A low computational complexity fully parallel fast acquisition method for DS-SS signals

Abstract

This invention discloses a low-computational-complexity, fully parallel, fast acquisition method for direct-sequence spread spectrum (DSSS) signals. The method involves first performing parallel FFT acquisition operations on multiple consecutive segments of equal-length received signals with a local PN code, then accumulating the correlation results of each signal segment. The peak value of the accumulated result is used for acquisition determination to obtain the code phase value. Based on the code phase value, the received signal is despread and FFT is performed to obtain the Doppler frequency offset value. This invention provides a fully parallel acquisition method for DSS signals by accumulating correlation results multiple times, which not only reduces computational complexity but also decreases resource consumption. Simulation analysis demonstrates that the multiple accumulation method of this invention achieves a higher acquisition probability than the PMF-FFT method under the same acquisition bandwidth and PN code length.

Description

Technical Field

[0001] This invention belongs to the field of measurement and control technology, specifically relating to a method for fast, fully parallel acquisition of direct-sequence spread spectrum signals with low computational complexity. Background Technology

[0002] Direct-sequence spread spectrum (DSSS) communication is widely used in military and civilian communications due to its excellent security and anti-jamming capabilities. For example, the currently used GPS (Global Positioning System), W-CDMA (Wideband Code Division Multiple Access) mobile communication system, and wireless local area networks all utilize DSSS technology. DSSS technology spreads the spectrum by multiplying the baseband data with a pseudo-noise (PN) code at the transmitting end, hiding the power spectrum of the transmitted signal within the noise power spectrum within the signal bandwidth, thus achieving both security and anti-jamming. Before data demodulation at the receiving end, the PN code modulated on the baseband data and the carrier Doppler frequency offset generated by the channel must be stripped before the baseband data can be demodulated using conventional modulation and demodulation methods. Therefore, the key technology in a DSSS communication system lies in the two-dimensional acquisition of the PN code phase and carrier Doppler frequency offset of the DSSS signal at the receiving end.

[0003] DSSS signal acquisition methods can be categorized into four types based on the serial / parallel operation of each dimension: serial code phase and serial carrier phase, serial code phase and parallel carrier phase, parallel code phase and serial carrier phase, and parallel code phase and parallel carrier phase. Each type of acquisition method has its own application scenarios depending on the DSSS communication system parameters and design requirements. For example, when a fast acquisition time is not required, a fully serial approach can be used, consuming the least resources. When the search interval in one dimension is small, a parallel approach can be considered for that dimension, while a serial approach can be used for the other dimension. When fast acquisition is required, and the search intervals in both dimensions are large, a fully parallel acquisition method must be considered; however, a fully parallel acquisition method inevitably involves high computational load and hardware resource consumption.

[0004] Modern DSSS communication systems are mostly high-dynamic, short-duration burst signals. These signals are not only short-duration bursts but also carry Doppler frequency offsets generated by high dynamics. Furthermore, each frame of the signal must undergo a complete acquisition and tracking process. If the code phase and frequency offset of the signal cannot be correctly acquired within the pilot signal's duration, frame loss will occur. In this scenario, rapid acquisition is crucial. Due to the significant Doppler frequency offsets generated by high dynamics, parallel acquisition of the frequency offset is often achieved using FFT. For this scenario, a commonly used and easily implemented acquisition method is the Partial Matched Filter (PMF) plus FFT analysis method, i.e., the PMF-FFT acquisition method. However, this method requires correlation operations on the received signal at each code phase, followed by segmented summation and then FFT calculation, resulting in significant computational and hardware resource consumption. Additionally, the frequency offset acquisition bandwidth of this method is limited by the number of FFT calculation points; when increasing the acquisition bandwidth is required, the increase in computational load and hardware resource consumption becomes even more pronounced.

[0005] In summary, existing technologies for fast acquisition of direct-sequence spread spectrum (DSSS) signals often come at the cost of high computational complexity and resource consumption. This makes such fast acquisition methods unsuitable for engineering implementation in some resource-constrained and power-constrained scenarios. Therefore, there is a need for a fully parallel fast acquisition method for DSS signals that reduces the resource consumption of algorithm implementation without increasing computational load or reducing acquisition performance. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a low-computational-complexity fully parallel fast acquisition method for direct-sequence spread spectrum (DSSS) signals. First, multiple consecutive segments of equally long received signals are acquired in parallel using FFT with the local PN code. Then, the correlation results of each signal segment are accumulated, and the peak value of the accumulated result is used for acquisition determination to obtain the code phase value. Based on the code phase value, the received signal is despread and FFT is performed to obtain the Doppler frequency offset value.

[0007] To achieve the above-mentioned technical effects, the present invention is implemented through the following technical solution: a low computational complexity fully parallel fast acquisition method for direct-sequence spread spectrum signals, characterized by comprising the following steps:

[0008] S1. After the direct-sequence intermediate frequency signal is sampled by AD, down-converted, filtered and down-sampled, it is divided into multiple signal segments. The length of each signal segment is the same as the code period. Then, FFT operation is performed on each signal segment.

[0009] S2. Multiply the FFT result of each signal segment in S1 with the conjugate result of the pre-calculated local PN code FFT operation accordingly.

[0010] S3. Perform IFFT operation on the calculation results in S2 and sum the results of each segment;

[0011] The length of the accumulated results in S4 and S3 is the same as the PN code period. The captured PN code phase value can be obtained based on the peak position of the results.

[0012] S5. Using the PN code phase value obtained in S4, generate a PN code sequence to despread the signal in S1 and perform FFT operation on the result. The frequency value corresponding to the peak position of the FFT operation result is the captured Doppler frequency offset value, thus completing the two-dimensional parallel acquisition of the PN code phase and Doppler frequency offset of the direct-sequence spread signal.

[0013] The beneficial effects of this invention are:

[0014] This invention provides a fully parallel acquisition method for direct-sequence spread spectrum (DSSS) signals. First, the data is divided into multiple continuous segments with the same code period. Second, the code phase of each segment is acquired in parallel using FFT. Third, the acquisition results of each segment are accumulated multiple times, and the code phase value is obtained from the peak position of the accumulated results. Finally, the data is despread using the obtained code phase value, and the Doppler frequency offset is obtained by FFT operation on the despread result, thus completing the acquisition of the DSS signal. This method not only reduces computational complexity but also decreases resource consumption. Simulation analysis demonstrates that the multiple accumulation method of this invention achieves a higher acquisition probability than the PMF-FFT method under the same acquisition bandwidth and PN code length. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the principle of multi-accumulation fully parallel capture in Embodiment 3 of the present invention;

[0017] Figure 2 This is a graph showing the acquisition probability curves of different methods when the Doppler frequency offset is 2kHz, as shown in Embodiment 4 of the present invention. Detailed Implementation

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Example 1

[0020] A low-computational-complexity, fully parallel, fast acquisition method for direct-sequence spread spectrum signals, characterized by the following steps:

[0021] S1. After the direct-sequence intermediate frequency signal is sampled by AD, down-converted, filtered and down-sampled, it is divided into multiple signal segments. The length of each signal segment is the same as the code period. Then, FFT operation is performed on each signal segment.

[0022] S2. Multiply the FFT result of each signal segment in S1 with the conjugate result of the pre-calculated local PN code FFT operation accordingly.

[0023] S3. Perform IFFT operation on the calculation results in S2 and sum the results of each segment;

[0024] The length of the accumulated results in S4 and S3 is the same as the PN code period. The captured PN code phase value can be obtained based on the peak position of the results.

[0025] S5. Using the PN code phase value obtained in S4, generate a PN code sequence to perform despreading operation on the signal in S1 and perform FFT operation on the result. The frequency value corresponding to the peak position of the FFT operation result is the captured Doppler frequency offset value, thus completing the two-dimensional parallel acquisition of the PN code phase and Doppler frequency offset of the direct-sequence spread signal.

[0026] Example 2

[0027] Based on the above embodiment 1, this embodiment is an analysis of the constraints of the multiple accumulation method of a fully parallel fast acquisition method for direct-sequence spread spectrum signals with low computational complexity;

[0028] The main feature of this method is that in the acquisition of the PN code phase, a short signal is used for correlation to reduce the impact of Doppler frequency offset on the correlation peak and increase the acquisition bandwidth. Then, the acquisition performance is improved by accumulating the correlation peak results of multiple signal segments. The impact of Doppler frequency offset on the PN code is manifested in its effect on the chip width T. c The compression or expansion, as long as the Doppler frequency shift relative to T within the time of the accumulated signal length. c If the compression or expansion is less than one chip time, then the maximum position of the correlation peak of each signal segment can be guaranteed to be the same, which means that accumulation operation can be performed to improve acquisition performance. Define the RF communication carrier as f. r The chip width after the Doppler frequency offset is T c 'Then the constraint condition for the cumulative signal length N is...'

[0029] |T c -T c '|·N<T c

[0030] According to T c 'and f r and f d The relationship can be further simplified to the above formula as follows:

[0031]

[0032] At a certain f r From the above formula, we can obtain the cumulative signal length N and f that satisfy the correlation peak accumulation operation. d The relationship. For example, when f r =5GHz, f d At 500kHz, the cumulative signal length N is approximately 10000, meaning that the accumulation operation is not only applicable to large Doppler frequency offsets, but also has very little constraint on the cumulative signal length.

[0033] Example 3

[0034] Based on the above embodiment 1, this embodiment is a hardware resource consumption analysis of a low computational complexity direct-sequence spread spectrum signal fully parallel fast acquisition method;

[0035] Assume the required accumulated signal length N = 4096 and code period L = 127 for the required frequency offset acquisition accuracy. Since the multiple accumulation acquisition method processes only signals of length L each time, it only requires one RAM storage resource of depth N, allowing simultaneous data storage and FFT operations. In contrast, the PMF-FFT method requires storing all data at once, then reading it out for correlation and FFT operations on the PN code. When processing the received data, it must use another RAM to store subsequent received data, requiring at least two RAM storage resources of depth N.

[0036] See Figure 1 As shown, after receiving all the data in one pass, the PMF-FFT method must perform L correlation operations of length N and L FFT operations of P points within the next data reception time of the same length to ensure that continuously received data is not lost. However, in engineering implementations, a sampling precision of 2 sampling points per chip is generally used, which doubles the number and length of the corresponding correlation and FFT operations. Therefore, when processing the received data, multiple correlators must be used for parallel correlation operations, and multiple FFT cores must be used for parallel operations. The specific degree of parallelism is related to the system clock and the PN code rate. The multiple accumulation method can store the pre-calculated FFT and conjugate operation results of the local PN code, that is, this part does not consume an FFT core; the FFT and IFFT operations of the data are processed sequentially in time, that is, this part only consumes one FFT core; in addition, the FFT operation after signal despreading requires an FFT core.

[0037] The above analysis shows that the PMF-FFT method consumes more resources when the PN code rate is high, while the multiple accumulation method not only consumes less resources than the PMF-FFT method, but also does not depend on the system clock and the PN code rate.

[0038] Example 4

[0039] Based on the above embodiment 1, this embodiment is a performance simulation and analysis of a fully parallel fast acquisition method for direct-sequence spread spectrum signals with low computational complexity;

[0040] This section primarily simulates the capture probabilities of two capture methods in the MATLAB environment. The simulation parameters are set as follows:

[0041] The system clock is 16MHz, the intermediate frequency carrier of the spread spectrum signal is 4MHz, the PN code rate is 1MHz, the code length is 127, the LPF bandwidth is 2MHz, the downsampling factor is 16 times (i.e., 1 chip = 1 sampling point), and the channel is an AWGN channel. The signal accumulation length N is 4096. In the PMF-FFT method, M is 128, and P is 32; in the multiple accumulation method, the accumulation number is 32.

[0042] Based on the above parameters, the Monte Carlo method was used to simulate the acquisition probabilities of different methods, as follows: Figure 2 As shown.

[0043] Based on the simulation parameter settings Figure 2 The PMF-FFT method in this paper involves 128-point partial correlation and 32-point FFT operations. The multiple accumulation method performs 128-point FFT and IFFT operations each time, then accumulates the results of the 32 operations to capture the PN code phase. After capturing the PN code phase, the signal is despread and then subjected to an FFT operation. The number of operation points and data length are the same as the PMF-FFT method to ensure the same capture bandwidth. The calculated capture bandwidth is approximately 3.9 kHz. The computational complexity of the two methods is analyzed in Table 1. The computational complexity mainly includes real-number multiplication and real-number addition in the FFT and correlation operations.

[0044] Table 1 Comparison of computational complexity of different capture methods

[0045]

[0046] Simulation results and computational comparisons show that, with a Doppler frequency offset of 2kHz, the acquisition probability of the multiple accumulation method is higher than that of the PMF-FFT method, with a maximum gain improvement of approximately 2dB. Simultaneously, the computational cost of the multiple accumulation method is about 1 / 10 that of the PMF-FFT method. The computational complexity of the PMF-FFT method mainly lies in the 127 correlation operations across a 4096-point signal length, while the computational cost of the multiple accumulation method is solely based on the FFT operation.

[0047] In summary, the direct-sequence spread spectrum (DSS) signal acquisition method provided by this invention, which performs multi-stage accumulation of correlation results in a fully parallel acquisition process, not only reduces computational complexity but also decreases resource consumption. Simulation analysis demonstrates that the multi-stage accumulation method of this invention achieves a higher acquisition probability than the PMF-FFT method under the same acquisition bandwidth and PN code length.

[0048] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0049] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A low-computational-complexity, fully parallel, fast acquisition method for direct-sequence spread spectrum signals, characterized in that: Specifically, the following steps are included: S1. After the direct-sequence intermediate frequency signal is sampled by AD, down-converted, filtered and down-sampled, it is divided into multiple signal segments. The length of each signal segment is the same as the code period. Then, FFT operation is performed on each signal segment. S2. Multiply the FFT result of each signal segment in S1 with the conjugate result of the pre-calculated local PN code FFT operation accordingly. S3. Perform IFFT on the calculation results in S2 and accumulate the results of each segment; the constraint condition for the accumulated signal length N during accumulation is: ； In the formula, For chip width, For radio frequency carrier frequency; The length of the accumulated results in S4 and S3 is the same as the PN code period. The captured PN code phase value can be obtained based on the peak position of the results. S5. Using the PN code phase value obtained in S4, generate a PN code sequence to perform despreading operation on the signal in S1 and perform FFT operation on the result. The frequency value corresponding to the peak position of the FFT operation result is the captured Doppler frequency offset value, thus completing the two-dimensional parallel acquisition of the PN code phase and Doppler frequency offset of the direct-sequence spread signal.

2. The application of the low computational complexity fully parallel fast acquisition method for direct-sequence spread spectrum signals as described in claim 1 in signal acquisition.

Images