CN102162852A - Method and system for capturing weak GNSS (Global Navigation Satellite System) signal under condition of large-scale frequency deviation - Google Patents

Abstract

The invention discloses a weak GNSS signal acquisition method and system thereof under large-scale frequency offset. The maximum carrier frequency search step is determined according to the pre-detection integration time, carrier frequency and sampling rate, and then the entire frequency range to be searched is determined by the maximum frequency. The search step is divided into multiple carrier frequency units to realize a rough search of the carrier frequency. The received satellite signal is mapped to each carrier frequency unit, and the local pseudo-code rate is adjusted according to the carrier frequency unit to reduce the influence of code Doppler. Adopt flexible zero padding method, calculate correlation through FFT/IFFT, and realize code phase parallel search. Next, FFT operation is performed on different correlation values corresponding to the same code phase, so as to achieve the purpose of finely searching the carrier frequency in the carrier frequency unit. The method and the system of the invention can improve the acquisition speed and the acquisition sensitivity of the receiver, and enable the receiver to complete the pseudo-code acquisition work of the weak navigation signal under large-scale frequency deviation.

Description

Method and system for weak GNSS signal acquisition under large-scale frequency offset

technical field

The invention relates to the technical field of Global Navigation Satellite System (GNSS), in particular to a weak GNSS signal acquisition method and system thereof under large-scale frequency offset.

Background technique

Global Navigation Satellite System (GNSS) mainly includes Global Positioning System (GPS) (United States), Global Navigation Satellite System (GLONASS) (Russia), Galileo Navigation Satellite System (GALILEO) (EU) and Beidou Global Navigation System (COMPASS) (China) ). These systems all broadcast satellite signals to users around the world by way of code division multiple access. In this way, the user must receive the satellite signal normally after completing the pseudo-code acquisition. In environments such as cities, mountains, and forests, satellite signals are often blocked by surrounding buildings, vegetation, etc., the signal will become very weak, and it will become very difficult to capture pseudo codes. Furthermore, when the user moves at a high speed, it will become more difficult to capture pseudocodes affected by the Doppler effect. Therefore, fast acquisition of weak navigation signals under large-scale frequency offset has become one of the research hotspots in the field of satellite navigation.

Several pseudo-code acquisition algorithms commonly used in the GNSS field at home and abroad mainly include: segmental zero-padding algorithm, extended copy overlap acquisition technology, direct averaging algorithm, etc.

Segmented zero-padding algorithm is a commonly used pseudo-code capture algorithm in the world. Utilizing the relationship between correlation and convolution and convolution and fast Fourier transform (FFT), the correlation operation can be realized by fast Fourier transform. If the zero padding operation can be performed on the received signal before the correlation is realized, then the FFT can be used to achieve the purpose of searching the code phase in parallel.

As shown in Figure 1, it is a schematic diagram of the subsection zero padding algorithm. Firstly, the analog-to-digital conversion is performed on the received signal, and then the digital signal after sampling at N points is selected, zero-filled to 2N points, and FFT and conjugate processing are performed. Select 2N point local signals and perform FFT processing. The result after FFT processing of the received signal is multiplied by the result of FFT processing of the local signal, and then inverse Fourier transform (IFFT) processing is performed, and the first N+1 points of the obtained results are retained. Non-coherent accumulation is performed according to the strength of the received signal. If there is a correlation value greater than or equal to the threshold in the obtained result, it means that the acquisition is successful; otherwise, search the next code phase and carrier frequency unit until a correlation value greater than or equal to the threshold is found.

The core idea of the extended copy overlap capture technology is to combine multiple local signals into one segment by means of segmentation and folding, so as to directly reduce the number of code phase units to be searched, and thus speed up the search speed. Shown in Fig. 2 is the schematic diagram of this method. Select the received signal after sampling at N points, and perform FFT and conjugate processing on the signal. The code phase range to be searched is determined according to the time uncertainty of the received signal, and a local pseudo code signal is generated. Select the local signal of MN points, divide it into M segments with equal length, and then add the corresponding segments to generate N-point folded local signals, and perform FFT processing. The result after the FFT processing of the received signal is multiplied by the result of the FFT of the local signal, and then IFFT processing is performed. Non-coherent accumulation is performed according to the signal-to-noise ratio of the received signal. Then, threshold detection is performed, and if a correlation value greater than or equal to the threshold is found, it indicates that the code phase of the received signal may fall within the local pseudo-code segment. Code phase detection is performed on the local signal in turn until the code phase that is aligned with the received signal is found—deambiguity. If no correlation value greater than or equal to the threshold is found during threshold detection, then move the phase of the local code and generate the next section of local pseudo code and then search until one larger than the threshold is found.

As shown in Figure 3 is the schematic diagram of the direct average algorithm. Select the received signal after sampling at MN points, and then average the signals of each adjacent M point to obtain the averaged received signal at N points, and fill in zeros to 2N points. Perform FFT and conjugate processing on this signal. According to the time uncertainty of the received signal, the code phase range to be searched is determined, and a local pseudo code signal is generated. Select local signals of 2MN points, and then average each adjacent M points to obtain averaged local signals of 2N points. Perform FFT processing on this signal. The result after the FFT of the received signal is multiplied by the result of the FFT of the local signal, and then IFFT processing is performed. Non-coherent accumulation is performed according to the signal-to-noise ratio of the received signal. Then, threshold detection is performed, and if a correlation value greater than or equal to the threshold is found, it indicates that the code phase of the received signal may fall within the local pseudo-code segment. In the case of not performing average processing on the received signal and the local signal, code phase detection is performed on the local signal in turn to find out the local code phase aligned with the received signal. This process is called deblurring. If no correlation value greater than or equal to the threshold is found during threshold detection, then move the phase of the local code and generate the next section of local pseudo code and then search until one larger than the threshold is found.

Compared with the piecewise zero-padding algorithm, the outstanding advantage of this method is that the number of code phase units to be searched is directly reduced by averaging the received signal and the local signal, thereby improving the search speed. However, because the averaging operation will deteriorate the pseudo-code correlation, and because there is a random code phase offset when the pseudo-code is selected during the averaging process, the acquisition performance of this method is poor. Compared with the segmented zero padding algorithm, the difference in detection performance can reach up to 6dB.

The pseudo-code capture algorithm described above considers normally received signals. When the signal is blocked or interfered, the signal will become very weak. At this time, the above algorithm will face severe challenges. When the signal is very weak, it is necessary to extend the pre-detection integration time, including extending the coherent integration time and increasing the number of non-coherent accumulations. After prolonging the pre-detection integration time, the code Doppler effect, which is not considered under normal receiving conditions, will become very significant. Code Doppler will cause the code phases of different signal segments to slide relative to each other, so that the correlation peaks of different signal segments cannot be added to the same code phase, which will seriously affect the acquisition performance. Therefore, the above algorithm is not applicable to weak navigation signals, and it is urgent to study new methods.

Contents of the invention

(1) Technical problems to be solved

The technical problem to be solved by the present invention is to improve the acquisition speed and acquisition sensitivity of the receiver, so that the receiver can complete the pseudo-code acquisition work of the weak navigation signal under large-scale frequency deviation.

(2) Technical solutions

In order to solve the above problems, the present invention provides a method for capturing weak GNSS signals under large-scale frequency offset, the method comprising steps:

S 1. Determine the maximum carrier frequency search step size f _step according to the coherent integration time T _coh set by the receiver, the number of non-coherent accumulation N _non , the carrier frequency f _carr , and the sampling frequency F _s , and set the carrier frequency to be searched The range is divided into several carrier frequency cells according to the maximum carrier frequency search step size f _step ;

S2. Divide the received signal into M segments, each segment contains N points, and map each segment signal to the selected carrier frequency cell;

S3. According to the selected carrier frequency cell, adjust the local pseudo-code rate, generate a local pseudo-code signal, sample the local pseudo-code signal with F _s , select the local pseudo-code signal after point P sampling, and It performs FFT operation;

S4. Select the first segment of signal mapped to the carrier frequency cell in step S2, fill it with zeros to point P, and perform FFT and conjugate operation on the zero-filled sequence;

S5. Multiply the result of step S3 and step S4, and perform IFFT operation to obtain a correlation value sequence with a length of P, store the first P-N point correlation values in the correlation value sequence as a row, and discard the remaining N point correlation values ;

S6. Process all received signals mapped to the carrier frequency cell in all M segments to obtain a correlation value matrix of M rows and P-N columns;

S7. Carry out zero padding, FFT operation and modulus to each column data in the correlation value matrix, obtain Correlation value matrix of row PN column;

行P-N列的非相干累加值矩阵；S8. Repeat steps S2-S7 for a total of N _non times, and non-coherently accumulate the results obtained each time to obtain a size of

A matrix of non-coherent accumulated values of rows and columns;

S9. If the maximum value in the non-coherent accumulated value matrix is greater than or equal to the preset threshold, then determine the pseudo code phase of the received signal according to the column where the maximum value is located, and determine the pseudocode phase of the received signal according to the row where the maximum value is located and the carrier mapped to the received signal Frequency cell, determine carrier frequency offset and code Doppler, otherwise execute step S10;

S10. If all the carrier frequency unit cells have been searched, then slide the P-N point local pseudo-code phase, and re-execute steps S2-S9, otherwise, map the received signal to the next carrier frequency unit cell, and continue to execute steps S3-S9.

T_short＝1/f_step，若T_short＞T_coh，则取T_short＝T_coh，

表示不小于T_coh/T_short的最小整数，N＝F_s×T_short。in,

T _short =1/f _step , if T _short >T _coh , then take T _short =T _coh ,

Represents the smallest integer not smaller than T _coh /T _short , N=F _s ×T _short .

Wherein, step S1 further includes:

S1.1 Determine the carrier frequency search step size f _step according to the receiver coherent integration time T _coh , non-coherent accumulation times N _non , carrier frequency f _carr , and sampling frequency F _s :

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; step</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; carr</span>}}}{{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; coh</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; non</span>}}}<span\; class="notranslate">\; ,</span>$

If f _step is greater than the carrier frequency range f _search to be searched, then set f _step = f _search ;

(i＝0，2，...，K-1)；S1.2 According to the f _step , divide the carrier frequency range to be searched into K=f _search /f _step carrier frequency cells, and record these carrier frequency cells as

(i=0,2,...,K-1);

S1.3 Determine the short integration time T _short =1/f _step according to f _step , if T _short >T _coh , then set T _short =T _coh ;

S1.4 Determine the number of segments of the received signal according to T _coh and T _short

S1.5 Reset the short integration time T _short =T _coh /M, according to T _short and F _s , determine N=F _s ×T _short and

上，得到新的信号

其中，Δt＝1/F_s表示采样时间间隔，

Wherein, in step S2, the _digital signal r ₀ (n) (n=0, 1, 2, ..., N-1, N, N+1, ..., 2N- 1, 2N, 2N+1,..., MN-1), the total length of r ₀ (n) is MN points, and the M segment signals are mapped to the first carrier frequency cell

on, get a new signal

Among them, Δt=1/F _s represents the sampling time interval,

Wherein, step S3 further includes:

S3.1 adjust the local pseudo-code rate according to the carrier frequency cell mapped in step S2 And generate a local pseudocode signal, f _chip is the pseudocode rate;

S3.2 Sampling the local pseudo-code signal with F _s , selecting the sampled local pseudo-code signal c(m) (m=0, 1, 2..., P-1) at point P, and then performing FFT operation, get:

$\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; c</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="j"\; filenames="HSA00000383226300021.png,HSA00000383226300032.png"\; state="\{\{state\}\}">j</figure-callout></span>}\frac{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}\mathrm{<span\; class="notranslate">\; mk</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0,1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; .</span>$

(m＝0，1，2，...，P-1)，对该序列进行FFT和共轭运算得到：Wherein, in step S4, the sequence after the zero padding is:

(m=0, 1, 2, ..., P-1), perform FFT and conjugate operation on this sequence to obtain:

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="j"\; filenames="HSA00000383226300021.png,HSA00000383226300032.png"\; state="\{\{state\}\}">j</figure-callout></span>}\frac{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}\mathrm{<span\; class="notranslate">\; mk</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; *</span>}$

() ^* indicates complex conjugate operation.

Wherein, in step S6, the processing of all M segments of received signals mapped to the carrier frequency unit cell is as follows: each time another segment of signal mapped to the carrier frequency unit cell in step S2 is selected, step S4 is performed Afterwards, a new P-point local pseudo-code signal is generated by sliding N-point local pseudo-code phases, and then returns to step S3, and goes to step S5.

Wherein, step S7 further includes:

个零，得到补零后的相关值矩阵；S7.1 Supplement each column of data in the correlation value matrix

zeros to get the correlation value matrix after padding with zeros;

行P-N列的相关值矩阵。S7.2 performs FFT operation and modulus on the correlation value matrix after the zero padding, and obtains

A matrix of correlation values for rows and columns.

The present invention also provides a weak GNSS signal acquisition system under large-scale frequency offset, the system comprising:

The carrier frequency unit is used to determine the maximum carrier frequency search step f _step according to the coherent integration time T _coh set by the receiver, the number of non-coherent accumulation N _non , the carrier frequency f _carr , and the sampling frequency F _s . The carrier frequency range is divided into several carrier frequency unit cells according to the maximum carrier frequency search step size f _step , and the received signal is divided into M segments, each segment contains N points, and each segment signal is mapped to the selected carrier on the frequency cell;

The local pseudo-code generation unit is used to adjust the local pseudo-code rate according to the selected carrier frequency unit cell, generate the local pseudo-code signal, sample the local pseudo-code signal with F _s , and select the local pseudo-code signal after point P sampling Pseudo-code signal, and perform FFT operation on it;

The first processing unit is used to select the first signal segment mapped to the carrier frequency cell, pad it to point P with zeros, and perform FFT and conjugate operations on the sequence after zero padding;

A correlation operation unit, for multiplying the result of the local pseudo code generation unit and the result of the first processing unit, and performing an IFFT operation to obtain a correlation value sequence with a length of P, and correlating the first P-N points in the correlation value sequence The value is stored as a row, and the remaining N point related values are discarded;

A correlation value matrix generation unit, used to control and process all M segments mapped to the received signal on the carrier frequency cell, and obtain a correlation value matrix of M rows and P-N columns;

The second processing unit is used to perform zero padding, FFT operation and modulus for each column of data in the correlation value matrix, to obtain Correlation value matrix of row PN column;

行P-N列的非相干累加值矩阵；The non-coherent accumulation unit is used to non-coherently accumulate the different correlation values corresponding to the same code bit to obtain a size of

A non-coherent accumulated value matrix of rows and columns;

Threshold detection unit, for judging whether the maximum value in the non-coherent accumulated value matrix is greater than or equal to the preset threshold, if greater than or equal to the preset threshold, then determine the phase of the pseudo code of the received signal according to the column where the maximum value is located, and according to the maximum Determine the carrier frequency offset and code Doppler in the line where the value is located and the carrier frequency cell mapped to the received signal; otherwise, judge whether all the carrier frequency cells have been searched, and if the search has been completed, slide the P-N point local pseudo-code phase, Re-execute the loop processing, otherwise, map the received signal to the next carrier frequency cell, and continue to execute the loop processing.

T_short＝1/f_step，若T_short＞T_coh，则取T_short＝T_coh，

表示不小于T_coh/T_short的最小整数，N＝F_s×T_short。in,

T _short ＝1/f _step , if T _short ＞T _coh, take T _short ＝T _coh ,

Represents the smallest integer not smaller than T _coh /T _short , N=F _s ×T _short .

(3) Beneficial effects

The method and system of the present invention have the following advantages:

(1) The carrier frequency range to be searched is classified into coarse search and fine search, which can not only realize the parallel search of carrier frequency to the greatest extent, but also reduce the impact of carrier frequency on the acquisition performance.

(2) Roughly divide the carrier frequency unit according to the coherent integration time and the non-coherent accumulation times, and then compensate the code Doppler accordingly, which can reduce the influence of the code phase slip on the acquisition performance and improve the acquisition sensitivity.

(3) A flexible zero padding method is adopted to make the length of FFT/IFFT be a power of 2 and improve the efficiency of FFT/IFFT operation.

Description of drawings

Fig. 1 is a block diagram of zero padding algorithm in the prior art;

Fig. 2 is a schematic diagram of the extended copy overlapping capture technology in the prior art;

Fig. 3 is a schematic diagram of the direct averaging algorithm in the prior art;

Fig. 4 is a flow chart of a weak GNSS signal acquisition method under large-scale frequency offset according to an embodiment of the present invention;

Fig. 5 (a)-5 (b) respectively is when receiving signal signal to noise ratio is-46dB, coherent integration time is 1ms, non-coherent accumulation number of times is 100 times, segmental zero padding algorithm and method capture performance schematic diagram of the present invention;

Fig. 6 (a)-6 (b) respectively is when receiving signal SNR is-46dB, coherent integration time is 1ms, non-coherent accumulation number of times is 300 times, segmental zero padding algorithm and method capture performance schematic diagram of the present invention;

Figures 7(a)-7(b) are schematic diagrams of the segmental zero-padding algorithm and the method capture performance of the present invention when the signal-to-noise ratio of the received signal is -46dB, the coherent integration time is 2ms, and the number of non-coherent accumulations is 300 times;

Fig. 8 is a schematic diagram of the acquisition performance of the method of the present invention when the signal-to-noise ratio of the received signal is -52dB, the coherent integration time is 2ms, and the number of non-coherent accumulations is 300 times;

Fig. 9 is a schematic diagram of a weak GNSS signal acquisition system under large-scale frequency offset according to an embodiment of the present invention.

Detailed ways

The method and system for acquiring weak GNSS signals under large-scale frequency offset proposed by the present invention are described as follows in conjunction with the accompanying drawings and embodiments.

The method of the present invention determines the maximum carrier frequency search step size according to the pre-detection integration time (the product of coherent integration time and non-coherent accumulation times), carrier frequency, and sampling rate, and then divides the entire frequency range to be searched by the maximum frequency search step size into Multiple carrier frequency units to realize rough search of carrier frequency. The received satellite signal is mapped to each carrier frequency unit, and the local pseudo-code rate is adjusted according to the carrier frequency unit to reduce the influence of code Doppler. Adopt flexible zero padding method, calculate correlation through FFT/IFFT, and realize code phase parallel search. Next, FFT operation is performed on different correlation values corresponding to the same code phase, so as to achieve the purpose of finely searching the carrier frequency in the carrier frequency unit.

As shown in FIG. 4, a weak GNSS signal acquisition method under a large-scale frequency offset according to an embodiment of the present invention includes steps:

S1. Determine the maximum carrier frequency search step size f _step according to the coherent integration time T _coh set by the receiver, the number of non-coherent accumulation N _non , the carrier frequency f _carr , and the sampling frequency F _s , and set the carrier frequency range to be searched Divide into several carrier frequency unit cells according to the maximum carrier frequency search step size f _step . This step further includes:

S1.1 According to receiver coherent integration time T _coh , non-coherent accumulation times N _non , carrier frequency f _carr , and sampling rate F _s , determine the step size f _step of carrier frequency rough search:

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; step</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; carr</span>}}}{{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; coh</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; non</span>}}}<span\; class="notranslate">\; ,</span>$

If f _step is greater than the carrier frequency range f _search to be searched, then set f _step = f _search ;

S1.2 Take f _step as the step size, divide the carrier frequency range to be searched into K=f _search /f _step carrier frequency cells, and record these carrier frequency cells as

S1.3 Determine the short integration time T _short =1/f _step according to the frequency search step size f _step . If T _short > T _coh , then set T _short = T _coh ;

其中，符号

表示不小于T_coh/T_short的最小整数。S1.4 Determine the number of signal segments according to the coherent integration time T _coh and the short integration time T _short

Among them, the symbol

Indicates the smallest integer not smaller than T _coh /T _short .

S1.5 Reset the short integration time T _short = T _coh /M. According to short integration time T _short and sampling rate F _s , determine parameter N=F _s ×T _short and parameter

上，得到新的信号

其中，Δt＝1/F_s表示采样时间间隔，

总长度为MN点，可分为M段长度为N点的信号。S2. Receive the _digital signal r ₀ (n) (n=0, 1, 2, ..., N-1, N, N+1, ..., 2N- 1, 2N, 2N+1, ..., MN-1). r ₀ (n) The total length is MN points, there are M segments in total, and each segment has N points. Then map this M segment signal to the first carrier frequency cell

on, get a new signal

Among them, Δt=1/F _s represents the sampling time interval,

The total length is MN points, which can be divided into M segments with a length of N points.

S3. According to the selected carrier frequency cell, adjust the local pseudo-code rate, generate a local pseudo-code signal, sample the local pseudo-code signal with F _s , select the local pseudo-code signal after point P sampling, and It performs FFT operation. This step further includes:

并生成本地伪码信号，f_chip表示伪码速率；S3.1 adjust the local pseudo-code rate according to the carrier frequency cell mapped in step S2

And generate a local pseudo code signal, f _chip represents the pseudo code rate;

S3.2 Sampling the local pseudo-code signal at the sampling frequency F _s , selecting the sampled local pseudo-code signal c(m) (m=0, 1, 2..., P-1) at point P, and then performing FFT operation, get:

$\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; c</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="j"\; filenames="HSA00000383226300021.png,HSA00000383226300032.png"\; state="\{\{state\}\}">j</figure-callout></span>}\frac{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}\mathrm{<span\; class="notranslate">\; mk</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0,1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>$

Among them, k represents the index.

对该序列进行FFT和共轭运算。S4. Select the first segment of the mapped received signal from step S2, fill the received signal with zeros to point P, and obtain the sequence

Perform FFT and conjugate operations on this sequence.

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="j"\; filenames="HSA00000383226300021.png,HSA00000383226300032.png"\; state="\{\{state\}\}">j</figure-callout></span>}\frac{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}\mathrm{<span\; class="notranslate">\; mk</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; *</span>}$

() ^* indicates complex conjugate operation.

S5. Multiply the results of step S3 and step S4, and then perform IFFT operation to obtain a correlation value sequence d(m) of length P.

Y(k)=R(k)C(k)

$\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; Y</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="e\; j"\; filenames="HSA00000383226300021.png,HSA00000383226300032.png"\; state="\{\{state\}\}">e</figure-callout></span><figure-callout\; id="2"\; label="e\; j"\; filenames="HSA00000383226300021.png,HSA00000383226300032.png"\; state="\{\{state\}\}">\; </figure-callout>}}^{\mathrm{<span\; class="notranslate">\; </span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0,1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}$

Store the correlation values of the first P-N points in d(m) as a row, and discard the remaining N point correlation values.

S6. Processing all received signals mapped to the carrier frequency cell in all M segments, and then obtaining a correlation value matrix of M rows and P-N columns;

The processing of all M segments mapped to the received signal on the carrier frequency cell is as follows:

Select the second segment of the mapped received signal from step S2, and then repeat step S4; slide the N-point local pseudo-code phase to regenerate the P-point local signal, then repeat step S3, and then repeat step S5.

Select the third segment of the mapped received signal from step S2, and then repeat the above processing until the Mth segment of the mapped received signal in step S2 is processed.

个零，使每列数据长度为

得到补零后的相关值矩阵H_i，对矩阵H_i按列进行FFT运算并取模，得到一个行P-N列的相关值矩阵G_i。S7. Supplement each column of data

zeros, so that the data length of each column is

Get the correlation value matrix H _i after zero padding, perform FFT operation on the matrix H _i column by column and take the modulo, and get a Correlation value matrix G _i of rows PN columns.

q=0, 1, 2, . . . , P-N-1,

d=0,1,2,...,

G _i (d, q) represents the element in row d and column q in matrix G _i , H _i (m, q) represents the element in row m and column q in matrix H _i , and ‖‖ represents a modulo operation.

行P-N列的非相干累加值矩阵 S8. Repeat step S2 to step S7 for a total of N _non times, and then perform non-coherent accumulation to obtain a size of

Incoherent summation matrix of rows PN columns

S9. If the maximum value in the non-coherent accumulated value matrix is greater than or equal to the preset threshold, then determine the pseudo code phase of the received signal according to the column where the maximum value is located, and determine the pseudocode phase of the received signal according to the row where the maximum value is located and the carrier mapped to the received signal Frequency cell, determine carrier frequency offset and code Doppler, otherwise execute step S10;

S10. If all the carrier frequency cells have been searched, slide the P-N point local pseudo-code phase, and re-execute steps S2-S9, otherwise, map the received signal to the next carrier frequency cell, and continue to execute steps S3-S9.

Taking the segmented zero padding algorithm commonly used in the GNSS field as an example, the acquisition performance of the acquisition method of the present invention and the algorithm is compared. Simulation conditions: the carrier frequency is 1575.42MHz, the sampling rate is 62MHz, the intermediate frequency is 46MHz, and the preset pseudocode phase position is 30720. Figures 5 to 8 show the correlation values of the row where the maximum correlation peak is located in the non-coherent correlation value matrix.

From Figure 5(a)-Figure 5(b), it can be seen that under the condition of SNR of -46dB, using 1ms coherent integration time and 100 times of non-coherent accumulation, the piecewise zero-padding algorithm cannot find the correct correlation Peak, that is, the pseudocode capture work cannot be completed. At this time, if the method of the present invention is adopted, an obvious correlation peak can be seen, that is, the invented method can complete the pseudo code capture work. Figure 6(a)-Figure 6(b) again compare the capture performance of the two capture methods when the number of non-coherent accumulations is increased to 300. It can be seen from the figure that despite increasing the number of non-coherent accumulations, the segmented zero-padding algorithm still cannot find the correct correlation peak, and still cannot complete the pseudo-code capture work. However, the method of the present invention presents a more obvious correlation peak after increasing the number of non-coherent accumulations. It can be seen that the method of the present invention has better capture performance.

Fig. 7 (a)-Fig. 7 (b) have compared the performance of the new method proposed by the present invention and the subsection zero-filling algorithm when the signal-to-noise ratio of the received signal is-46dB, the coherent integration time is 2ms, and the number of times of non-coherent accumulation is 300 times Capturing performance. It can be seen from the figure that the segmented zero padding algorithm cannot find the correct correlation peak, but the algorithm of the present invention presents a very obvious correlation peak. Further, FIG. 8 shows the capture performance of the method of the present invention when the signal-to-noise ratio of the received signal is -52dB. It can be seen from the figure that the method of the present invention can still find the correct correlation peak at this time, and successfully completes the pseudo code capture work. This further proves that the method of the present invention has better capture performance.

The present invention also provides a weak GNSS signal acquisition system under large-scale frequency offset, the system comprising:

The front-end AD conversion unit is used for digital sampling of the received analog signal.

The carrier frequency unit is used to roughly divide the carrier frequency to be searched, that is, determine the maximum carrier frequency search according to the coherent integration time T _coh set by the receiver, the number of non-coherent accumulation N _non , the carrier frequency f _carr , and the sampling frequency F _s The step size f _step divides the carrier frequency range to be searched into several carrier frequency unit cells according to the maximum carrier frequency search step size f _step , and divides the received signal into M segments, each segment contains N points, and each The segment signal is mapped to the selected carrier frequency cell.

The local pseudo-code generation unit is used to adjust the local pseudo-code rate according to the selected carrier frequency unit cell, generate the local pseudo-code signal, sample the local pseudo-code signal with F _s , and select the local pseudo-code signal after point P sampling Pseudo-code signal, and perform FFT operation on it. This unit further comprises: Compensation sub-unit, is used for calculating the current local pseudo-code rate that should be adjusted according to the carrier frequency unit cell of current rough search; Local pseudo-code generator, is used for generating the pseudo-code sequence to be searched; P point data selection The subunit is used to select the P-point local pseudo-code signal; the first FFT sub-unit is used to perform fast Fourier transform on the P-point pseudo-code signal.

The first processing unit is configured to select the first segment of signal mapped to the carrier frequency cell, pad it to point P with zeros, and perform FFT and conjugate operations on the zero-padded sequence. The unit further includes: an N-point data selection subunit, used to select the mapped N-point received signal; a zero-filling to P-point subunit, supplementing P-N zeros to the N-point received signal; an FFT and conjugate operation subunit for Perform a fast FFT on the received signal after zero padding at point P and take the complex conjugate.

A correlation operation unit, for multiplying the result of the local pseudo code generation unit and the result of the first processing unit, and performing an IFFT operation to obtain a correlation value sequence with a length of P, and correlating the first P-N points in the correlation value sequence Values are stored as one row, and the remaining N points of related values are discarded. The unit further includes: an IFFT subunit, which is used to multiply the result of the local pseudocode generation unit with the result of the first processing unit, and perform IFFT operation to obtain a correlation value sequence with a length of P; the previous P-N point correlation value operator A unit used to retain the top N points among the obtained P point correlation values.

The correlation value matrix generation unit is used to control and process all M segments mapped to the received signal on the carrier frequency cell, and obtain a correlation value matrix with M rows and P-N columns.

行P-N列的相关值矩阵。The second processing unit is used to perform zero padding, FFT operation and modulus for each column of data in the correlation value matrix, to obtain A matrix of correlation values for rows and columns. The unit further includes: a zero-padding subunit, which is used to perform zero-padding operation on each column of the correlation value matrix; a second FFT subunit, which is used to perform fast Fourier transform column-by-column to the zero-padding correlation value matrix, to obtain

A matrix of correlation values for rows and columns.

The non-coherent accumulation unit is used to non-coherently accumulate the different correlation values corresponding to the same code bit to obtain a size of A matrix of non-coherent summaries of rows and columns.

Threshold detection unit, for judging whether the maximum value in the non-coherent accumulated value matrix is greater than or equal to the preset threshold, if greater than or equal to the preset threshold, then determine the phase of the pseudo code of the received signal according to the column where the maximum value is located, and according to the maximum Determine the carrier frequency offset and code Doppler in the line where the value is located and the carrier frequency cell mapped to the received signal; otherwise, judge whether all the carrier frequency cells have been searched, and if they have been searched, slide the P-N point local pseudo-code phase , re-execute the loop processing, otherwise, map the received signal to the next carrier frequency cell, and continue to execute the loop processing.

Fig. 9 is a working principle block diagram of the GNSS signal acquisition system proposed by the present invention.

Example

In this embodiment, the carrier frequency is 1575.42MHz, the sampling rate is 62MHz, the intermediate frequency is 46MHz, the pseudo-code rate is 10.23Mchip/s, the coherent integration time is 2ms, the number of non-coherent accumulations is 10, and the carrier frequency range to be searched is -5kHz Taking 5kHz as an example, further illustrate the implementation steps of the weak GNSS signal acquisition method under the large-scale frequency offset of the present invention:

1. According to the coherent integration time of the receiver, the number of non-coherent accumulations, the carrier frequency, and the sampling rate, determine that the step size for the rough search of the carrier frequency is 1270 Hz.

2. Divide the carrier frequency range to be searched into 8 carrier frequency cells at intervals of 1270 Hz. Note that the frequencies of these carrier frequency cells are respectively -5000+1270k (k=0, 1, 2, . . . , 7).

3. According to the frequency search step length to determine the short integration time is 0.78ms.

4. Determine the number of signal segments M=3 according to the coherent integration time and the short integration time.

5. Reset the short integration time to T _short =2ms/3=0.6667ms. According to the short integration time and sampling rate, the parameter N=41335 and the parameter P=131072 are determined.

6. Receive 3 segments of digital signal r ₀ (n) each with a length of 41335 points sampled by the front-end AD, and map them to the -5000Hz carrier frequency cell.

7. Slow down the local pseudocode generator by 32.4675Hz and generate local pseudocode signals.

8. Sampling the local pseudo-code signal at a sampling rate of 62MHz, and selecting the sampled local pseudo-code signal of 131072 points, and then performing FFT operation.

9. Select the first segment of the mapped received signal from step 6 and fill it with zeros to 131072 points, then perform FFT and conjugate operations.

10. Multiply the results of step 8 and step 9, and then perform IFFT operation to obtain a correlation value sequence with a length of 131072.

11. Keep the first 89737 points in the result obtained in step 10, and store them as one row, and discard the remaining 41335 points.

12. Select the second segment of the mapped received signal from step 6, and repeat step 9. Slide the 41335-point local pseudo-code signal to regenerate the 131072-point local signal, and then repeat step 8. Repeat steps 10 and 11.

13. Select the third segment of the mapped received signal from step 6, and repeat step 12.

14. After processing in step 13, a correlation value matrix with 3 rows and 89737 columns will be obtained.

15. Add one zero to each column of data in the correlation value matrix, so that the length of each column of data is 4, and obtain a zero-padded correlation value matrix.

16. Carry out FFT operation and modulus on the column-by-column correlation value matrix after zero padding to obtain a new correlation value matrix with 4 rows and 89737 columns.

17. Repeat step 6 to step 16 for a total of 10 times, and then perform non-coherent accumulation to obtain a non-coherent accumulation value matrix with 4 rows and 89737 columns.

18. If the maximum value in the non-coherent cumulative value matrix is greater than or equal to the preset threshold, then determine the pseudo code phase of the received signal according to the column where the maximum value is located; according to the row where the maximum value is located and the carrier frequency grid mapped to the received signal in step 6 (ie -5000Hz) to determine carrier frequency offset and code Doppler.

19. If the maximum value in the non-coherent accumulated value matrix is less than the threshold, then in step 6, map the received signal to the next carrier frequency cell (ie -3730Hz) and repeat steps 6 to 18 until the 8 in step 2 is searched up to the carrier frequency unit.

20. After searching the 8 carrier frequency cells, if there is a correlation value greater than or equal to the preset threshold in the non-coherent cumulative value matrix, then determine the pseudo code phase, carrier frequency offset and code multiplicity of the received signal according to the method in step 18 Puller. Otherwise, slide the 89737-point local pseudo-code phase, and then repeat steps 6 to 19 to search for the next pseudo-code phase until a correlation value greater than or equal to the preset threshold appears in the non-coherent accumulated value matrix.

The above embodiments are only used to illustrate the present invention, but not to limit the present invention. Those of ordinary skill in the relevant technical field can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, all Equivalent technical solutions also belong to the category of the present invention, and the scope of patent protection of the present invention should be defined by the claims.

Claims (8)

1. A method for capturing weak GNSS signals under large-scale frequency offset is characterized by comprising the following steps:

s1, coherent integration time T set by receiver_cohNumber of times of sum noncoherent accumulation N_nonCarrier frequency f_carrAnd a sampling frequency F_sDetermining a maximum carrier frequency search step f_stepSearching the carrier frequency range to be searched according to the maximum carrier frequency searching step length f_stepDividing the carrier frequency into a plurality of carrier frequency cells;

s2, dividing the received signals into M sections, wherein each section comprises N points, and mapping each section of signals to a selected carrier frequency cell;

s3, according to the selected carrier frequency cell, adjusting the local pseudo code rate to generate a local pseudo code signal, and F_sSampling the local pseudo code signal, selecting the P-point sampled local pseudo code signal, and performing FFT operation on the local pseudo code signal;

s4, selecting the first section of signal mapped to the carrier frequency cell in the step S2, carrying out zero filling on the first section of signal to a point P, and carrying out FFT and conjugate operation on the sequence after zero filling;

s5, multiplying the results of the step S3 and the step S4, performing IFFT operation to obtain a correlation value sequence with the length of P, storing the previous P-N point correlation values in the correlation value sequence as a row, and discarding the rest N point correlation values;

s6, processing all M sections of received signals mapped to the carrier frequency unit grids to obtain a correlation value matrix of M rows and P-N columns;

s7, carrying out zero filling and FFT operation on each line of data in the correlation value matrix and carrying out modulus obtaining to obtain

A correlation value matrix of rows P-N columns;

s8, repeating the steps S2-S7 for N_nonSecondly, the results obtained each time are subjected to incoherent accumulation to obtain the value of

A row P-N column incoherent accumulated value matrix;

s9, if the maximum value in the incoherent accumulated value matrix is larger than or equal to a preset threshold, determining the pseudo code phase of the received signal according to the column of the maximum value, determining the carrier frequency offset and code Doppler according to the row of the maximum value and the carrier frequency cell mapped by the received signal, and otherwise, executing the step S10;

s10, if all carrier frequency cells are searched, sliding P-N local pseudo code phases, and executing steps S2-S9 again, otherwise, mapping the received signal to the next carrier frequency cell, and continuing to execute steps S3-S9. (Note: corresponding to the abstract attached figure)

Wherein,

T_short＝1/f_stepif T is_short＞T_cohThen get T_short＝T_coh，

Means not less than T_coh/T_shortN is F_s×T_short。

2. The method for weak GNSS signal acquisition under large scale frequency offset according to claim 1, wherein the step S1 further comprises:

s1.1 coherent integration time T according to receiver_cohNumber of incoherent accumulations N_nonCarrier frequency f_carrAnd a sampling frequency F_sDetermining a carrier frequency search step f_step：

<math><mrow><msub><mi>f</mi><mi>step</mi></msub><mo>=</mo><mfrac><msub><mi>f</mi><mi>carr</mi></msub><mrow><msub><mi>F</mi><mi>s</mi></msub><mo>&times;</mo><msub><mi>T</mi><mi>coh</mi></msub><mo>&times;</mo><msub><mi>N</mi><mi>non</mi></msub></mrow></mfrac><mo>,</mo></mrow></math>

If f_stepGreater than the carrier frequency range f to be searched_searchThen f is set_step＝f_search；

S1.2 according to said f_stepDividing the carrier frequency range to be searched into K ═ f_search/f_stepIndividual carrier frequency cells, which are respectively denoted as

(i＝0，2，...，K-1)；

S1.3 according to f_stepDetermining short integration time T_short＝1/f_stepIf T is_short＞T_cohThen set T_short＝T_coh；

S1.4 according to T_cohAnd T_shortDetermining the number of segments of a received signal

S1.5 resetting short integration time T_short＝T_coh/M according to T_shortAnd F_sDetermining N ═ F_s×T_shortAnd

3. the method as claimed in claim 1, wherein in step S2, T is received_cohLong sampled digital signal r₀(n)(n＝0，1，2，...，N-1，N，N+1，...，2N-1，2N，2N+1，...，MN-1)，r₀(n) with a total length of MN points, mapping the M signals to a first carrier frequency cell

To obtain a new signalWherein, Δ t is 1/F_sWhich represents the time interval between the sampling of the samples,

4. the method for weak GNSS signal acquisition under large scale frequency offset according to claim 1, wherein the step S3 further comprises:

s3.1 adjusting the local pseudo code rate according to the carrier frequency cell mapped by the step S2And generating a local pseudo-code signal, f_chipIs the pseudo code rate;

s3.2 by F_sSampling the local pseudo code signal, selecting a P-point sampled local pseudo code signal c (m) (0, 1, 2., P-1), and then performing FFT operation to obtain:

<math><mrow><mi>C</mi><mrow><mo>(</mo><mi>k</mi><mo>)</mo></mrow><mo>=</mo><munderover><mi>&Sigma;</mi><mrow><mi>m</mi><mo>=</mo><mn>0</mn></mrow><mrow><mi>P</mi><mo>-</mo><mn>1</mn></mrow></munderover><mi>c</mi><mrow><mo>(</mo><mi>m</mi><mo>)</mo></mrow><msup><mi>e</mi><mrow><mo>-</mo><mi>j</mi><mfrac><mrow><mn>2</mn><mi>&pi;</mi></mrow><mi>P</mi></mfrac><mi>mk</mi></mrow></msup><mo>,</mo><mi>k</mi><mo>=</mo><mn>0,1,2</mn><mo>,</mo><mo>.</mo><mo>.</mo><mo>.</mo><mo>,</mo><mi>P</mi><mo>-</mo><mn>1</mn><mo>.</mo></mrow></math>

5. the method for weak GNSS signal acquisition under large-scale frequency offset according to claim 4, wherein in step S4, the zero-padded sequence is:

(m-0, 1, 2.., P-1), performing FFT and conjugate operations on the sequence to yield:

<math><mrow><mi>R</mi><mrow><mo>(</mo><mi>k</mi><mo>)</mo></mrow><mo>=</mo><msup><mrow><mo>(</mo><munderover><mi>&Sigma;</mi><mrow><mi>m</mi><mo>=</mo><mn>0</mn></mrow><mrow><mi>P</mi><mo>-</mo><mn>1</mn></mrow></munderover><msubsup><mi>r</mi><mn>2</mn><mrow><mo>(</mo><mn>0</mn><mo>)</mo></mrow></msubsup><mrow><mo>(</mo><mi>m</mi><mo>)</mo></mrow><msup><mi>e</mi><mrow><mo>-</mo><mi>j</mi><mfrac><mrow><mn>2</mn><mi>&pi;</mi></mrow><mi>P</mi></mfrac><mi>mk</mi></mrow></msup><mo>)</mo></mrow><mo>*</mo></msup></mrow></math>

()^*representing a complex conjugate operation.

6. The method for weak GNSS signal acquisition under large scale frequency offset according to claim 1, wherein in step S6, the processing on all M pieces of received signals mapped onto the carrier frequency cell is: every time another signal mapped to the carrier frequency cell in the step S2 is selected, after the step S4 is executed, a new P-point local pseudo code signal is generated by sliding the N-point local pseudo code phase, the step S3 is executed again, and the process goes to the step S5.

7. The method for weak GNSS signal acquisition under large scale frequency offset according to claim 1, wherein the step S7 further comprises:

s7.1 supplementing each column of data in the correlation value matrix

Zero counting is carried out, and a correlation value matrix after zero padding is obtained;

s7.2, performing FFT operation on the zero-filled correlation value matrix according to columns and performing modulus obtaining to obtain the correlation value matrix

And the correlation value matrix of the rows P-N.

8. A weak GNSS signal acquisition system under large scale frequency offset, the system comprising:

carrier frequency unit for coherent integration time T set by receiver_cohNumber of times of sum noncoherent accumulation N_nonCarrier frequency f_carrAnd a sampling frequency F_sDetermining a maximum carrier frequency search step f_stepSearching the carrier frequency range to be searched according to the maximum carrier frequency searching step length f_stepDividing the signal into a plurality of carrier frequency cells, dividing the received signal into M sections, wherein each section comprises N points, and mapping each section of signal to a selected carrier frequency cell;

a local pseudo code generating unit for adjusting the local pseudo code rate according to the selected carrier frequency cell to generate a local pseudo code signal, as F_sSampling the local pseudo code signal, selecting the P-point sampled local pseudo code signal, and performing FFT operation on the local pseudo code signal;

the first processing unit is used for selecting the first section of signals mapped to the carrier frequency cell, carrying out zero filling on the first section of signals to a point P, and carrying out FFT and conjugate operation on the sequence after zero filling;

the correlation operation unit is used for multiplying the result of the local pseudo code generation unit and the result of the first processing unit, carrying out IFFT operation to obtain a correlation value sequence with the length of P, storing the previous P-N point correlation values in the correlation value sequence into a line, and discarding the rest N point correlation values;

the correlation value matrix generating unit is used for controlling and processing all M sections of received signals mapped to the carrier frequency cell and obtaining a correlation value matrix of M rows and P-N columns;

a second processing unit, configured to perform zero padding, FFT operation, and modulus extraction on each column of data in the correlation value matrix to obtain

A correlation value matrix of rows P-N columns;

an incoherent accumulation unit for performing incoherent accumulation on different correlation values corresponding to the same code bit to obtain a value of

A row P-N column incoherent accumulated value matrix;

the threshold detection unit is used for judging whether the maximum value in the incoherent accumulated value matrix is greater than or equal to a preset threshold, if so, determining the pseudo code phase of the received signal according to the column of the maximum value, and determining the carrier frequency offset and code Doppler according to the row of the maximum value and the carrier frequency cell mapped by the received signal; otherwise, judging whether all carrier frequency cells are searched, if so, sliding the P-N local pseudo code phase, and executing the cyclic processing again, otherwise, mapping the received signal to the next carrier frequency cell, and continuing to execute the cyclic processing.

Wherein,

T_short＝1/f_stepif T is_short＞T_cohThen get T_short＝T_coh，

Means not less than T_coh/T_shortN is F_s×T_short。

Images