CN104199057A - BOC signal unambiguous acquisition algorithm based on compressed sensing technology - Google Patents

Abstract

The invention discloses an unambiguous acquisition algorithm of BOC (binary offset carrier) modulated signal in GNSS (Global Navigation Satellite System) based on compressed sensing technology. The method divides the acquisition process of navigation signals into two stages to complete, and In the first stage, compressive sensing technology is used to process the received signal within a pseudo-code period, which greatly reduces the range of the initial phase estimation value of the received signal. In the second stage, GRASS (General Removing Ambiguity via Sidepeak Suppression) is only used for the reduced range data. Algorithms are used to process and complete signal capture. This method uses the combination of compressed sensing technology and GRASS algorithm to complete the unambiguous capture of BOC signals, which is suitable for all kinds of BOC signals, and compared with the traditional GRASS algorithm, the required hardware resources are greatly reduced.

Description

A BOC Signal Unambiguous Acquisition Algorithm Based on Compressed Sensing Technology

technical field

The present invention relates to the technical field of capturing navigation signals by receivers in GNSS (Global Navigation Satellite System), and more specifically, relates to a method for eliminating BOC (Binary Offset Carrier, binary offset) in GNSS based on compressed sensing technology. A capture algorithm for the ambiguity caused by the multi-peak characteristic of the autocorrelation function of the modulated signal.

Background technique

Since the global satellite navigation system can provide quasi-all-weather, high-precision, and automatic measurement navigation services, it has significant utilization value in both military and civilian fields.

However, with the continuous development and improvement of GNSS (Global Navigation Satellite System, Global Navigation Satellite System), the status of satellite navigation systems in the military and civilian fields is becoming more and more important, which directly leads to the tightness of the originally limited frequency band resources. And the mutual interference between navigation signals is becoming more and more serious. Therefore, BOC (Binary Offset Carrier, binary offset carrier) modulation came into being. The BOC modulated signal is a signal obtained by modulating the PRN sequence in the navigation signal by using a square wave subcarrier, which can be expressed as BOC(f _sc , f _c ), where f _sc =m×1.023MHz is the square wave frequency, f _c =n×1.023MHz is the spreading code rate, so we often abbreviate the BOC modulated signal as BOC(m,n). The BOC modulation signal has two significant advantages: one is spectrum splitting. The splitting characteristic of the spectrum of the BOC modulation signal makes its energy concentrated and distributed at the edge of the frequency band, so that it can share the frequency band with the BPSK signal and make full use of the frequency band resources. Each satellite navigation system will not interfere with each other on the basis of sharing the frequency band; the second is that the BOC modulated satellite signal has a sharper autocorrelation peak than the BPSK modulated satellite signal, which means that the BOC modulated signal has a stronger Anti-interference and higher positioning accuracy. Therefore, once BOC modulation was proposed, it attracted the attention of major navigation countries and organizations.

At present, the Galileo system has partially adopted the BOC modulation method, and in the GPS modernization signal, the newly added military signal and civilian signal will also adopt the BOC modulation method, and my country also announced it at the working group meeting of the International Committee on Global Satellite Navigation Systems The modulation method of the Beidou 2nd generation signal will all adopt the BOC modulation method.

The proposal of BOC modulation can well solve the problem of frequency congestion caused by the development of global satellite navigation systems, but the ambiguity caused by the multi-peak characteristics of the autocorrelation function of BOC modulation signals can easily make the navigation receiver difficult to capture and track. Missed detection and false detection occur, which makes the capture of BOC modulated signal much more complicated than traditional BPSK modulated signal.

At present, the methods to eliminate the ambiguity of the BOC modulated signal mainly include: 1) The sideband processing method (BPSK-like) uses three filters to filter out the upper and lower main lobes and the local code main lobe of the BOC signal respectively, and then performs BPSK capture on the signal, However, this method will cause 3dB energy loss during the implementation process, and at the same time, the processed correlation peak will also become wider, so that the BOC signal loses the advantage of high code tracking accuracy. 2) The basic idea of the Bump-Jumping method is to monitor the amplitude difference between the current locking correlation peak and its correlation peak in real time to determine whether it is false locking. However, this method has a high probability of missed detection and false alarms. Recovery time is long. 3) The subcarrier phase cancellation method (SCPC) can obtain a single triangular peak similar to the BPSK signal, but the triangular peak is wider than the autocorrelation peak of the original BOC signal, which loses the advantage of high precision of the BOC signal. 4) Autocorrelation Peak Cancellation Technology (ASPeCT) uses a new function to eliminate the secondary peak of the BOC signal. This method can retain the advantages of the BOC modulation signal and has become a widely used method in recent years, but this method is only applicable to BOC ( n,n) family of signals. 5) The side peak removal method (SCM) can completely eliminate the side peaks, but the relevant peaks will be reduced after capture. 6) The GRASS (General Removing Ambiguity via Sidepeak Suppression) algorithm is an extension of ASPeCT and is applicable to any BOC signal, but it requires a huge amount of storage space and resources, which is unbearable for embedded platforms with extremely limited hardware resources.

Contents of the invention

In view of the ambiguity brought by the BOC modulation signal and the above-mentioned other problems in the prior art, the purpose of the present invention is to provide a BOC signal ambiguity-free capture algorithm based on compressed sensing technology. The purpose of the present invention is achieved like this:

A. The antenna of the radio frequency front-end processing (1) receives the satellite signal, and then obtains the intermediate frequency signal through amplification and down-conversion processing, and provides it to the A/D conversion (2);

B, A/D conversion unit (2) determines the sampling frequency according to the modulation index α of the BOC modulation signal, the sampling frequency f _s =α f _c (f _c is the PN code rate in the navigation signal), and the received signal is sent after signal digitization Incoming carrier elimination processing unit (4);

C. The local carrier generator unit (3), generates the local carrier signal according to the Doppler frequency estimate value Δf returned by the ||w ₁ || ₀ > 0 decision unit (7) and w ₂ > r ₂ decision unit (10) c(n), and c(n) is sent to the carrier cancellation processing unit (4)

D, the carrier elimination processing unit (4), the local carrier c (n) that the local carrier generator (3) produces is multiplied with the received signal, filters out the high-frequency part, completes the carrier elimination of the received signal, and gets a pseudo-code period The length of the received signal r={r[0],r[1],...,r[αL-1]} ^T , send r to the first stage of observation processing (5);

E, the first-stage observation processing unit (5), first uses the transformation matrix Ψ to convert the signal to its sparse domain, obtains the sparse domain signal R, and sends R to the second-stage observation processing unit (8), and simultaneously converts R and The first-stage observation matrix Φ ₁ is multiplied to obtain the first-stage observation signal y ₁ , and the first-stage observation signal y ₁ is sent to the first-stage FWHT transformation unit (6);

F. The first-stage FWHT transformation unit (6), complete the FWHT transformation of the first-stage observation signal y ₁ to obtain the signal Z ₁ , find out the number of Z ₁ that is greater than the threshold r ₁ and not more than N _p , and use w ₁ Store the coordinates of these numbers in Z ₁ ;

G. Carry out ||w ₁ || ₀ > 0 judgment unit (7), if ||w ₁ || ₀ > 0, according to w ₁ and the modulation index α of the BOC modulation signal and the compression rate η of the first-stage observation process Calculate W ₁ , and send w ₁ and W ₁ to the second-stage observation processing unit (8); if ||w ₁ || ₀ <0, increase the estimated value of Doppler frequency shift Δf by a fixed step Δp Send into local carrier generator unit (3), enter step C;

H, the second-stage observation processing unit (8), according to w ₁ , W ₁ and GRASS algorithm, generates the second-stage observation matrix Φ ₂ and the second-stage sparse domain signal R', and then combines R' with the second-stage observation matrix R ' is multiplied to obtain the second-stage observation signal y ₂ , and the second-stage observation signal y ₂ is sent to the second-stage FWHT transformation unit (9);

1, the second stage FWHT transformation unit (9), completes the FWHT transformation of y ₂ , obtains signal Z ₂ , finds out maximum value w ₂ in Z ₂ ;

J. Carry out w ₂ >r ₂ judgment unit (10), compare w ₂ with the r ₂ threshold, if w ₂ <r ₂ , increase the Doppler frequency shift estimated value Δf by a fixed step Δp and send it to the local carrier The generator unit (3) enters step C, if w ₂ >r ₂ , then send the spreading code phase W ₁ (w ₂ ) and Doppler frequency shift Δf to the tracking module unit (11), and the acquisition is successful.

Compared with other technologies, the present invention has the following advantages:

①A method based on compressed sensing is used to complete the acquisition of BOC signals. Compared with parallel acquisition, the number of correlators required is less, and compared with FFT-based acquisition methods, the amount of calculation is less.

②A combination of compressed sensing technology and GRASS algorithm is used to complete the unambiguous capture of BOC signals. This method is suitable for all types of BOC signals, and compared with the traditional GRASS algorithm, the required hardware resources are greatly reduced.

Description of drawings

The invention and its advantages and other features can be better understood by reading the following detailed description of the invention in conjunction with the accompanying drawings, in which:

The accompanying drawing shows the process of capturing the BOC navigation signal by the unambiguous capture algorithm based on compressed sensing technology.

Detailed ways

In order to better understand the present invention, specific embodiments of the present invention will be described in detail below.

The accompanying drawing shows the process of capturing BOC navigation signals by the unambiguous capture algorithm based on compressed sensing technology

A. The antenna of the radio frequency front-end processing (1) receives the satellite signal, and then obtains the intermediate frequency signal through amplification and down-conversion processing, and provides it to the A/D conversion (2);

The BOC navigation signal received by the receiving antenna can be expressed as

A, τ, f _IF , f _D and They are amplitude, PN code phase, intermediate frequency, Doppler frequency shift and carrier phase. D(t), n(t) are respectively the navigation data and the bilateral power spectral density as N ₀ /2 Gaussian noise, C(t) is the PN code with code rate f _c =n×1.023MHz and period L=1023, sc (t) is the subcarrier, and its mathematical expression is

$\mathrm{<span\; class="notranslate">\; sc</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; sgn</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; sc</span>}}\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; BOC</span>}\\ \mathrm{<span\; class="notranslate">\; sgn</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; cos</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; sc</span>}}\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; BOC</span>}\end{array}\right.$

Where f _sc =m×1.023MHz. α=2m/n is called the modulation index of BOC(m,n).

B. In order to maintain the characteristics of high tracking accuracy of BOC(m,n), the sampling frequency of A/D conversion (2) is: f _s = αf _c , after A/D conversion (2) digitizes the signal, it is used for carrier cancellation process(4);

C. The local carrier generator (3), according to the Doppler frequency estimated value Δf returned by ||w ₁ || ₀ > 0 judgment (7) and w ₂ > r ₂ judgment (10), adopts Generate the local carrier signal c(n), send the local carrier signal c(n) to the carrier cancellation process (4), where f _d is the Doppler frequency shift initially estimated by the receiving end, and the initial value of Δf is 0;

D, carrier elimination process (4), the local carrier c (n) that local carrier generator (3) produces is multiplied with received signal, filters out high-frequency part, finishes the carrier elimination of received signal, gets a pseudo-code cycle length The received signal r={r[0],r[1],...,r[αL-1]} ^T , send r to the first stage of observation processing (5);

E. The first stage of observation processing (5), first use the transformation matrix Ψ to convert the signal to its sparse domain, and obtain its sparse domain signal R, send R to the second stage of observation processing (8), and at the same time combine R with the first The stage observation matrix Φ ₁ is multiplied to obtain the first stage observation signal y ₁ , and the first stage observation signal y ₁ is sent to the first stage FWHT transformation (6).

Compressed Sensing technology believes that as long as the signal is sparse in a certain orthogonal space, the signal can be sampled at a lower frequency, and the signal can be reconstructed with a high probability, because the GNSS signal has a strong autocorrelation domain. Sparsity, so the use of compressed sensing technology can reduce the amount of data that needs to be processed by the receiver during the GNSS acquisition process.

The transformation matrix Ψ of the GNSS signal can be obtained by the following formula:

Here B(t)=C(t)sc(t), l,k∈{0,1,2,...,αL-1}, the received signal and when the Doppler frequency estimated by the receiver is close to the received signal When Doppler shifted, the BOC signal can be converted to its sparse domain by multiplying the received signal r by the transformation matrix Ψ. At the same time, according to the compressed sensing theory, we use the observation matrix Φ ₁ to observe the sparse domain signal R. Each element in Φ ₁ can be obtained by the following formula:

1≤m≤M ₁ ,0≤l≤αL-1, is a Hadamard matrix of size M ₁ ×M ₁ We define η as the compression ratio, Multiply the observation matrix Φ ₁ with the sparse domain signal R to obtain the first-stage observation signal y ₁ .

y ₁ =Φ ₁ R=Φ ₁ Ψr

F. The first stage of FWHT transformation (6), complete the FWHT transformation of the observation signal y ₁ in the first stage to obtain the signal Z ₁ , find out the number of Z ₁ that is greater than the threshold r ₁ and not more than N _p , and store it in w ₁ The coordinates of these numbers in _Z1 ;

${\mathrm{<span\; class="notranslate">\; Z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; point\; FWHTof</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

G. Carry out ||w ₁ || ₀ > 0 judgment (7), calculate the number of values ||w ₁ || ₀ in Z ₁ greater than the threshold r ₁ , if ||w ₁ || ₀ > 0, according to w ₁ and the modulation index α and compression ratio η of the BOC modulation signal to calculate W ₁ , $W_1=\underset{w_1\&Subset;w_1}{\&cup;}\{{\mathrm{\&eta;w}}_1+1,{\mathrm{\&eta;w}}_1+2,...,\min \{\mathrm{\&eta;}(w_1+1),\mathrm{\&alpha;\; L}\left\}\right\}$ And send w ₁ and W ₁ to the second stage of observation processing (8); if ||w ₁ || ₀ <0, increase the estimated value of Doppler frequency shift Δf with a fixed step Δp and send it to the local carrier generator (3), enter step C;

H, second-stage observation processing (8), according to w ₁ , W ₁ and GRASS algorithm, generate the second-stage observation matrix Φ ₂ and the second-stage sparse domain signal R′, and then combine R′ with the second-stage observation matrix R ′ multiplied to obtain the second-stage observation signal y ₂ , and send the second-stage observation signal y ₂ into the second-stage FWHT transformation (9).

We view the BOC(m,n) signal as a vector of shape The SCS signal of the PN code, by multiplying each chip of the PN code by d _B , we can get the corresponding BOC signal.

We use a new correlation function R' to capture:

${\mathrm{<span\; class="notranslate">\; R</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; R</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

Here R _B/L = Ψ _L r, L(t) is the local SCS signal, its shape vector $d_L={[d_0,d_1,...,d_{\mathrm{\&alpha;}-1}]}_{\mathrm{\&alpha;}\&times;1}^T,$ here

$\left\{\begin{array}{c}{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}}}\\ {\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\frac{{<span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; .</span>\end{array}\right.$

In the second stage of observation processing, we use a simplified GRASS algorithm, first let And the coordinates of each element in W ₁ are ν, and the generated size is Hadamard matrix Then calculate the second-stage correlation function R′ and transformation matrix Φ ₂ , $R_{l.m}^{\&prime;}=R^2\left(l\right)-R_{B/L}^2\left(l\right),$ l=W ₁ (ν), ${\mathrm{\&Phi;}}_2=W_{m_{02}}/\sqrt{m_2},$ Here ν={1,2,…,||W ₁ || ₀ }, is the element in row m and column l in R ₂ , R(l), R _B/L (l) are the l elements in R and R _B/L respectively, and finally calculate the second-stage observation signal y ₂ , y ₂ = Φ ₂ R'.

1, second stage FWHT conversion (9), complete the FWHT conversion of y ₂ , obtain signal Z ₂ , find out maximum value _{w 2} in Z 2;

w ₂ ＝{arg max Z ₂ }∩{arg Z ₂ ≥r ₁ }

J. Carry out w ₂ > r ₂ judgment (10), compare w ₂ with the r ₂ threshold, if w ₂ > r ₂ , then shift the spreading code phase W ₁ (w ₂ ) and Doppler frequency f _d +Δf is sent to the tracking module (11), and the acquisition is successful; if w ₂ <r ₂ , the estimated Doppler frequency shift value Δf is increased by a fixed step Δp and then sent to the local carrier generator (3), and enters step C.

According to the above algorithm flow, the received GNSS signal is converted to its sparse domain by using the corresponding transformation matrix, and the first stage of observation processing (5), the first stage of FHWT transformation (6) and ||w ₁ || ₀ >0 After judgment (6), the estimated phase value of the received signal can be locked within the range of L/M ₁ PN symbols. In the second stage of observation processing, only the L/M ₁ PN symbols need to be The αL/M ₁ data in the symbol is processed by the GRASS algorithm, which greatly reduces the amount of calculation compared with the traditional GRASS algorithm, and the required hardware resources are also greatly reduced.

It should be noted:

1. For GNSS signals, when the carrier frequency is determined, the maximum Doppler frequency shift of the signal is within a fixed range. Therefore, if the estimated Doppler frequency value Δf exceeds this range, it indicates that the capture has failed.

2. In the first stage of observation processing (5), an observation matrix with a size of M ₁ ×M ₁ is used, where the size of M ₁ determines the compression rate of the algorithm, and also affects the detection probability of the algorithm, so M The size of ₁ can be determined according to the actual situation.

3. In the first stage of FHWT transformation (6), after the signal is changed by FHWT, it is necessary to find out the number of Z ₁ that is greater than the threshold r ₁ and not more than N _p , where the size of N _p is important for the second stage observation processing In (8), the amount of processed data and the detection probability are affected, so the size of N _p should be adjusted and set according to the actual situation.

Claims (1)

1. the BOC signal based on compressed sensing technology is without a fuzzy acquisition algorithm, at receiving end by the navigation satellite signal of radio-frequency front-end antenna reception through amplifying, after down-converted, obtain intermediate-freuqncy signal, signal capture process comprises the following steps:

A, determine sample frequency according to the modulation index α of BOC modulation signal, intermediate-freuqncy signal is carried out to digitizing, produce local carrier signal according to Doppler frequency estimated value Δ f simultaneously, complete the carrier wave elimination that receives signal, get the reception signal r={r[0 of a pseudo-code Cycle Length], r[1] ..., r[α L-1] } ^t, L=1023;

B, first adopt transition matrix Ψ that reception signal is transformed into its sparse territory, obtain sparse territory signal R, simultaneously by R and first stage observing matrix Φ ₁multiply each other and obtain first stage observation signal y ₁; To y ₁carry out FWHT conversion and obtain signal Z ₁, find out Z ₁in be greater than thresholding r ₁and no more than N _pnumber, and use w ₁deposit these numbers at Z ₁in coordinate; Right || w ₁|| ₀> 0 adjudicates, if || w ₁|| ₀> 0, calculates W ₁, here $W_1=\underset{w_1\&Subset;w_1}{\&cup;}\{{\mathrm{\&eta;w}}_1+1,{\mathrm{\&eta;w}}_1+2,...,\min \{\mathrm{\&eta;}(w_1+1),\mathrm{\&alpha;L}\left\}\right\},$ If || w ₁|| ₀< 0, increases fixing stepping Δ p by Doppler frequency estimation value Δ f, returns to steps A, proceeds to catch;

It is characterized in that:

C, order and W ₁in the coordinate of each element be ν, ν ∈ 1,2 ..., || W ₁|| ₀, generation size is hadamard matrix generate subordinate phase observing matrix Φ ₂with the sparse territory of subordinate phase signal R ', then calculate subordinate phase observation signal y ₂: y ₂=Φ ₂r '; Φ ₂expression formula be the capable l column element of m of R ' can be expressed as: $R_{m.l}^{\&prime;}=\left\{\begin{array}{cc}{R}^2\left(l\right)-{\mathrm{aR}}_{B/L}^2\left(l\right),& l=W_1\left(\mathrm{\&nu;}\right)\\ 0,& l\&NotEqual;W_1\left(\mathrm{\&nu;}\right)\end{array}\right.,$ M ∈ 1,2 ..., M ₂, a=2 α-3, R (l), R _b/L(l) be respectively R, R _b/Lin l element, R _b/Lit is the auxiliary signal being calculated by GRASS algorithm;

D, complete y ₂fWHT conversion, obtain signal Z ₂, find out Z ₂middle maximal value w ₂; By w ₂with thresholding r ₂compare, if w ₂< r ₂, by returning to steps A after fixing Doppler frequency estimation value Δ f increase stepping Δ p, proceed to catch, if w ₂> r ₂, obtain spreading code phase place W ₁(w ₂) and Doppler shift Δ f, acquisition success.