) represents the loss of incoherent accumulation, which is expressed by the following formula: Among them, D c It represents the ideal detection factor under a specific false alarm probability and a specific detection probability. -7 , when the detection probability is 0.9, D c ≈21. In order to ensure that the capture sensitivity and signal-to-noise ratio are not reduced, the accumulated coherent accumulation length needs to be strictly controlled; Step S214, determining a larger value from the non-coherent accumulation results corresponding to the odd-numbered data blocks and the non-coherent accumulation results corresponding to the even-numbered data blocks as a determination threshold; Take Y=max(Y A ,Y B ), that is, determine the larger value from the incoherent accumulation results corresponding to the odd-numbered data blocks and the incoherent accumulation results corresponding to the even-numbered data blocks, and use the obtained maximum value as the decision threshold. The decision threshold is used to be compared with the preset capture threshold to determine the final weak signal capture result.

6. A weak signal noise reduction and capture device, characterized in that: The device comprises: Signal noise reduction device: The signal noise reduction and amplification module collects weak signals and processes them through the segmented chaotic stochastic resonance mathematical model to obtain the noise-reduced amplified signal; The analog-to-digital conversion module is used to convert the processed analog signal into the required digital signal; Signal capture device: Digital signal acquisition module, carrier separation module, initial signal grouping module, coherent accumulation module, data block grouping module, incoherent accumulation module, judgment threshold determination module and comparison module. The method described in any one of claims 1 to 5 is used to implement weak signal processing.

Images