CN103163535A - Direct Acquisition Method of P-code Based on DSP - Google Patents

Abstract

The invention discloses a P code direct capturing method based on a digital signal processor (DSP) and solves the problem that power consumption is overhigh when a P code is parallelly captured by using a field programmable gate array (FPGA). According to the method, the P code is directly captured through coordination of two DSPs so as to ensure that navigation data are positioned and resolved, one of the two DSPs is used for processing analog/digital (A/D) input signals, and the other DSP is used for processing locally generated P code. Compared with the conventional method in which only the FPGA is used, the method has the advantages that power consumption of a system is reduced, and convenience is brought to design of the portable miniature system. Simultaneously, 3dB of loss generated by correlated peaks is eliminated through composition and average operations, the composition and average operations aim to reduce sampling points and operation quantity, but a small amount of noise can be correspondingly introduced. By the method, the noise introduction quantity is reduced by reducing the composition frequency and the number of average points, so that the capture probability is increased.

Description

Direct Acquisition Method of P-code Based on DSP

technical field

The invention belongs to the technical field of satellite navigation, in particular to a fast and direct capture method of P code based on a digital signal processor (DSP, Digital Signal Processor) in a satellite navigation system.

Background technique

The Beidou navigation system, like the Global Positioning System (GPS, Global Positioning System), can provide users with standard positioning services and precise positioning services. The two services are based on C/A code and P code respectively. In the Beidou navigation system, the rate of the two is the same (10.23M Chip/s), but the precise code period is very long, and it has strong anti-jamming and anti-spoofing capabilities. Usually the P code capture is to obtain the navigation message through the locking of the C/A code, and then use the information provided by the transfer word in the navigation message to capture the P code. However, because the code length of the C/A code is very short, it is very susceptible to interference and spoofing, and it is necessary to start capturing the P code after obtaining the transfer word of a certain frame, and its capture speed is also slow (the transfer word Appearance interval is 0.6S). The P code period is as long as one week, and it has much stronger anti-interference performance than the C/A code. In complex geographical or electromagnetic environments, direct capture of the P code for the Beidou navigation system can further improve the positioning and timing accuracy and use navigation. The ability of the system to carry out electronic warfare and navigation warfare has high practical value.

In recent years, the research on P-code direct acquisition technology can be roughly divided into two categories, one is the time-domain processing method based on large-scale parallel correlators, represented by STS Y-EXPRESS ASIC, and the other is based on FFT Various frequency domain processing methods of processing technology, represented by XFAST.

The core idea of P code acquisition based on parallel correlator is to use parallel correlator to search multiple time-frequency units at the same time. Before searching, first determine the search time and frequency according to the GPS system time, position information and satellite position of the current receiver Range, and then determine the integration period according to the interference situation determined by AGC and the determined acquisition probability, false alarm probability, etc., which also determines the time-frequency range and search times that each parallel search can cover. Afterwards, the satellite signal processed by the front end of the receiver is correlated with the local code in the correlator. The phase offset of the spreading code and the Doppler frequency offset corresponding to the peak value greater than the threshold are given to the tracking module. Taking the STS Y-EXPRESS receiver as an example, it can search 32704 time-frequency units (511 time units and 64 frequency units) at the same time, the search accuracy is 1/2 chip, and the coverage time range can reach ±12.5μs. Frequency coverage up to ±20kHz. In order to improve the anti-interference ability, Y-EXPRESS ASIC can strip the GPS message to make the coherent integration time exceed 20ms. And the coherent integration period can be extended to 200ms by using FFT. In the literature "Wolfert R, Chen S, Kohli S, Direct P(Y)-code acquisition under a jamming environment, Proceedings of IEEE PLANS, April1998, 228-235", the Y-EXPRESS performance analysis and simulation test show that using a single Y -When the EXPRESS chip performs P code direct capture, when the J/S exceeds 50dB, the uncertainty is ±1ms, and the frequency search range is ±315kHz, the capture time is 600s, and most military receivers require that the P code capture time be within When the degree of certainty is less than or equal to 60s when the degree of certainty is ±1s, the capture time is far from meeting the requirements. Only by increasing the number of parallel correlators can the degree of parallelism be increased to improve capture efficiency. But this will increase power consumption and hardware resource overhead exponentially.

A P-code direct capture method proposed by Lin M David et al. This method first divides the satellite signal and the local code signal into M segments according to the frequency search accuracy and frequency uncertainty, and each segment has N points of long data, and then the satellite signal and the corresponding segment of the local code are correlated to obtain M points of data, and then Perform FFT operation on the data, and then compare it with the threshold. If it exceeds the threshold, the capture is successful, otherwise move a sampling point and repeat the above process. This method is a serial search method in terms of code phase search, and the frequency parallel search is realized by performing FFT on the correlated result, and the search speed is relatively slow.

The Extended Copy Overlapping Acquisition Search Technology (XFAST) not only takes advantage of frequency domain correlation, but also takes advantage of the low cross-correlation characteristics of P codes. In order to improve the search rate of the uncertain interval, the XFAST method divides the local uncertain interval into M sub-intervals, and each segment is N in length, and then each segment is correspondingly superimposed to construct a parent sequence, which is finally correlated with the input sequence, thereby speeding up the capture rate M times, but after a large number of stacking, the signal-to-noise ratio is extremely reduced, which affects the capture performance.

The direct averaging method is to average every N points of the satellite signal and the local code to obtain a new sampling point. Then perform FFT operations on them separately. The difference between this method and XFAST is that every N points are averaged to obtain a new sampling point instead of segmented superposition. The code phase accuracy is N, and then from the N code phases to search. This method can directly reduce the calculation amount by N times, but the disadvantage is that when the code phase offset is N/2, the correlation peak will produce 3dB loss.

Because the above P code serial capture method has serious defects in capture speed or performance, most receivers use the FPGA-based P code parallel capture method, but due to the high power consumption of FPGA, for similar individual handheld receivers For small receivers, high power consumption devices will reduce the receiver's standby capability.

Contents of the invention

The purpose of the present invention is to solve the problem of high power consumption when using FPGA to realize the parallel capture method of P code, and proposes a direct capture method of P code based on DSP.

In order to achieve the above object, the technical solution of the present invention is: a kind of P code based on DSP fast direct capture method, specifically comprises the steps:

S1. The received satellite signal after A/D conversion is input to the first block of DSP, and then I/Q separation, down-conversion and down-sampling are performed, and the sequence a obtained by sampling is stored in the internal RAM of the first block of DSP middle;

S2. In the second block of DSP, according to the time information provided by the local clock, determine the number of satellites and satellite numbers that can be captured at the current time, and the uncertainty range of the P code sequence with the smallest satellite number in the corresponding satellites that can be captured, and cost the local P code b, store it in external RAM;

S3. The sequence a obtained by sampling is taken out from the internal RAM of the first DSP, and the local P code b is taken out from the external RAM of the second DSP, and superposition and average operations are performed on them respectively to obtain A, B, superposition and average The specific process of operation is as follows:

S31. Segment the sampled sequence a and the local P code b by N points and one segment, wherein, sequence a is divided into M segments, sequence b is divided into K segments, and sequence a is offset by N points to obtain a', and sequence b is shifted by N/2 and N points to obtain b' and b'' respectively, superimposed a and a' to obtain sequence c, superimposed b, b', b', b'' and then averaged to obtain sequence d;

S32. Average each section of the superimposed two-way signal sequence c and sequence d to form a new sampling point, and obtain the A/D input signal A of M points and the local code L of K points after averaging;

S33. The local code L with a length of K is divided into J segments according to the M points of each segment, and then all segments are superimposed to form the local code B;

S4. In the first block of DSP, zero-fill the A obtained in step S32, then take the conjugate, and perform an FFT operation, and in the second block of DSP, perform zero-fill on the B obtained in step S33, and then perform an FFT operation , input the operation result in the second block DSP to the first block DSP, then multiply it with the operation result in the first block DSP and do IFFT;

S5. Compare the IFFT result with the threshold value, if it exceeds the threshold value, then search more precisely for the N/2 code phases exceeding the threshold value, after that, transfer the star to the tracking module and judge whether all satellites that can be captured are completed If there are still captureable satellites that have not been searched, start to capture the next satellite. If all the searches are completed, the capture is completed, and the second block of DSP is transferred to the navigation data positioning solution;

S6. If the threshold value is not exceeded, regenerate the local P code after shifting by one chip according to the uncertainty range, and repeat steps S3-S5.

Beneficial effects of the present invention: the present invention realizes P code quick direct capture by adopting two pieces of DSP, and one DSP is responsible for the processing of A/D input signal, and another piece of DSP is responsible for the processing of local generation P code, compared with FPGA, has reduced the system Power consumption is conducive to the realization of portable and miniaturized design of the system. In addition, the 3dB loss generated by the correlation peak can be eliminated through step S31 in the capture phase. The principle is to eliminate the The loss of the code offset to the correlation peak value can be seen through steps S32 and S33. Since the purpose of the average and superposition operations is to reduce the sampling point and reduce the amount of calculation, but due to the influence of the P code cross-correlation will introduce a small amount of noise, so This method reduces the amount of capture calculation at the cost of introducing noise, thereby reducing the capture time. This method also reduces the introduction of noise by reducing the number of stacking times and the number of average points, thereby increasing the capture probability.

Description of drawings

Fig. 1 is a schematic diagram of N/2 point superimposition before and after signal segmentation in the method of the present invention.

Fig. 2 is a schematic diagram of the local code J-segment superimposition in the method of the present invention.

Fig. 3 is a schematic diagram of the processing flow of capturing P codes in the present invention.

Fig. 4 is a schematic diagram of the internal structure and connection relationship of the positioning receiver system of the navigation system using the method of the present invention.

Detailed ways

The present invention will be further elaborated below in conjunction with the accompanying drawings and specific embodiments.

Because there are two signals that need to be processed for the capture of P code, A/D input signal and locally generated P code. Because DSP is a serial processing device, if only a single-core DSP is used, one of the signals must be processed earlier and then another, such a structure will affect the real-time performance of signal processing, and the method of the present invention is designed based on the above considerations. Two pieces of DSP are used to jointly capture the P code. The two DSPs here can be two single-core high-performance DSPs, or one multi-core DSP.

This embodiment adopts a multi-core DSPTI C6000 series high-performance signal processor, whose single-core main frequency can reach more than 1GHz, which provides a strong guarantee for processing the huge amount of calculations of P-code direct capture. In this embodiment, the DSP is mainly responsible for the following functions: I/Q separation, down-conversion, down-sampling, data storage, local code generation, overlapping and average FFT processing, frequency offset search, and threshold detection.

The processing flow of capturing the P code is shown in Figure 3, specifically as follows:

S1. The satellite signal received after the A/D conversion is input in the DSP core 1, then I/Q separation is performed, down-converted and down-sampled, and the sequence a obtained by sampling is stored in the internal RAM of the core 1;

S2. In the DSP core 2, according to the time information provided by the local clock, determine the number and satellite number of the satellites that can be captured at the current time and the minimum P code sequence uncertainty range of the corresponding satellite number in the satellites that can be captured, and then generate the local P code b, and store it in external RAM.

Specifically, the local P code b can be generated by a precision ranging code chip (PRM).

S3. the sequence a obtained by sampling is taken out from the internal RAM of the DSP core 1, and the local P code b is taken out from the external RAM, and superposition and average operations are carried out to it respectively to obtain A, B, and the specific process of superposition and average operations is as follows:

S31. As shown in Figure 1, the sequence a obtained by sampling and the local P code b are segmented by N points and one segment, wherein sequence a is divided into M segments, sequence b is divided into K segments, and sequence a is offset by N points Obtain a', offset the sequence b by N/2, and N points to obtain b' and b'' respectively, superimpose a and a' to obtain sequence c, superimpose b, b', b', b'' and then average to obtain Sequence d; those skilled in the art should realize that N here can be selected according to actual conditions.

S32. Average each section of the superimposed two-way signal sequence c and sequence d to form a new sampling point, and obtain the A/D input signal A of M points and the local code L of K points after averaging;

S33. As shown in Figure 2, the local code L with a length of K is divided into J sections according to every M point, and then all sections are superimposed to form the local code B;

S4. In DSP core 1, the A obtained in step S32 is zero-filled, then conjugated, and FFT operation is performed. In DSP core 2, the B obtained in step S33 is zero-filled, and then FFT operation is performed, and the DSP The operation result in core 2 is input to DSP core 1, and then it is multiplied by the operation result in DSP core 1 to perform IFFT;

S5. Comparing the IFFT result with the threshold value. If the threshold value is exceeded, the phases of the N/2 codes exceeding the threshold are further precisely searched. After that, transfer the star to the tracking module and judge whether all the satellites that can be captured have been searched. If there are still satellites that can be captured, start to capture the next satellite. If all the searches are completed, the capture is completed, and the DSP transfers the navigation data positioning solution.

S6. If the threshold value is not exceeded, regenerate the local P code after shifting by one chip according to the uncertainty range, and repeat steps S3-S5.

Specifically, the local P code can be regenerated by controlling the PRM to offset by one chip according to the uncertainty range.

As can be seen from the method introduction of the previous background technology, the method of the present invention eliminates the loss of 3dB that the correlation peak produces by step S31 in the capture phase, its principle can be seen by formula (1), same as to input signal a and local code b Doing the offset superposition operation can completely eliminate the loss of the code offset to the correlation peak value. It can be seen from steps S32 and S33 that since the purpose of the average and superposition operations is to reduce the sampling point to reduce the amount of calculation, but due to the cross-correlation of the P code A small amount of noise will be introduced due to the influence of nature, and if the threshold value is not exceeded after three repeated captures, the star will be abandoned and other visible stars will be captured instead.

Cor _peak =max{∑[(x _m +x _m+N )×(y ₀ +y _N/2 +y _N/2 +y _N )/4]}

=max{∑[x _m y ₀ +2(x _m y _N/2 +x _m+N y _N/2 )+x _m y _N +x _m+N y ₀ +x _m+N y _N ]/4 }

=max{∑[(x _m y ₀ +x _m+N y ₀ )+2(x _m y _N/2 +x _m+N y _N/2 )+(x _m+N y ₀ +x _m+N y _N )]/4} formula (1)

=max{∑[x ₀ y ₀ +2x ₀ y ₀ +x ₀ y ₀ ]/4}

=max{∑x ₀ y ₀ }

Utilize the navigation system positioning receiver system internal structure of the inventive method and connection relationship as shown in Figure 4, at first the antenna receiving signal enters the preamplifier to amplify the signal, then down-converts, analog-to-digital conversion, and then enters the DSP to carry out P code capture, capture After completion, the captured satellites are sent to the tracking module for carrier synchronization and P code phase synchronization. Since the positioning calculation requires at least 4 satellites, the tracking module must track at least 4 satellites at the same time, so use FPGA to take advantage of its parallel processing to build this tracking module, and perform positioning calculation when the number of satellites is more than 4. The DSP core 2 is responsible for this module, and recaptures the lost satellite at a certain time interval. After the positioning calculation is completed, the result is input to the user interface for the user to use.

In summary, the direct acquisition method of P code in the present invention adopts multi-core DSP to replace commonly used FPGA to reduce receiver power consumption and hardware resources, and improves the overlapping average zero-padding algorithm to adapt to this platform, so it will not increase the capture rate substantially. time. This method is especially suitable for handheld navigation and positioning receivers.

Those skilled in the art will appreciate that the embodiments described herein are to help readers understand the principles of the present invention, and it should be understood that the protection scope of the invention is not limited to such specific statements and embodiments. All possible equivalent replacements or changes made according to the above descriptions are deemed to belong to the protection scope of the claims of the present invention.

Claims (1)

1. a kind of direct capture method of P code based on DSP, specifically comprises the following steps: S1. The received satellite signal after A/D conversion is input to the first block of DSP, and then I/Q separation, down-conversion and down-sampling are performed, and the sequence a obtained by sampling is stored in the internal RAM of the first block of DSP middle; S2. In the second block of DSP, according to the time information provided by the local clock, determine the number of satellites and satellite numbers that can be captured at the current time, and the uncertainty range of the P code sequence with the smallest satellite number in the corresponding satellites that can be captured, and cost the local P code b, store it in external RAM; S3. The sequence a obtained by sampling is taken out from the internal RAM of the first DSP, and the local P code b is taken out from the external RAM of the second DSP, and superposition and average operations are performed on them respectively to obtain A, B, superposition and average The specific process of operation is as follows: S31. Segment the sampled sequence a and the local P code b by N points and one segment, wherein, sequence a is divided into M segments, sequence b is divided into K segments, and sequence a is offset by N points to obtain a', and sequence b is shifted by N/2 and N points to obtain b' and b'' respectively, superimposed a and a' to obtain sequence c, superimposed b, b', b', b'' and then averaged to obtain sequence d; S32. Average each section of the superimposed two-way signal sequence c and sequence d to form a new sampling point, and obtain the A/D input signal A of M points and the local code L of K points after averaging; S33. The local code L with a length of K is divided into J segments according to the M points of each segment, and then all segments are superimposed to form the local code B; S4. In the first block of DSP, zero-fill the A obtained in step S32, then take the conjugate, and perform an FFT operation, and in the second block of DSP, perform zero-fill on the B obtained in step S33, and then perform an FFT operation , input the operation result in the second block DSP to the first block DSP, then multiply it with the operation result in the first block DSP and do IFFT; S5. Compare the IFFT result with the threshold value, if it exceeds the threshold value, then search more precisely for the N/2 code phases exceeding the threshold value, after that, transfer the star to the tracking module and judge whether all satellites that can be captured are completed If there are still captureable satellites that have not been searched, start to capture the next satellite. If all the searches are completed, the capture is completed, and the second block of DSP is transferred to the navigation data positioning solution; S6. If the threshold value is not exceeded, regenerate the local P code after shifting by one chip according to the uncertainty range, and repeat steps S3-S5.

Images