A high dynamic stable tracking method for non-spread spectrum continuous wave signals

Abstract

The application belongs to the technical field of wireless communication, and specifically discloses a high-dynamic stable tracking method for a non-spread spectrum continuous wave signal. The method comprises the following steps: obtaining a digital signal based on ADC sampling, and performing digital frequency conversion on all sampling points of a current symbol to obtain a digital baseband signal according to a current estimated frequency result; obtaining a digital baseband signal under 4 times oversampling through multi-stage downsampling, and performing matched filtering processing on the digital baseband signal under 4 times oversampling; determining the output result of a preset position in the matched filtering output as an optimal sampling point to serve as the output of a tracking loop; performing timing error estimation according to the output result of the matched filtering, and adjusting the sampling deviation of next data by using the timing error estimation result; and obtaining a phase discrimination value according to the optimal sampling point, and adjusting the digital frequency conversion of the next data by using the phase discrimination value. The scheme solves the technical problem that the non-spread spectrum continuous wave signal cannot be continuously and stably tracked in a high-dynamic scene.

Description

Technical Field

[0001] This invention belongs to the technical field of wireless communication, specifically relating to a high dynamic stability tracking method for non-spread spectrum continuous wave signals. Background Technology

[0002] The relative motion between the transmitting and receiving parties causes the Doppler effect. As the relative velocity and acceleration between the two parties increase, the Doppler frequency offset and frequency offset acceleration of the received signal increase, and the code offset of both parties also changes. Besides being related to relative motion, the code offset is also related to the inconsistency between the crystal oscillators of the transmitting and receiving parties. If the receiver cannot accurately and stably track the changes in the frequency offset and code offset of the received signal, it will affect the receiver's demodulation and may even cause communication failure or interruption.

[0003] Currently, most common high dynamic range tracking (HMR) methods are based on spread spectrum systems. The core idea is a carrier loop-assisted code loop, where the NCO estimated by the carrier loop is fed back to the code loop to help it eliminate the effects of high dynamic range. Carrier loops include phase-locked loops (PLLs) and frequency-locked loops (FLLs). However, non-spread spectrum systems cannot recover the original information using the code tracking loop, therefore the aforementioned algorithms cannot be used.

[0004] Common carrier synchronization algorithms for non-spread spectrum systems often assume that the frequency offset remains constant over a period of time, making them suitable for situations where the residual frequency offset is less than the symbol rate. However, in high-dynamic scenarios, where the received signal's frequency offset is greater than the symbol rate, current algorithms cannot guarantee stable tracking of non-spread spectrum continuous wave signals. Summary of the Invention

[0005] The purpose of this invention is to provide a high dynamic stability tracking method for non-spread spectrum continuous wave signals, so as to solve the technical problem that non-spread spectrum continuous wave signals cannot be continuously and stably tracked in high dynamic scenarios.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A high dynamic stability tracking method for non-spread spectrum continuous wave signals includes: obtaining a digital signal based on ADC sampling, and performing digital frequency conversion on all sampling points of the current symbol to obtain a digital baseband signal according to the current estimated frequency result; obtaining a digital baseband signal under 4x oversampling after multi-stage downsampling, and performing matched filtering on the digital baseband signal under 4x oversampling, determining the output result at a preset position in the matched filter output as the optimal sampling point, and using it as the output of the tracking loop; estimating the timing error based on the output result of the matched filter, and adjusting the sampling deviation of the next segment of data using the timing error estimation result; simultaneously obtaining the phase detection value based on the optimal sampling point, and adjusting the digital frequency conversion of the next segment of data using the phase detection value.

[0008] A high dynamic stability tracking system for non-spread spectrum continuous wave signals includes: a digital frequency conversion module for obtaining a digital signal based on ADC sampling, and performing digital frequency conversion on all sampling points of the current symbol to obtain a digital baseband signal according to the currently estimated frequency result; a matched filtering module for obtaining a digital baseband signal under 4x oversampling after multi-stage downsampling, and performing matched filtering on the digital baseband signal under 4x oversampling, determining the output result at a preset position in the matched filter output as the optimal sampling point, which is used as the output of the tracking loop; a timing error estimation module for estimating the timing error based on the output result of the matched filter, and adjusting the sampling deviation of the next data segment using the timing error estimation result; and a phase detector module for obtaining the phase detection value based on the optimal sampling point, and adjusting the digital frequency conversion of the next data segment using the phase detection value.

[0009] A computer-readable storage medium storing a computer program configured to execute the above-described high dynamic stability tracking method for non-spread spectrum continuous wave signals at runtime.

[0010] An electronic device includes a memory and a processor, wherein the memory stores a computer program and the processor is configured to execute the high dynamic stability tracking method for the non-spread spectrum continuous wave signal through the computer program.

[0011] In this invention, a digital signal is obtained based on ADC sampling, and a digital baseband signal is obtained by digitally converting all sampling points of the current symbol according to the current estimated frequency result. After multi-stage downsampling, a digital baseband signal with 4x oversampling is obtained, and matched filtering is performed on the digital baseband signal with 4x oversampling. The output result at a preset position in the matched filter output is determined as the optimal sampling point, which is used as the output of the tracking loop. Timing error is estimated based on the output result of the matched filter, and the sampling deviation of the next data segment is adjusted using the timing error estimation result. At the same time, the phase detection value is obtained based on the optimal sampling point, and the phase detection value is used to adjust... The next segment of data undergoes digital frequency conversion, which gradually improves the signal-to-noise ratio of the received signal through multi-stage filtering and decimation, thereby enhancing the accuracy of timing deviation estimation. The sampling deviation is adjusted based on the timing deviation estimation results, and the number of sampling points for the symbol is adjusted at high oversampling multiples to reduce the phase error caused by adjusting the number of sampling points. Simultaneously, a combination of code offset tracking loop and frequency offset tracking loop is used to achieve stable tracking and compensation of code offset and frequency offset in high dynamic scenarios for non-spread spectrum continuous wave signals. This solves the technical problem of the inability to achieve continuous and stable tracking of non-spread spectrum continuous wave signals in high dynamic scenarios, achieving the technical effect of continuous and stable tracking of non-spread spectrum continuous wave signals in high dynamic scenarios. Attached Figure Description

[0012] Figure 1This is a flowchart illustrating a high dynamic stability tracking method for non-spread spectrum continuous wave signals according to an embodiment of the present invention.

[0013] Figure 2 This is a schematic diagram of the implementation process of a high dynamic stability tracking method for non-spread spectrum continuous wave signals in an embodiment of the present invention;

[0014] Figure 3 This is a schematic diagram of the constellation diagram simulation results of the matched filter output in an embodiment of the present invention;

[0015] Figure 4 This is a schematic diagram of the timing error estimation process in an embodiment of the present invention;

[0016] Figure 5 This is a schematic diagram showing the number of sampling points for each symbol after timing deviation adjustment in an embodiment of the present invention;

[0017] Figure 6 This is a schematic diagram of the phase-locked loop phase detector in an embodiment of the present invention;

[0018] Figure 7 This is a schematic diagram of the output result of the phase-locked loop phase detector in an embodiment of the present invention;

[0019] Figure 8 This is a schematic diagram of a high dynamic stability tracking system for non-spread spectrum continuous wave signals according to an embodiment of the present invention. Detailed Implementation

[0020] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become clearer from the following description and claims. It should be noted that the drawings are all in a very simplified form and are not to a precise scale, and are only used to facilitate and clarify the illustration of the embodiments of the present invention.

[0021] It should be noted that, in order to clearly illustrate the content of this invention, several embodiments are provided to further explain different implementations of the invention. These embodiments are enumerated rather than exhaustive. Furthermore, for the sake of brevity, content mentioned in the preceding embodiments is often omitted in the following embodiments. Therefore, content not mentioned in the later embodiments can be referred to in the preceding embodiments.

[0022] Example 1

[0023] A high dynamic stability tracking method for non-spread spectrum continuous wave signals, such as Figure 1 As shown, the method includes:

[0024] S102: Based on ADC sampling, a digital signal is obtained, and according to the current estimated frequency result, all sampling points of the current symbol are digitally converted to obtain a digital baseband signal.

[0025] S104, after multiple downsampling stages, obtains a digital baseband signal under 4x oversampling, and performs matched filtering on the digital baseband signal under 4x oversampling. The output result at the preset position in the matched filter output is determined as the optimal sampling point, which is used as the output of the tracking loop.

[0026] S106: Based on the output of the matched filter, the timing error is estimated, and the sampling deviation of the next data segment is adjusted using the timing error estimation result. At the same time, the phase detection value is obtained based on the optimal sampling point, and the digital frequency conversion of the next data segment is adjusted using the phase detection value.

[0027] As an optional implementation, after obtaining the digital signal based on ADC sampling, the method further includes: buffering the digital signal obtained by ADC sampling in RAM, wherein the address of the data stored in RAM is incremented by one for each input sampling point; and reading the corresponding sampling point from RAM.

[0028] Specifically, the high dynamic stability tracking method for non-spread spectrum continuous wave signals is not limited to including the following steps:

[0029] S1 buffers the digital signal sampled by the ADC in RAM, and reads a certain number of sampling points from RAM when needed.

[0030] S2, based on the current estimated frequency result, perform digital frequency conversion on all sampling points of the current symbol to achieve frequency offset compensation, eliminate the frequency offset effect caused by high dynamics, and obtain the digital baseband signal.

[0031] S3, through multiple levels of downsampling, gradually reduces the sampling rate of the digital baseband signal, eventually reducing it to 4 times oversampling.

[0032] S4 performs matched filtering on the digital baseband signal oversampled by 4x, and is not limited to determining the 0th, 4th, 8th, 12th, 16th, ... output points as the optimal sampling points. The optimal sampling points are then used as the output of the tracking loop and fed into subsequent decoding modules.

[0033] S5. Based on the output of the matched filter, perform timing error estimation, convert the timing error estimation result into the number of sampling points to be adjusted, and feed it back to step S1 to adjust the sampling deviation of the next segment of data, thereby achieving stable tracking of the code deviation of the high dynamic BPSK continuous signal.

[0034] S6: Based on the optimal sampling point of the matched filter output in step S4, estimate the residual phase deviation to obtain the phase detection value. After passing through the loop filter, the phase detection value is fed back to the digital frequency conversion in step S2 to construct a phase-locked loop feedback loop, thereby achieving stable tracking of the frequency deviation of the high dynamic BPSK continuous signal.

[0035] As an optional implementation, reading the corresponding sampling point from RAM includes: determining the starting address of the first sampling point of the first symbol, and determining a preset number of consecutive sampling points as a symbol based on the starting address; updating the address information of the current sampling point according to the address of the sampling point of the previous symbol and the timing error estimation result, and reading the corresponding sampling point from RAM.

[0036] The address of the sampling point read from RAM is not limited to determining the starting address of the first sampling point of the first symbol based on the capture result. Starting from that address, OSR samples are read consecutively as one symbol, where OSR represents the oversampling factor of the received signal. Based on the address of the sampling point of the previous symbol and the address adjustment amount obtained from the timing error estimation module, the address information of the current sampling point is updated, and the corresponding sampling point is read from RAM.

[0037] Furthermore, in addition to determining the starting address and number of sampling points of the current symbol before reading, this process specifically includes:

[0038] The starting address of the current symbol is updated based on the starting address of the previous symbol and the address adjustment amount obtained from the timing error estimation module.

[0039] Update the number of sampling points for the current symbol based on the starting position of the current symbol and the starting position of the previous symbol.

[0040] Based on the starting address of the current symbol and the number of sampling points of the current symbol, all sampling points of the current symbol can be read from RAM.

[0041] As an optional implementation, a digital baseband signal with 4x oversampling is obtained by multi-level downsampling, including: reducing the sampling rate of the digital baseband signal by a preset factor step by step, and reducing the sampling rate to 4 times the symbol rate through multi-level downsampling of the preset factor to obtain a digital baseband signal with 4x oversampling.

[0042] As an optional implementation, the sampling rate of the digital baseband signal is reduced by a preset factor step by step, including: obtaining the first filtering result after inputting a preset number of sampling points for the current symbol based on the number of sampling points of the current symbol; performing convolution operation when outputting the filtering result at the current sampling point; and outputting a filtering result for each preset number of input sampling points starting from the second segment of data of the current symbol.

[0043] Downsampling involves multi-stage filtering and decimation, where the first stage of filtering and decimation needs to consider the impact of the sampling point number not being equal to the oversampling factor (OSR). Specifically, it includes the following steps:

[0044] After downsampling by a factor of z, the sampling rate is reduced to OSR / z times the symbol rate. Considering the influence of code offset, the number of sampling points for each symbol is not necessarily OSR. The number of sampling points for the nth symbol oscillates around the OSR, and the specific oscillation range is related to the crystal oscillator stability and relative motion velocity.

[0045] After multiple z-fold downsampling stages, the sampling rate is reduced to four times the symbol rate.

[0046] Furthermore, code bias causes the number of sampling points for each symbol to be not necessarily OSR, and the first filtering and decimation process for each symbol is inconsistent with subsequent processes, specifically including the following steps:

[0047] Based on the number of sampling points, the first filtered result can be obtained after taking m sampling points from the current symbol input. There is a corresponding relationship between m and the number of sampling points.

[0048] When the current sampling point outputs the filtered result, a convolution operation is performed. Let the filter order be N. fir The filter coefficients are fir_coe0, fir_coe1, fir_coe2, ..., fir_coe Nfir-1 If the current sampling point is the k-th point, the first-stage filter outputs the l-th result.

[0049] The current symbol starts from the second segment, and a filtering result is output for every z input sampling points.

[0050] As an optional implementation, timing error estimation is performed based on the output of matched filtering, and the sampling deviation of the next data segment is adjusted using the timing error estimation result. This includes: performing timing error estimation based on the output of matched filtering, converting the timing error estimation result into the number of sampling points to be adjusted, and adjusting the sampling deviation of the next data segment.

[0051] As an optional implementation, timing error estimation is performed based on the output of matched filtering, and the timing error estimation result is converted into the number of sampling points to be adjusted and the sampling deviation of the next data segment is adjusted. This includes: using the output of matched filtering to perform timing error estimation and obtaining the normalized symbol timing error estimation result; calculating the sampling deviation of the next data segment based on the timing error estimation result and determining the corresponding address adjustment amount.

[0052] The normalized symbol timing error is estimated based on the output of the matched filter, and the obtained normalized symbol timing error estimate is used to compensate for the timing deviation of the next L symbols, which is the address adjustment amount.

[0053] As an optional implementation, the phase detection value is obtained based on the optimal sampling point, and the digital frequency conversion of the next data segment is adjusted using the phase detection value. This includes: inputting the optimal sampling point into the phase detector to obtain the phase detection value; inputting the phase detection value into the loop filter; and using the frequency offset estimate of the loop filter output to control the digital frequency conversion frequency of the next data segment.

[0054] As an optional implementation, the optimal sampling point is input into the phase detector to obtain the phase detection value, including: inputting the optimal sampling point into the arctan arctangent phase detector, compensating the phase detection result, and compensating again after the loop is locked to obtain the phase detection value that eliminates sudden phase detection errors.

[0055] Phase detection is performed based on the optimal sampling point of the matched filter output. The phase detection result is then fed back to the digital frequency converter after loop filtering. The specific steps include:

[0056] S6.1: Input the optimal sampling point of the matched filter output in step S4 into the phase detector to obtain the phase detection value.

[0057] S6.2: The phase detection result is input into the loop filter, and the output of the loop filter is fed back to step S2 to control the frequency of the digital frequency converter. The phase detection result is updated once for each symbol, and the frequency deviation estimate of the loop filter output is updated accordingly and fed back to the digital frequency converter module to eliminate the frequency deviation of the next symbol and eliminate the frequency deviation effect caused by high dynamics.

[0058] Considering that the frequency offset estimate is updated once per symbol, and step S2 performs frequency offset compensation for each sampling point, the result fcw fed back to S2 is... n With frequency offset estimation result fd n There is a corresponding relationship, which distributes the phase deviation caused by the frequency offset within a symbol to each sampling point.

[0059] In step S6.1, the phase detection process includes arctan calculation and two phase compensations, specifically including the following steps:

[0060] S6.1.1: An arctan phase detector is used, which is not sensitive to 180° phase jumps and is suitable for BPSK signals.

[0061] S6.1.2: Compensate for the obtained phase detection result to overcome the limitation of the arctan arctangent phase detection result from -π / 2 to π / 2. S6.1.3: After the loop is locked, further compensate for the obtained phase to reduce sudden phase detection errors caused by noise under low signal-to-noise ratio conditions and improve loop stability. No compensation is performed when the loop is not locked. After the loop is locked, it is assumed that each phase deviation will not exceed the limitation of -π / 2 to π / 2, therefore compensation is performed to limit the phase detection result to the range of -π / 2 to π / 2.

[0062] In a specific embodiment, the communication system uses a non-spread spectrum BPSK continuous wave signal, with a symbol rate denoted as R. s The sampling rate is denoted as f. s The Doppler frequency shift caused by high dynamics is in [-f dmax ,f dmax Within the range of [-f], the frequency offset acceleration is in the range of [-f] amax ,f amax Within the range. Wherein, the sampling rate f s =40000×R s This is 40,000 times the symbol rate, where OSR = 40,000 represents the oversampling factor, f dmax =780×R s The maximum frequency range is greater than the symbol rate, f amax =16×R s The maximum frequency offset acceleration is greater than the symbol rate. The frequency offset exhibits a sinusoidal waveform, with the frequency offset acceleration being minimum when the frequency offset is maximum and maximum when the frequency offset is minimum, which aligns with real-world satellite communication scenarios.

[0063] The high dynamic stability tracking method for non-spread spectrum continuous wave signals in this embodiment is not limited to, for example... Figure 2 As shown, it includes: S1, data buffering; S2, digital frequency conversion; S3, multi-stage decimation filtering; S4, matched filtering; S5, timing error estimation; S6, phase detection and loop filtering. Specifically, the steps are as follows:

[0064] S1: Buffer the digital signal sampled by the ADC in RAM, and read a certain number of sampling points from RAM when needed.

[0065] Furthermore, S1.1: The digital signal sampled by the ADC is buffered in RAM. Specifically, for each input sampling point, the address of the data stored in RAM is incremented by one.

[0066] S1.2: Read a certain number of sampling points from RAM when needed. Specifically, the addresses of the sampling points in RAM are as follows:

[0067] Furthermore, S1.2.1: Determine the starting address of the first sampling point of the first symbol based on the capture result, and continuously read OSR=40000 sampling points as one symbol starting from this address.

[0068] S1.2.2: Update the address information of the current sampling point based on the address of the previous symbol sampling point and the address adjustment amount obtained by the timing error estimation module, and read the corresponding sampling point from RAM.

[0069] Further, S1.2.2.1: Update the starting address of the current symbol based on the starting address of the previous symbol and the address adjustment amount obtained from the timing error estimation module. Specifically, the update formula is:

[0070]

[0071] In the above formula, AddrStart n AddrStart represents the starting address of the nth symbol. n-1 AddrAdjust represents the starting address of the (n-1)th symbol. n This represents the address adjustment amount estimated by the timing error estimation module. This indicates a floor operation. According to S1.2.1, the starting address of the first symbol is given by the capture result; starting from the second symbol, the starting position is calculated by equation (1).

[0072] S1.2.2.2: Update the number of sampling points for the current symbol based on the starting position of the current symbol and the starting position of the previous symbol. Specifically, the update formula is:

[0073] ΔSampleNum n =AddrStart n -AddrStart n-1 (2)

[0074] In the above formula, ΔSampleNum n This represents the number of sampling points for the nth symbol. According to S1.2.1, the number of sampling points for the first symbol is OSR = 40000; starting from the second symbol, the number of sampling points is calculated by equation (2).

[0075] S1.2.2.3: Based on the starting address of the current symbol, AddrStart n And the number of sampling points ΔSampleNum for the current symbol n This allows the system to read all sample points of the current symbol from RAM.

[0076] S2: Based on the current estimated frequency result, perform digital frequency conversion on all sampling points of the current symbol to achieve frequency offset compensation, eliminate the frequency offset effect caused by high dynamics, and obtain the digital baseband signal.

[0077] S3: After multiple levels of downsampling, the sampling rate of the digital baseband signal is gradually reduced, eventually reaching 4 times oversampling.

[0078] Furthermore, S3.1: After a 10x downsampling, the sampling rate is reduced to 4000 times the symbol rate. Considering the impact of code bias, the number of sampling points per symbol is not necessarily OSR. The number of sampling points ΔSampleNum for the nth symbol... n It oscillates around OSR=40000, and the specific oscillation range is related to the crystal stability and relative motion speed.

[0079] Furthermore, S3.1.1: According to ΔSampleNum n The value is obtained by taking m samples of the current symbol as input and then outputting the first filtered result. If the current symbol has ΔSampleNum n =osr=40000 sampling points, then the first-stage filter of the current symbol outputs a filtering result every 10 sampling points; if the current symbol has ΔSampleNum n =osr-1=39999 sampling points, then the first-stage filter of the current symbol outputs a filtering result every 9 sampling points for the first time; if the current symbol has ΔSampleNum n =osr+1=40001 sampling points, then the first-stage filter of the current symbol outputs a filtering result every 11 sampling points for the first time; ΔSampleNum n The same logic applies when taking other values. (m and ΔSampleNum) n The specific correspondence is as follows:

[0080]

[0081] S3.1.2: When the current sampling point needs to output the filtered result, a convolution operation is performed. Let the filter order be N. fir The filter coefficients are If the current sampling point is the k-th point, the first-level filter outputs the l-th result. The convolution operation process is as follows:

[0082]

[0083] S3.1.3: Starting from the second segment, the current symbol outputs a filtering result for every 10 input sampling points. After adjustment in S3.1.1, the current symbol has 39990 sampling points remaining, and a total of 3999 filtering results are output. The filtering process is a convolution operation, as shown in equation (4).

[0084] S3.2: After a 10x downsampling, the sampling rate is reduced to 400 times the symbol rate. After the 10x decimation filtering in step S3.1, there are still 4000 sampling points left for the current symbol. After another 10x downsampling, the second-stage decimation filtering outputs a total of 400 filtered results. The filtering process is a convolution operation, which is the same as shown in equation (4).

[0085] S3.3: After a 10x downsampling, the sampling rate is reduced to 40 times the symbol rate. After the 10x decimation filtering in step S3.2, there are still 400 sampling points left for the current symbol. After another 10x downsampling, the third-stage decimation filtering outputs a total of 40 filtered results. The filtering process is a convolution operation, which is the same as shown in equation (4).

[0086] S3.4: After a 10x downsampling, the sampling rate is reduced to 4 times the symbol rate. After the 10x decimation filtering in step S3.2, there are still 400 sampling points left for the current symbol. After another 10x downsampling, the fourth-stage decimation filtering outputs a total of 4 filtering results. The filtering process is a convolution operation, which is the same as shown in equation (4).

[0087] S4: Perform matched filtering on the digital baseband signal oversampled by 4x. The 0th, 4th, 8th, 12th, 16th, ... output of the matched filter are the optimal sampling points. These optimal sampling points are used as the output of the tracking loop and fed into subsequent decoding modules. The optimal sampling points of the matched filter output are not limited to... Figure 3 As shown, the specific result is a constellation diagram simulation.

[0088] S5: Based on the matched filtering result, estimate the timing error, convert the timing error estimate into the number of sampling points to be adjusted, and feed it back to step S1.2 to adjust the sampling deviation of the next data segment, thereby achieving stable tracking of the code offset of the high-dynamic BPSK continuous signal. Specific steps are not limited to... Figure 4 As shown.

[0089] Furthermore, S5.1: using x i The normalized symbol timing error estimation formula is given by the output of the matched filter.

[0090]

[0091] As shown in equation (5), the normalized symbol timing error is updated every 1024 symbols (corresponding to 1024×4 sampling points), and the estimation result is in the range of -0.5 to 0.5. Due to the influence of relative motion, the normalized symbol timing error will change with time. The error can be regarded as constant in a short period of time. Here, it is assumed that the timing error within 1024 symbols will not change.

[0092] S5.2: Based on the normalized symbol timing error estimation result obtained in S5.1, compensate for the timing deviation of the next 1024 symbols, that is, calculate the address adjustment required in step S1.2.2.1. The specific calculation formula is as follows:

[0093]

[0094] The range of n is (b+1)×1024≤n≤(b+2)×1024-1. The address adjustment for each segment of the received signal uses the normalized symbol timing error value obtained from the previous segment. The initial time is not limited to the assumption that AddrAdjust... n Equivalent to OSR, the equivalent initial time is assumed to have a timing error estimate of 0.

[0095] After timing deviation adjustment, the number of sampling points for each symbol is not limited to, for example, Figure 5 As shown.

[0096] S6: Based on the optimal sampling point output from the matched filter in step S4, estimate the residual phase deviation to obtain the phase detection value. This phase detection value, after passing through a loop filter, is fed back to the digital frequency conversion in step S2 to construct a phase-locked loop feedback loop, achieving stable tracking of the frequency offset of the high-dynamic BPSK continuous signal. Specific steps are not limited to... Figure 6 As shown

[0097] Further, S6.1: Input the optimal sampling point of the matched filter output in step S4 into the phase detector module to obtain the phase detection value.

[0098] Specifically, S6.1.1: Employs an arctan tangent phase detector, which is insensitive to 180° phase transitions and is suitable for BPSK signals.

[0099] θ1(n)=arctanx n (7)

[0100] In equation (7), x n This represents the optimal sampling point of the matched filter output; there is one optimal sampling point for each symbol.

[0101] S6.1.2: Compensate the phase detection result obtained by equation (7) to break through the limitation of arctan arctangent phase detection result from -π / 2 to π / 2. The compensation formula is as follows:

[0102]

[0103] S6.1.3: After the loop is locked, the phase obtained by equation (8) is further compensated to reduce the sudden phase detection error caused by noise under low signal-to-noise ratio and improve the stability of the loop. The compensation formula is as follows:

[0104]

[0105] When the loop is not locked, the compensation shown in equation (9) is not performed. After the loop is locked, it is assumed that the phase deviation of each time will not exceed the limit of -π / 2 to π / 2, so compensation is required to limit the phase detection result to the range of -π / 2 to π / 2.

[0106] S6.2: The phase detection result is input into the loop filter, and the output of the loop filter is fed back to step S2 to control the frequency of the digital frequency converter. The phase detection result is updated once for each symbol, and the frequency offset estimate of the loop filter output is updated accordingly. This estimate is fed back to the digital frequency converter module to eliminate the frequency deviation of the next symbol, thus eliminating the frequency offset effect caused by high dynamics. Considering that the frequency offset estimate is updated once per symbol, and step S2 needs to perform frequency offset compensation for each sampling point, the result fcw fed back to S2... n With frequency offset estimation result fd nThe relationship is:

[0107]

[0108] Equation (10) is used to distribute the phase deviation caused by the frequency offset within a symbol equally to each sampling point.

[0109] The output of the phase-locked loop phase detector is not limited to, for example, Figure 7 As shown.

[0110] In this embodiment, multi-stage filtering and decimation are used to gradually improve the signal-to-noise ratio (SNR) of the received signal, ensuring the accuracy of timing deviation estimation. The timing deviation estimate is re-estimated periodically, and the number of sampling points for each symbol is adjusted accordingly, achieving stable tracking and compensation of code offset in high-dynamic scenarios for non-spread spectrum continuous wave signals. Adjusting the number of sampling points for each symbol at high oversampling factors reduces the phase error caused by adjusting the number of sampling points, thereby reducing the impact on the carrier loop and improving the stability of the carrier tracking loop, achieving stable tracking and compensation of frequency offset in high-dynamic scenarios. The combination of code offset tracking loop and frequency offset tracking loop achieves stable tracking and compensation of code offset and frequency offset in high-dynamic scenarios for non-spread spectrum continuous wave signals. This method can be used for reliable tracking under low SNR conditions and for demodulation of non-spread spectrum continuous wave signals. This embodiment is simple to implement, highly operable and portable, and requires few resources, making it very suitable for FPGA implementation and offering good engineering benefits.

[0111] Example 2

[0112] A high dynamic stability tracking system for non-spread spectrum continuous wave signals, such as Figure 8 As shown, the system includes:

[0113] The digital frequency conversion module 802 is used to obtain a digital signal based on ADC sampling, and to perform digital frequency conversion on all sampling points of the current symbol to obtain a digital baseband signal according to the current estimated frequency result;

[0114] The matched filter module 804 is used to obtain a digital baseband signal with 4 times oversampling after multiple downsampling stages, and to perform matched filtering on the digital baseband signal with 4 times oversampling. The output result at a preset position in the matched filter output is determined as the optimal sampling point, which is used as the output of the tracking loop.

[0115] The timing error estimation module 806 is used to estimate the timing error based on the output of the matched filter, and to adjust the sampling deviation of the next segment of data using the timing error estimation result.

[0116] The phase detector module 808 is used to obtain the phase value based on the optimal sampling point and to adjust the digital frequency conversion of the next data segment using the phase value.

[0117] Optionally, the high dynamic stability tracking system for non-spread spectrum continuous wave signals described above, after obtaining the digital signal based on ADC sampling, further includes: buffering the digital signal obtained by ADC sampling in RAM, wherein the address of the data stored in RAM is incremented by one for each input sampling point; and reading the corresponding sampling point from RAM.

[0118] Optionally, the high dynamic stability tracking system for the above-mentioned non-spread spectrum continuous wave signal reads the corresponding sampling point from RAM, including: determining the starting address of the first sampling point of the first symbol, and determining a preset number of consecutive sampling points as a symbol based on the starting address; updating the address information of the current sampling point according to the address of the sampling point of the previous symbol and the timing error estimation result, and reading the corresponding sampling point from RAM.

[0119] Optionally, the matched filtering module 804 obtains a digital baseband signal with 4 times oversampling by performing multiple downsampling stages, including: reducing the sampling rate of the digital baseband signal by a preset factor step by step, and reducing the sampling rate to 4 times the symbol rate by multiple preset factors to obtain a digital baseband signal with 4 times oversampling.

[0120] Optionally, the matched filtering module 804 reduces the sampling rate of the digital baseband signal by a preset factor step by step, including: obtaining the first filtering result after inputting a preset number of sampling points for the current symbol based on the number of sampling points of the current symbol; performing convolution operation when outputting the filtering result at the current sampling point; and outputting a filtering result for each preset number of input sampling points starting from the second segment of data of the current symbol.

[0121] Optionally, the timing error estimation module above performs timing error estimation based on the output of the matched filter, and uses the timing error estimation result to adjust the sampling deviation of the next segment of data, including: performing timing error estimation based on the output of the matched filter, converting the timing error estimation result into the number of sampling points to be adjusted, and adjusting the sampling deviation of the next segment of data.

[0122] Optionally, the timing error estimation module 806 performs timing error estimation based on the output of the matched filter, converts the timing error estimation result into the number of sampling points to be adjusted, and adjusts the sampling deviation of the next data segment, including: performing timing error estimation using the output of the matched filter to obtain the normalized symbol timing error estimation result; calculating the sampling deviation of the next data segment based on the timing error estimation result, and determining the corresponding address adjustment amount.

[0123] Optionally, the phase detector module 808 obtains the phase detection value based on the optimal sampling point and uses the phase detection value to adjust the digital frequency conversion of the next segment of data, including: inputting the optimal sampling point into the phase detector to obtain the phase detection value; inputting the phase detection value into the loop filter, and using the frequency offset estimate of the output of the loop filter to control the digital frequency conversion of the next segment of data.

[0124] Optionally, the phase detector module 808 inputs the optimal sampling point into the phase detector to obtain the phase detection value, including: inputting the optimal sampling point into the arctan arctangent phase detector, compensating the phase detection result, and performing compensation again after the loop is locked to obtain the phase detection value that eliminates sudden phase detection errors.

[0125] In this embodiment, a digital signal is obtained based on ADC sampling, and a digital baseband signal is obtained by digitally converting all sampling points of the current symbol according to the current estimated frequency result. After multiple levels of downsampling, a digital baseband signal with 4x oversampling is obtained, and matched filtering is performed on the digital baseband signal with 4x oversampling. The output result at a preset position in the matched filter output is determined as the optimal sampling point, which is used as the output of the tracking loop. Timing error is estimated based on the output result of the matched filter, and the sampling deviation of the next data segment is adjusted using the timing error estimation result. At the same time, the phase detection value is obtained based on the optimal sampling point, and the phase detection value is used... The digital frequency conversion of the next data segment is adjusted to gradually improve the signal-to-noise ratio of the received signal through multi-stage filtering and extraction, thereby improving the accuracy of timing deviation estimation. The sampling deviation is adjusted based on the timing deviation estimation results. Under high oversampling multiples, the number of sampling points of the symbol is adjusted to reduce the phase error caused by adjusting the number of sampling points. At the same time, the combination of sampling code offset tracking loop and frequency offset tracking loop enables stable tracking and compensation of code offset in high dynamic scenarios of non-spread spectrum continuous wave signals. This solves the technical problem that non-spread spectrum continuous wave signals cannot be continuously and stably tracked in high dynamic scenarios, and achieves the technical effect of continuous and stable tracking of non-spread spectrum continuous wave signals in high dynamic scenarios.

[0126] Example 3

[0127] In another aspect, the present invention provides an electronic device for implementing the above-described high dynamic stability tracking method for non-spread spectrum continuous wave signals. This electronic device is not limited to a terminal device or server in a system. The electronic device includes, but is not limited to, a memory and a processor. The memory stores a computer program, and the processor is configured to execute the steps of any of the above method embodiments via the computer program.

[0128] Example 4

[0129] In another aspect, the present invention provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in various alternative embodiments of the high dynamic stability tracking method for non-spread spectrum continuous wave signals described above. The computer program is configured to execute the steps in any of the above method embodiments during runtime.

Claims

1. A high dynamic stability tracking method for non-spread spectrum continuous wave signals, characterized in that, include: The digital signal is obtained by sampling with an ADC, and the digital baseband signal is obtained by digital frequency conversion of all sampling points of the current symbol based on the current estimated frequency result. After multiple downsampling stages, a digital baseband signal with 4x oversampling is obtained. The digital baseband signal with 4x oversampling is then subjected to matched filtering. The output result at a preset position in the matched filter output is determined as the optimal sampling point and used as the output of the tracking loop. The timing error is estimated based on the output of the matched filter, and the sampling deviation of the next data segment is adjusted using the timing error estimation result. At the same time, the phase detection value is obtained based on the optimal sampling point, and the digital frequency conversion of the next data segment is adjusted using the phase detection value. Timing error is estimated based on the output of the matched filter, and the sampling deviation of the next data segment is adjusted using the timing error estimation result. This includes: estimating the timing error using the output of the matched filter, obtaining the normalized sign timing error estimation result, and using x... i The normalized symbol timing error estimation formula is given by the formula for the output of the matched filter: Calculate the sampling deviation of the next data segment based on the timing error estimation result, and determine the corresponding address adjustment amount. The specific calculation formula is as follows: The value of n is in the range of (b+1)×1024≤n≤(b+2)×1024-1, and OSR represents the oversampling factor of the received signal.

2. The high dynamic stability tracking method for non-spread spectrum continuous wave signals as described in claim 1, characterized in that, After obtaining the digital signal based on ADC sampling, the process also includes: The digital signal sampled by the ADC is buffered in RAM. For each input sampling point, the address of the data stored in RAM is incremented by one. Read the corresponding sampling points from RAM.

3. The high dynamic stability tracking method for non-spread spectrum continuous wave signals as described in claim 2, characterized in that, Read the corresponding sampling points from RAM, including: Determine the starting address of the first sampling point of the first symbol, and determine a preset number of consecutive sampling points as one symbol based on the starting address; Based on the address of the previous symbol sampling point and the timing error estimation result, update the address information of the current sampling point and read the corresponding sampling point from RAM.

4. The high dynamic stability tracking method for non-spread spectrum continuous wave signals as described in claim 1, characterized in that, After multiple levels of downsampling, a digital baseband signal with a 4x oversampling factor is obtained, including: The sampling rate of the digital baseband signal is gradually reduced by a preset factor. After multiple levels of downsampling by preset factors, the sampling rate is reduced to 4 times the symbol rate, resulting in a digital baseband signal under 4 times oversampling.

5. The high dynamic stability tracking method for non-spread spectrum continuous wave signals as described in claim 4, characterized in that, The sampling rate of the digital baseband signal is gradually reduced by a preset factor, including: Based on the number of sampling points of the current symbol, obtain the first filtering result after the current symbol inputs a preset number of sampling points; Perform convolution operation when outputting the filtered result at the current sampling point; Starting from the second segment of data in the current symbol, a filtering result is output for every preset number of input sampling points.

6. The high dynamic stability tracking method for non-spread spectrum continuous wave signals as described in claim 1, characterized in that, The phase detection value is obtained based on the optimal sampling point, and the digital frequency conversion of the next data segment is adjusted using the phase detection value, including: Input the optimal sampling point into the phase detector to obtain the phase detection value; The phase detection value is input into the loop filter, and the frequency offset estimate of the loop filter output is used to control the digital frequency conversion of the next data segment.

7. The high dynamic stability tracking method for non-spread spectrum continuous wave signals as described in claim 6, characterized in that, The optimal sampling point is input into the phase detector to obtain the phase detection value, including: The optimal sampling point is input into the arctan arctangent phase detector to compensate for the phase detection result. After the loop is locked, compensation is performed again to obtain the phase detection value that eliminates sudden phase detection errors.

8. A high dynamic stability tracking system for non-spread spectrum continuous wave signals, used to implement the high dynamic stability tracking method for non-spread spectrum continuous wave signals as described in claim 1, characterized in that, include: The digital frequency conversion module is used to obtain a digital signal based on ADC sampling, and to perform digital frequency conversion on all sampling points of the current symbol to obtain a digital baseband signal according to the current estimated frequency result; The matched filtering module is used to obtain a digital baseband signal with 4 times oversampling after multiple downsampling stages, and to perform matched filtering on the digital baseband signal with 4 times oversampling. The output result at a preset position in the matched filtering output is determined as the optimal sampling point, which is used as the output of the tracking loop. The timing error estimation module is used to estimate the timing error based on the output of the matched filter, and to adjust the sampling deviation of the next segment of data using the timing error estimation result. The phase detector module is used to obtain the phase value based on the optimal sampling point and to adjust the digital frequency conversion of the next data segment using the phase value.

Images