CN104579414A - Method for adaptively acquiring P codes on basis of averaging technology - Google Patents

Abstract

The invention provides a method for adaptively acquiring P codes on the basis of an averaging technology. The method is an algorithm for quickly directly acquiring the P codes on the basis of circular correlation theories, each N points in pseudo code sequences of received signals are sequentially subjected to averaging processing to form a new sequence, local pseudo code sequences are subjected to similar processing to form a new sequences, each two corresponding new sequences are subjected to circular correlation processing, particular code intervals where code elements in the received signals are located can be judged when correlation peak values are higher than judgment thresholds, and the exact positions of the code elements can be determined. The method is equivalent to the basic averaging technology when the signals are high in intensity and has the advantage of calculation economization.

Description

A kind of P code self-adapting catching method based on averaging method

Technical field

The invention belongs to signal capture field, be specifically related to a kind of P code self-adapting catching method based on averaging method.

Background technology

Global positioning system (GPS) can provide standard positioning services (SPS) and precision positioning service (PPS) for user.PPS can provide more accurate positioning precision and time service precision.Thering is provided of SPS is based on C/A code, and providing of PPS is based on P code.Catching of general P code is captured as basis with C/A code, and catch P coded signal according to the information that C/A code provides, but C/A code is easily interfered, the Direct Acquisition therefore for P code just becomes extremely important.P code is the pseudo noise code that a kind of cycle is long, speed is high, and the difficulty of Direct Acquisition is comparatively large, needs the capture time grown very much and huge operand.

At present, a lot of research has been carried out to the algorithm of P code Direct Acquisition abroad.1997, J.B.Lozow analyzed adopting the P Code acquisition of serial and parallel search during incoherent detection, gives mean acquisition time.Direct Acquisition Methods for P code can be roughly divided into two classes: 1. time domain acquisition algorithm, RobertWolfert and Steve Chen proposed in 1998 the method realizing P code Direct Acquisition with Y-EXPRESS processor, and the method mainly realizes based on large-scale parallel correlator.2. acquisition in frequency domain algorithm, Jing Pang proposed directly average and overlapping average algorithm in 2003.This average Direct Acquisition technology makes to carry out catching of P code with FFT becomes possibility, but introduces larger multi-access inference while reduction Direct Acquisition complexity.In recent years domestic, also many to the research of P code Direct Acquisition Methods, replace FFT to carry out time domain to frequency-domain calculations with wavelet transformation, can data volume be reduced, roughly estimate Doppler frequency; Large quantity research concentrates on the specific implementation to existing algorithm in addition.

The method of the P code Direct Acquisition that domestic and international appearance is a lot, what wherein have certain using value is averaging method, but realizing and seldom considering the snr loss that algorithm itself causes in application process, makes, when signal is more weak, cannot meet contact conditions.

Summary of the invention

The object of the invention is to overcome above-mentioned deficiency, provide a kind of P code self-adapting catching method based on averaging method, when signal is stronger, tool averaging method amount of calculation is less; And when signal is more weak, by irrelevant cumulative mode, increase detection signal-to-noise ratio, still can carry out catching of signal when averaging method cannot meet and catch requirement.

In order to achieve the above object, the present invention includes following steps:

Step one: adopt receiver reception from the signal of satellite, obtain the signal to noise ratio of satellite-signal, if signal to noise ratio is less than 40dB, then continues to receive; If signal to noise ratio is greater than 40dB, the signal received is carried out down-conversion and A/D conversion, become digital intermediate frequency sampling signal;

Step 2: introduce white noise in digital intermediate frequency sampling signal, and carry out the despreading of m long code after adopting correlator and digital mixer to remove intermediate frequency amplitude;

Step 3: the pseudo-code sequence in the signal after despreading is carried out average value processing successively according to every N point, forms new signal pseudo-code sequence, introduces local pseudo-code sequence and carries out average value processing successively according to every N point equally, form new local pseudo-code sequence;

Step 4: new signal pseudo-code sequence and local pseudo-code sequence are carried out circular correlation process, when there is the correlation peak higher than decision threshold, just the code element correspondence in signal can be gone out specific code in local pseudo-code sequence interval, and then determine accurate location, complete and catch.

In described step one, signal to noise ratio is according to the ICD file of gps system, and the transmitting equivalent isotropically radiated power of the P coded signal of L1 frequency range is EIRP _l1-P, gps signal is sent to ground from the low-orbit satellite apart from ground R to come, and its free space attenuation is wherein λ is the wavelength of gps signal, and the decay of real atmosphere layer is about A ≈ 2dB, and therefore the signal strength signal intensity on L1 frequency range P code arrival ground is:

P _L1-P＝EIRP _L1-P-F-A；

The principal element affecting GPS noise is ambient temperature and noise of equipment, and therefore the noise temperature of unitized GPS is equivalent to 513K, and therefore the noise density of GPS is N ₀=kT, wherein k is Boltzmann constant, and T is thermodynamic temperature;

Therefore the noise power of known P coded signal is wherein B is the distance of gps signal to ground;

P coded signal arrives the signal strength signal intensity on ground, and therefore carrier-to-noise ratio is signal to noise ratio is

In described step 2, despreading process is as follows:

Be located at time period T, have digitised Intermediate Frequency sampled signal S ₁, S ₂, S _{m × n}, after removal intermediate frequency, amplitude is V _{s (1)}, V _{s (2)}..., V _{s (m × n)}, then $\left|V_{S\left(1\right)}\right|\≈\left|V_{S\left(2\right)}\right|\≈...\≈\left|V_{S(m\×n)}\right|\≈\left|{\stackrel{\‾}{V}}_S\right|,$ Signal power is:

$\begin{array}{c}{P}_{\mathrm{SIN}}=\frac{V_{S\left(1\right)}^2+V_{S\left(2\right)}^2...+V_{S(m\×n)}^2}{m\×n}\\ =m\×n\frac{{\stackrel{\‾}{V}}_S^2}{m\×n}n={\stackrel{\‾}{V}}_S^2\end{array}$

Noise N ₁, N ₂..., N _{m × n}power is:

$\begin{array}{c}{P}_{\mathrm{NIN}}=\frac{V_{n\left(1\right)}^2+V_{n\left(2\right)}^2...+V_{n(m\×n)}^2}{m\×n}\\ =E(m\×n\·\frac{V_{n\left(k\right)}^2}{m\×n})\\ =E\left(V_{n\left(k\right)}^2\right)\\ ={\mathrm{\δ}}^2\end{array}$

Before despreading, signal to noise ratio is

Carry out the despreading of m long symbols, then signal power is:

$P_{\mathrm{SOUT}}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{S\left(i\right)}\right)}^2}{n}=m^2\×n\frac{{\stackrel{\‾}{V}}_S^2}{n}=m^2\×{\stackrel{\‾}{V}}_S^2$

Carry out the despreading of m long symbols, then noise power is:

$P_{\mathrm{NOUT}}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{N\left(i\right)}\right)}^2}{n}=n\×\frac{E\left({\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{N\left(i\right)}\right)}^2\right)}{n}=m\×{\mathrm{\δ}}^2$

After despreading, signal to noise ratio is therefore the gain that spread spectrum brings is 10log (m).

In described step 3, the digital medium-frequency signal sequence after down-converted is G ₁=[S ₁, S ₂..., S _{m × n}], average value processing is carried out to it, carry out L point average after:

$G_2=[\frac{\underset{i=1}{\overset{L}{\mathrm{\Σ}}}{S}_i}{L},\frac{\underset{i=L+1}{\overset{2\×L}{\mathrm{\Σ}}}{S}_i}{L}...,\frac{\underset{i=(K-1)\×L+1}{\overset{K\×L}{\mathrm{\Σ}}}{S}_i}{L}]$

K×L＝m×n，K,L,m,n∈N

The local random code sequence produced is G ₃=[P ₁, P ₂..., P _{m × n}], carry out L point average after:

$G_4=[\frac{\underset{i=1}{\overset{L}{\mathrm{\Σ}}}{P}_i}{L},\frac{\underset{i=L+1}{\overset{2\×L}{\mathrm{\Σ}}}{P}_i}{L}...,\frac{\underset{i=(K-1)\×L+1}{\overset{K\×L}{\mathrm{\Σ}}}{P}_i}{L}].$

In described step 3, after average value processing is carried out to the pseudo-code sequence in the signal after despreading and local pseudo-code sequence, signal signal to noise ratio after average value processing being less than to 40dB carries out irrelevant cumulative, and irrelevant cumulative gain is that relevant cumulative gain deducts irrelevant cumulative loss:

G _i(n)＝G _e(n)-L(n)

G _e＝10lg(n)

$L\left(n\right)=101g\left[\frac{1+\sqrt{1+\frac{9.2n}{{[{\mathrm{erfc}}^{-1}\left({2P}_f\right)-{\mathrm{erfc}}^{-1}\left({2P}_d\right)]}^2}}}{1+\sqrt{1+\frac{9.2}{{[{\mathrm{erfc}}^{-1}\left({2P}_f\right)-{\mathrm{erfc}}^{-1}\left({2P}_d\right)]}^2}}}\right]$

In formula: n is accumulative frequency, p _ffor false alarm probability, P _dfor inspection probability.

Compared with prior art, the one quick P code Direct Acquisition algorithm that the present invention proposes based on the theory of circular correlation, every N point in Received signal strength pseudo-code sequence is carried out average value processing successively and forms new sequence, local pseudo-code sequence also carries out same process and forms new sequence, and then two new sequences carry out circular correlation process, when there is the correlation peak higher than decision threshold, just can judge that the specific code at the code element place in Received signal strength is interval, and then determine its accurate location, the present invention is equivalent to basic averaging method when signal is stronger, there is the advantage of saving amount of calculation.

Further, the present invention is more weak at signal, when basic averaging method cannot meet contact conditions, the present invention is based on averaging method self-adaptive direct and catches algorithm and also can be reached by the mode of irrelevant accumulation and carry out Direct Acquisition.

Accompanying drawing explanation

Fig. 1 is flow chart of the present invention;

Fig. 2 is the graph of a relation of L parameter of the present invention and snr loss;

Fig. 3 is averaging method correlation peak figure;

Fig. 4 is for catch algorithm correlation peak figure based on averaging method self-adaptive direct.

Embodiment

Below in conjunction with drawings and Examples, the present invention will be further described

See Fig. 1, the present invention includes following steps:

Step one: adopt receiver reception from the signal of satellite, signal to noise ratio is according to the ICD file of gps system, and the transmitting equivalent isotropically radiated power of the P coded signal of L1 frequency range is EIRP _l1-P, gps signal is sent to ground from the low-orbit satellite apart from ground R to come, and its free space attenuation is wherein λ is the wavelength of gps signal, and the decay of real atmosphere layer is about A ≈ 2dB, and therefore the signal strength signal intensity on L1 frequency range P code arrival ground is: P _l1-P=EIRP _l1-P-F-A;

The principal element affecting GPS noise is ambient temperature and noise of equipment, and therefore the noise temperature of unitized GPS is equivalent to 513K, and therefore the noise density of GPS is N ₀=kT, wherein k is Boltzmann constant, and T is thermodynamic temperature;

Therefore the noise power of known P coded signal is wherein B is the distance of gps signal to ground;

P coded signal arrives the signal strength signal intensity on ground, and therefore carrier-to-noise ratio is signal to noise ratio is

If signal to noise ratio is less than 40dB, then continue to receive; If signal to noise ratio is greater than 40dB, the signal received is carried out down-conversion and A/D conversion, become digital intermediate frequency sampling signal;

Step 2: introduce white noise in digital intermediate frequency sampling signal, and carry out the despreading of m long code after adopting correlator and digital mixer to remove intermediate frequency amplitude;

Despreading process is as follows, is located at time period T, has digitised Intermediate Frequency sampled signal S ₁, S ₂..., S _{m × n}, after removal intermediate frequency, amplitude is V _{s (1)}, V _{s (2)}, V _{s (m × n)}, then | V _{s (1)}| ≈ | V _{s (2)}| ≈ ... ≈ | V _{s (m × n)}| ≈ V _s|, signal power is:

$\begin{array}{c}{P}_{\mathrm{SIN}}=\frac{V_{S\left(1\right)}^2+V_{S\left(2\right)}^2...+V_{S(m\×n)}^2}{m\×n}\\ =m\×n\frac{{\stackrel{\‾}{V}}_S^2}{m\×n}n={\stackrel{\‾}{V}}_S^2\end{array}$

Noise N ₁, N ₂..., N _{m × n}power is:

$\begin{array}{c}{P}_{\mathrm{NIN}}=\frac{V_{n\left(1\right)}^2+V_{n\left(2\right)}^2...+V_{n(m\×n)}^2}{m\×n}\\ =E(m\×n\·\frac{V_{n\left(k\right)}^2}{m\×n})\\ =E\left(V_{n\left(k\right)}^2\right)\\ ={\mathrm{\δ}}^2\end{array}$

Before despreading, signal to noise ratio is

Carry out the despreading of m long symbols, then signal power is:

$P_{\mathrm{SOUT}}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{S\left(i\right)}\right)}^2}{n}=m^2\×n\frac{{\stackrel{\‾}{V}}_S^2}{n}=m^2\×{\stackrel{\‾}{V}}_S^2$

Carry out the despreading of m long symbols, then noise power is:

$P_{\mathrm{NOUT}}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{N\left(i\right)}\right)}^2}{n}=n\×\frac{E\left({\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{N\left(i\right)}\right)}^2\right)}{n}=m\×{\mathrm{\δ}}^2$

After despreading, signal to noise ratio is therefore the gain that spread spectrum brings is 10log (m)

Step 3: the pseudo-code sequence in the signal after despreading is carried out average value processing successively according to every N point, forms new signal pseudo-code sequence, the digital medium-frequency signal sequence after down-converted is G ₁=[S ₁, S ₂..., S _{m × n}], average value processing is carried out to it, carry out L point average after:

$G_2=[\frac{\underset{i=1}{\overset{L}{\mathrm{\Σ}}}{S}_i}{L},\frac{\underset{i=L+1}{\overset{2\×L}{\mathrm{\Σ}}}{S}_i}{L}...,\frac{\underset{i=(K-1)\×L+1}{\overset{K\×L}{\mathrm{\Σ}}}{S}_i}{L}]$

K×L＝m×n，K,L,m,n∈N

Introduce local pseudo-code sequence and carry out average value processing successively according to every N point equally, form new local pseudo-code sequence, the local random code sequence produced is G ₃=[P ₁, P ₂..., P _{m × n}], carry out L point average after:

$G_4=[\frac{\underset{i=1}{\overset{L}{\mathrm{\Σ}}}{P}_i}{L},\frac{\underset{i=L+1}{\overset{2\×L}{\mathrm{\Σ}}}{P}_i}{L}...,\frac{\underset{i=(K-1)\×L+1}{\overset{K\×L}{\mathrm{\Σ}}}{P}_i}{L}];$

After carrying out average value processing to the pseudo-code sequence in the signal after despreading and local pseudo-code sequence, the signal signal to noise ratio after average value processing being less than to 40dB carries out irrelevant cumulative, and irrelevant cumulative gain is that relevant cumulative gain deducts irrelevant cumulative loss:

G _i(n)＝Ge(n)-L(n)

G _e＝10lg(n)

$L\left(n\right)=101g\left[\frac{1+\sqrt{1+\frac{9.2n}{{[{\mathrm{erfc}}^{-1}\left({2P}_f\right)-{\mathrm{erfc}}^{-1}\left({2P}_d\right)]}^2}}}{1+\sqrt{1+\frac{9.2}{{[{\mathrm{erfc}}^{-1}\left({2P}_f\right)-{\mathrm{erfc}}^{-1}\left({2P}_d\right)]}^2}}}\right]$

In formula: n is accumulative frequency, p _ffor false alarm probability, P _dfor inspection probability;

Step 4: new signal pseudo-code sequence and local pseudo-code sequence are carried out circular correlation process, when there is the correlation peak higher than decision threshold, just the code element correspondence in signal can be gone out specific code in local pseudo-code sequence interval, and then determine accurate location, complete and catch.

Embodiment 1:

Step one: according to the ICD file of gps system, the transmitting equivalent isotropically radiated power EIRP of the P coded signal of L1 frequency range is:

EIRP _L1-P＝239.32W＝23.8dBW (1)

Gps signal is sent to ground from the low-orbit satellite apart from ground R=20000km to come, and its free space attenuation is:

$F={\left(\frac{\mathrm{\λ}}{4\mathrm{\πR}}\right)}^2=\left(\frac{0.19}{4\mathrm{\π}\×2\×{10}^7}\right)\≈5.73\×{10}^{19}\≈182.4\mathrm{dB}---\left(2\right)$

Wherein, λ is * *;

Be more than the result that vacuum obtains, real atmosphere layer decay to A ≈ 2dB, therefore L1 frequency range P code arrives the signal strength signal intensity on ground and is:

P _L1-P＝EIRP _L1-P-F-A＝23.8-182.4-2.0＝-160.6dBW＝-130.6dBmW (3)

The principal element affecting GPS noise is ambient temperature and noise of equipment, and therefore the noise temperature of unitized GPS is equivalent to 513K, and the noise density of GPS is:

N ₀＝kT＝10lg(1.38×10 ^-23×513)＝-201.5dBW/Hz＝-171.5dBmW/Hz (4)

Therefore the noise power of known P coded signal:

$\begin{array}{c}{P}_{N_0-P}=\mathrm{ktB}\\ =101g(1.39\×{10}^{-23}\×513\×2\×10^7)\\ =-128.5\mathrm{dBW}=-98.5\mathrm{dBmW}\end{array}---\left(5\right)$

Know that P coded signal arrives the signal strength signal intensity on ground by formula (3), therefore carrier-to-noise ratio is:

$\frac{P_{L1-P}}{N_0}=-130.6-(-171.5)=40.9\mathrm{dB}/\mathrm{Hz}---\left(6\right)$

Signal to noise ratio is:

$\frac{P_{L1-P}}{N_0-L1-P}=-130.6-(-98.5)=-32.\mathrm{dB}---\left(7\right)$

Signal carries out down-conversion and A/D conversion through radio-frequency module, becomes digital intermediate frequency sampling signal, transfers to baseband module to process.

Step 2: introduce white noise in digital intermediate frequency sampling signal, and carry out the despreading of m long code after adopting correlator and digital mixer to remove intermediate frequency amplitude,

Be located at time period T, have digitised Intermediate Frequency sampled signal S ₁, S ₂..., S _{m × n}, after removal intermediate frequency, amplitude is V _{s (1)}, V _{s (2)}..., V _{s (m × n)}, then signal power is:

$\begin{array}{c}{P}_{\mathrm{SIN}}=\frac{V_{S\left(1\right)}^2+V_{S\left(2\right)}^2...+V_{S(m\×n)}^2}{m\×n}\\ =m\×n\frac{{\stackrel{\‾}{V}}_S^2}{m\×n}n={\stackrel{\‾}{V}}_S^2\end{array}$

Noise N ₁, N ₂..., N _{m × n}power is:

$\begin{array}{c}{P}_{\mathrm{NIN}}=\frac{V_{n\left(1\right)}^2+V_{n\left(2\right)}^2...+V_{n(m\×n)}^2}{m\×n}\\ =E(m\×n\·\frac{V_{n\left(k\right)}^2}{m\×n})\\ =E\left(V_{n\left(k\right)}^2\right)\\ ={\mathrm{\δ}}^2\end{array}$

Before despreading, signal to noise ratio is

Carry out the despreading of m long symbols, then signal power is:

$P_{\mathrm{SOUT}}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{S\left(i\right)}\right)}^2}{n}=m^2\×n\frac{{\stackrel{\‾}{V}}_S^2}{n}=m^2\×{\stackrel{\‾}{V}}_S^2$

Carry out the despreading of m long symbols, then noise power is:

$P_{\mathrm{NOUT}}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{N\left(i\right)}\right)}^2}{n}=n\×\frac{E\left({\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{N\left(i\right)}\right)}^2\right)}{n}=m\×{\mathrm{\δ}}^2$

After despreading, signal to noise ratio is therefore the gain that spread spectrum brings is 10log (m).

Step 3: the pseudo-code sequence in the signal after despreading is carried out average value processing successively according to every N point, forms new signal pseudo-code sequence, the digital medium-frequency signal sequence after down-converted is G ₁=[S ₁, S ₂..., S _{m × n}], average value processing is carried out to it, carry out L point average after:

$G_2=[\frac{\underset{i=1}{\overset{L}{\mathrm{\Σ}}}{S}_i}{L},\frac{\underset{i=L+1}{\overset{2\×L}{\mathrm{\Σ}}}{S}_i}{L}...,\frac{\underset{i=(K-1)\×L+1}{\overset{K\×L}{\mathrm{\Σ}}}{S}_i}{L}]$

K×L＝m×n，K,L,m,n∈N

Introduce local pseudo-code sequence and carry out average value processing successively according to every N point equally, form new local pseudo-code sequence, the local random code sequence produced is G ₃=[P ₁, P ₂..., P _{m × n}], carry out L point average after:

$G_4=[\frac{\underset{i=1}{\overset{L}{\mathrm{\Σ}}}{P}_i}{L},\frac{\underset{i=L+1}{\overset{2\×L}{\mathrm{\Σ}}}{P}_i}{L}...,\frac{\underset{i=(K-1)\×L+1}{\overset{K\×L}{\mathrm{\Σ}}}{P}_i}{L}];$

See Fig. 2, in averaging method, can to find out that L is arranged larger for the settings of parameter L and the relation of snr loss, and snr loss is more serious, is pointwise correlation method as L=1.

Although averaging method can save amount of calculation, can cause snr loss, if received signal strength is more weak, then averaging method method cannot effectively be caught simultaneously.Irrelevant adding up is that a lot of input data are divided into many sections, is correlated with and is accumulated in every section of enforcement, and irrelevant adding up is by added together for relevant accumulation result, obtains higher signal to noise ratio.After carrying out average value processing to the pseudo-code sequence in the signal after despreading and local pseudo-code sequence, the signal signal to noise ratio after average value processing being less than to 40dB carries out irrelevant cumulative, and irrelevant cumulative gain is that relevant cumulative gain deducts irrelevant cumulative loss:

G _i(n)＝G _e(n)-L(n)

G _e＝10lg(n)

$L\left(n\right)=101g\left[\frac{1+\sqrt{1+\frac{9.2n}{{[{\mathrm{erfc}}^{-1}\left({2P}_f\right)-{\mathrm{erfc}}^{-1}\left({2P}_d\right)]}^2}}}{1+\sqrt{1+\frac{9.2}{{[{\mathrm{erfc}}^{-1}\left({2P}_f\right)-{\mathrm{erfc}}^{-1}\left({2P}_d\right)]}^2}}}\right]$

In formula: n is accumulative frequency, p _ffor false alarm probability, P _dfor inspection probability.

Step 4: new signal pseudo-code sequence and local pseudo-code sequence are carried out circular correlation process, when there is the correlation peak higher than decision threshold, just the code element correspondence in signal can be gone out specific code in local pseudo-code sequence interval, and then determine accurate location, complete and catch.

When therefore can find out in false alarm probability and check probability certain, increase irrelevant accumulative frequency n and finally can improve detection signal-to-noise ratio.For the feature of averaging method, irrelevant accumulation method is introduced existing averaging method is improved, whether the power ratio calculating cumulative rear correlation peak and noise meets is caught requirement, whether further decision proceeds irrelevant adding up is improved detection signal-to-noise ratio, finally determine specific code phase place, complete and catch.

Ensure detection probability P _d=90%, false alarm probability P _f=10-7, the detection signal-to-noise ratio after relevant is greater than 14dB, known according to formula (7), signal to noise ratio when signal arrival ground is the strongest is-32.1dB, suppose signal because the reason such as to block, SNR degradation is to-35dB, and antenna and radio frequency link gain are 2dB.Adopt the averaging method method of L=7 to carry out 4ms to catch, as shown in Figure 3, the signal to noise ratio after relevant is 7dB to correlated results, cannot reach and catch requirement.And the present invention adopts catching based on averaging method adaptive algorithm of L=4 under similarity condition, through repeatedly irrelevant cumulative after, as shown in Figure 4, the signal to noise ratio after relevant is 14.5dB to correlated results, can reach and catch requirement.

Claims (5)

1., based on a P code self-adapting catching method for averaging method, it is characterized in that: comprise the following steps:

Step one: adopt receiver reception from the signal of satellite, obtain the signal to noise ratio of satellite-signal, if signal to noise ratio is less than 40dB, then continues to receive; If signal to noise ratio is greater than 40dB, the signal received is carried out down-conversion and A/D conversion, become digital intermediate frequency sampling signal;

Step 2: introduce white noise in digital intermediate frequency sampling signal, and carry out the despreading of m long code after adopting correlator and digital mixer to remove intermediate frequency amplitude;

Step 3: the pseudo-code sequence in the signal after despreading is carried out average value processing successively according to every N point, forms new signal pseudo-code sequence, introduces local pseudo-code sequence and carries out average value processing successively according to every N point equally, form new local pseudo-code sequence;

Step 4: new signal pseudo-code sequence and local pseudo-code sequence are carried out circular correlation process, when there is the correlation peak higher than decision threshold, just the code element correspondence in signal can be gone out specific code in local pseudo-code sequence interval, and then determine accurate location, complete and catch.

2. a kind of P code self-adapting catching method based on averaging method according to claim 1, is characterized in that: in described step one, and signal to noise ratio is according to the ICD file of gps system, and the transmitting equivalent isotropically radiated power of the P coded signal of L1 frequency range is EIRP _l1-P, gps signal is sent to ground from the low-orbit satellite apart from ground R to come, and its free space attenuation is wherein λ is the wavelength of gps signal, and the decay of real atmosphere layer is about A ≈ 2dB, and therefore the signal strength signal intensity on L1 frequency range P code arrival ground is:

P _L1-P＝EIRP _L1-P-F-A；

The principal element affecting GPS noise is ambient temperature and noise of equipment, and therefore the noise temperature of unitized GPS is equivalent to 513K, and therefore the noise density of GPS is N ₀=kT, wherein k is Boltzmann constant, and T is thermodynamic temperature;

Therefore the noise power of known P coded signal is wherein B is the distance of gps signal to ground;

P coded signal arrives the signal strength signal intensity on ground, and therefore carrier-to-noise ratio is signal to noise ratio is

3. a kind of P code self-adapting catching method based on averaging method according to claim 1, is characterized in that: in described step 2, despreading process is as follows:

Be located at time period T, have digitised Intermediate Frequency sampled signal S ₁, S ₂..., S _{m × n}, after removal intermediate frequency, amplitude is V _{s (1)}, V _{s (2)}..., V _{s (m × n)}, then $\left|V_{S\left(1\right)}\right|\≈\left|V_{S\left(2\right)}\right|\≈...\≈\left|V_{S(m\×n)}\right|\≈\left|{\stackrel{\‾}{V}}_S\right|,$ Signal power is:

$\begin{array}{c}{P}_{\mathrm{SIN}}=\frac{V_{S\left(1\right)}^2+V_{S\left(2\right)}^2...+V_{S(m\×n)}^2}{m\×n}\\ =m\×n\frac{{\stackrel{\‾}{V}}_S^2}{m\×n}={\stackrel{\‾}{V}}_S^2\end{array}$

Noise N ₁, N ₂..., N _{m × n}power is:

$\begin{array}{c}{P}_{\mathrm{NIN}}=\frac{V_{n\left(1\right)}^2+V_{n\left(2\right)}^2...+V_{n(m\×n)}^2}{m\×n}\\ =E(m\×n\·\frac{V_{n\left(k\right)}^2}{m\×n})\\ =E\left(V_{n\left(k\right)}^2\right)\\ ={\mathrm{\δ}}^2\end{array}$

Before despreading, signal to noise ratio is

Carry out the despreading of m long symbols, then signal power is:

$P_{\mathrm{SOUT}}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{S\left(i\right)}\right)}^2}{n}=m^2\×n\frac{{\stackrel{\‾}{V}}_S^2}{n}=m^2\×{\stackrel{\‾}{V}}_S^2$

Carry out the despreading of m long symbols, then noise power is:

$P_{\mathrm{NOUT}}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{N\left(i\right)}\right)}^2}{n}=n\×\frac{E{\left(\left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{V}_{N\left(i\right)}\right)\right)}^2}{n}=m\×{\mathrm{\δ}}^2$

After despreading, signal to noise ratio is therefore the gain that spread spectrum brings is 10log (m).

4. a kind of P code self-adapting catching method based on averaging method according to claim 1, it is characterized in that: in described step 3, the digital medium-frequency signal sequence after down-converted is G ₁=[S ₁, S ₂..., S _{m × n}], average value processing is carried out to it, carry out L point average after:

$G_2=[\frac{\underset{i=1}{\overset{L}{\mathrm{\Σ}}}{S}_i}{L},\frac{\underset{i=L+1}{\overset{2\×L}{\mathrm{\Σ}}}{S}_i}{L}...,\frac{\underset{i=(K-1)\×L+1}{\overset{K\×L}{\mathrm{\Σ}}}{S}_i}{L}]$

K×L＝m×n，K,L,m,n∈N

The local random code sequence produced is G ₃=[P ₁, P ₂..., P _{m × n}], carry out L point average after:

$G_4=[\frac{\underset{i=1}{\overset{L}{\mathrm{\Σ}}}{P}_i}{L},\frac{\underset{i=L+1}{\overset{2\×L}{\mathrm{\Σ}}}{P}_i}{L}...,\frac{\underset{i=(K-1)\×L+1}{\overset{K\×L}{\mathrm{\Σ}}}{P}_i}{L}].$

5. a kind of P code self-adapting catching method based on averaging method according to claim 1, it is characterized in that: in described step 3, after average value processing is carried out to the pseudo-code sequence in the signal after despreading and local pseudo-code sequence, signal signal to noise ratio after average value processing being less than to 40dB carries out irrelevant cumulative, and irrelevant cumulative gain is that relevant cumulative gain deducts irrelevant cumulative loss:

G _i(n)＝G _e(n)-L(n)

G _e＝10lg(n)

$L\left(n\right)=10\lg \left[\frac{1+\sqrt{1+\frac{9.2n}{{[{\mathrm{erfc}}^{-1}\left(2P_f\right)-{\mathrm{erfc}}^{-1}\left(2P_d\right)]}^2}}}{1+\sqrt{1+\frac{9.2}{{[{\mathrm{erfc}}^{-1}\left(2P_f\right)-{\mathrm{erfc}}^{-1}\left(2P_d\right)]}^2}}}\right]$

In formula: n is accumulative frequency, p _ffor false alarm probability, P _dfor inspection probability.