CN101132191B - Baseband signal processing method for GNSS receiver - Google Patents

Abstract

This invention discloses a baseband signal processing method for GNSS receivers, particularly for common ranging-code signal for GNSS receiver, completing: the baseband signal catching, the tracking processing within frequency domain, easily for realization in large scale integrated circuit. During the signal catching step, FFT and IFFT calculation method is used for grouped calculation to simplify the time-domain related calculation. During the code-tracking, thus obtained related peak positions are found, then precision est1 mate of code phase is fulfilled by utilizing curve-fitting method. During the carrier tracking, DFT values of these points is calculated by using sliding DFT method, after the DFT point values ranges are determined based on rough carrier frequency estimate, and then proceeding estimate of carrier frequency, data demodulation by using IFFT output result.

Description

A kind of GNSS receiver baseband signal processing method

Technical field

The invention belongs to the communications field, relate to a kind of signal processing method, be specifically related to a kind of GNSS receiver baseband signal processing method.

Background technology

Adopt the navigation message of the modern satellite navigation system (GPS, the Big Dipper two generations, Galileo etc.) of spread spectrum ranging code pseudo noise code and carrier wave to be carried out twice modulation, form emission earthward continuously behind the radio wave with data mode.The user then intercepts and captures the satellite-signal in the visual field and then tries to achieve navigator fix information by receiver, and the communication infrastructure of system is spread spectrum communication, and this is the concise and to the point operation principle of modern satellite navigation system.

The signal structure of satellite navigation system is to determine by the design object of system, positioning accuracy request etc. are common.Satellite-signal comprises three kinds of signal components: PN (pseudo noise) sequence (be the PRN sequence, be called ranging code or subcarrier again), numeric data code and carrier wave.System utilizes ranging code to expand navigation message signals on frequency domain.

Ranging code comprises two kinds of common ranging code (thick catch code) and smart tellurium ranging codes.Two kinds of ranging codes that present satellite is adopted all belong to the PRN sign indicating number.Numeric data code claims navigation message again, also is a kind of Binary Zero, 1 yard, and it is different that each satellite navigation system numeric data code speed mostly is, but well below the ranging code bit rate, adopting low numeric data code speed is in order to obtain bigger spreading gain.The ranging code of satellite and numeric data code adopt BPSK or QPSK to come carrier wave is modulated.

Work such as base band signal process mainly comprises the catching of satellite-signal, follows the tracks of, decoding, purpose is to extract each observed quantity, by the decoding to satellite navigation message, finishes the extraction to the navigation message data, this is the basis that the receiver subsequent navigation is resolved work.

Figure 1 shows that the typical structure of GNSS receiver, wherein the digital baseband signal processor is the core of receiver.Antenna receives the GNSS signal, forms the digital intermediate frequency signal after the mixing afterwards, and inputs to the digital baseband signal processor through preposition amplification, RF/IF conversion, A/D conversion; This digital baseband signal processor output mainly is parameters such as pseudorange, Doppler frequency, local zone time.These parameters are calculated the needed information of user such as position, speed by the navigation signal processor of back by corresponding location algorithm.The function that digital baseband processor is mainly realized has: produce local with reference to pseudo-code; Acquiring pseudo code; Pseudo-code and carrier track; Data demodulates (navigation message demodulation); Obtain carrier doppler frequency (range rate) and carrier phase (variable in distance amount); Extract signal to noise ratio information etc. by each satellite-signal.

For the satellite-signal that adopts the spread spectrum ranging code multiple Acquisition Scheme is arranged, method for designing commonly used is to separate pseudo-code searching and searching carrier are organic, and the two is caught respectively.Can be divided into again the two process of searching for respectively: the search strategy of pseudo-code serial carrier wave serial; The search strategy that pseudo-code serial carrier wave is parallel; The search strategy of the parallel carrier wave serial of pseudo-code; The parallel search strategy of the parallel carrier wave of pseudo-code.Capture time corresponding to various search strategies has nothing in common with each other, complete walk abreast fastest, but to expend maximum software and hardware resources.This method that a plurality of 2D signals unit is searched for successively is called as sequential search.Can also adopt the FFT method in addition, can directly carry out two dimension and catch the GNSS satellite-signal.

After capturing satellite-signal, can utilize pseudo-code tracing ring and carrier tracking loop that signal is followed the tracks of, realize that receiver reproduction signal is accurately synchronous with input signal, make relevant output be in maximum rating all the time, simultaneously each observed quantity and navigation message are extracted.

Because when carrying out pseudo-code tracing, the frequency of carrier wave and phase place and inaccuracy are known, so traditional structure generally adopts leading-incoherent digital delay phase-locked loop of retarded type (DDLL).The related operation of this tracking loop has adopted two independently correlators: advanced code (sign indicating number early) correlator and the sign indicating number that lags behind (sign indicating number late) correlator.Input signal is divided into two-way: the one tunnel is relevant with leading local reference code; Another road is relevant with the local reference code that lags behind.Correlated results pass through again integration (or adding up), square, plus and minus calculation finishes phase demodulation.

Acting as of carrier tracking loop carried out demodulation to the output signal of code tracking loop, obtains the navigation message data, obtains carrier wave Doppler frequency displacement observed quantity simultaneously.At present, the method that is applied to carrier track is many, and commonly used have quadratic loop, phase-locked loop (PLL), a frequency locked loop (FLL) etc.The PLL dynamic is relatively more responsive, but can produce the observed quantity of the most accurate pseudorange rate of change.For given signal power, PLL can provide the data demodulates lower than FLL bit error rate.Therefore, PLL is applicable to the steady track pattern of GNSS receiver carrier tracking loop.When the signal initial acquisition, it is easier than realizing phase locking to realize frequency lock (FLL) under the dynamic environment, therefore method is also arranged with two kinds of loops incompatible carrier track that carries out that joins.FLL frequency discriminator algorithm commonly used has cross product automatic frequency tracking algorithm (CPAFC) etc.

Field capture tracing signal processing method when traditional, signal capture speed is limited, track band width, for the disposal ability of small-signal a little less than, be not suitable for the signal processing of carrying out under high dynamic environment or interference environment.

Summary of the invention

The present invention proposes a kind of method for processing baseband signal of GNSS receiver.Mainly be applicable to the reception of the common ranging code signal of GNSS receiver, and in frequency domain, finish baseband signal catch, follow the tracks of processing.In the code tracking process, find out a plurality of relevant peaks position after, utilize the method for curve fit to finish the accurate estimation of code phase.In the carrier track process, determine to carry out the sampled point scope of DFT computing, calculate the DFT value and carrier frequency is estimated, carry out the renewal of carrier frequency in the track loop again.When navigation message recovers, utilized the output result of IFFT in the code tracking process to finish the demodulation of data.This method all utilizes frequency domain to handle, and utilizes digital signal processing methods such as FFT, IFFT and DFT, has increased substantially relevant computational efficiency and the spectrum analysis precision of multiple branch circuit, and realization speed is faster, helps catching fast and recapturing of signal.It still has stronger robustness in complex environment, and is easy to use the large scale integrated circuit realization.

A kind of GNSS receiver baseband signal of the present invention processing method is characterized in that this method may further comprise the steps:

Step 1: antenna receives the GNSS signal, through forming digital intermediate frequency signal after preposition amplification, RF/IF conversion, the A/D conversion.

Step 2: digital baseband processor receives the digital intermediate frequency signal after the mixing, carries out signal capture, according to existing frequency domain quick capturing method, utilizes FFT and IFFT computational methods to carry out grouped calculation and simplifies time-domain related calculation.

Step 3: the signal after will catching is input to tracking module, carries out code tracking and carrier track.

Step 4: judge the upset of navigation message data according to the carrier phase difference of modulation navigation message.According to the output information of track loop,, judge the variation of navigation message data b it by judging the real part situation of change of code tracking loop IFFT computing at peak-peak place in two adjacent code phase place update cycles.Thereby carry out the demodulation of navigation message data.Simultaneously, the result after the demodulation is fed back to tracking module.

In the step 3, when carrying out the ranging code tracking, at first utilize FFT and IFFT computing to find out a plurality of relevant peaks position, utilize least square method to carry out the triangle match then, finish the accurate estimation of theoretical peak-peak position according to the position of triangular apex, and then finish the accurate estimation of code phase.

In the step 3, when carrying out carrier track, at first adopt multiple spot slip DFT algorithm, utilize the Fourier transform result of calculation of carrier signal sampled point in the last window, go to calculate the Fourier transform value of sampled point in this window, utilize interpolation DFT algorithm to carry out spectrum analysis then, finish the accurate estimation of carrier frequency.

In the step 3, when adopting the Fourier transform value of sampled point in this window of multiple spot slip DFT algorithm computation, at first, obtain frequency search step-length and rough estimating carrier frequencies value by Capture Circle, determine to need to calculate the scope of DFT value point then, utilize the DFT value of this part point of multiple spot slip DFT algorithm computation again.

The advantage of a kind of GNSS receiver baseband signal of the present invention processing method is:

(1) in the description of the inventive method, the operation of catching and follow the tracks of of GNSS signal all utilizes frequency domain to handle, increased substantially the relevant computational efficiency of multiple branch circuit, help catching fast and recapturing of signal, be very suitable for small-signal environment, interference environment and than signal capture under the high dynamic environment and tracking.

(2) in the description of the inventive method, GNSS signal capture and code tracking part all adopt FFT and IFFT to realize.Can adopt the pipeline module cascade structure when FFT and IFFT design, arithmetic speed is faster; Also can adopt ripe IP kernel to finish integrated circuit (IC) design, be easy to adopt large scale integrated circuit to realize.

When (3) application method steps of the present invention is carried out carrier track, frequency search step-length when at first basis is caught and the rough estimating carrier frequencies value that obtains, dwindle the scope of calculating DFT value point, required Fourier transform range value during applications exploiting slip DFT algorithm computation Frequency Estimation then, so just greatly reduced computation complexity, helped integrated circuit and realize.

Description of drawings

Fig. 1 is the system block diagram of existing GNSS receiver;

Fig. 2 is existing GNSS receiver frequency domain trapping module structure chart.

Fig. 3 is a kind of GNSS receiver baseband signal of the present invention process flow schematic diagram;

Fig. 4 is the intermediate-freuqncy signal ranging code tracking schematic flow sheet of a kind of GNSS receiver baseband signal of the present invention processing method;

Fig. 5 is the intermediate-freuqncy signal carrier wave tracing method schematic flow sheet of a kind of GNSS receiver baseband signal of the present invention processing method;

Fig. 6 is the existing multiple spot slip DFT method schematic diagram of a kind of GNSS receiver baseband signal of the present invention processing method;

Fig. 7 is the ranging code phase capturing of a kind of GNSS receiver baseband signal of the present invention processing method figure as a result;

Fig. 8 estimates to show figure as a result for ranging code in the code tracking process of a kind of GNSS receiver baseband signal of the present invention processing method;

Fig. 9 is the figure as a result of the carrier track in 20 navigation message cycles of a kind of GNSS receiver baseband signal of the present invention processing method;

Figure 10 is a transmitting terminal navigation message oscillogram for a kind of GNSS receiver baseband signal of the present invention processing method;

Figure 11 is a receiving terminal navigation message oscillogram for a kind of GNSS receiver baseband signal of the present invention processing method.

Embodiment

The present invention is described in further detail below in conjunction with accompanying drawing.

This method mainly is applicable to the reception of the common ranging code signal of GNSS receiver, and in frequency domain, finish baseband signal catch, follow the tracks of processing.In the acquisition procedure, utilize FFT and IFFT computational methods to carry out grouped calculation and simplify time-domain related calculation.In the code tracking process, at first utilize FFT and IFFT computing to find out a plurality of relevant peaks position, utilize the method for curve fit to finish the accurate estimation of code phase then.In the carrier track process, the rough estimating carrier frequencies value that obtains when at first basis is caught determines to calculate the scope of DFT value point, and utilize slip DFT method to calculate the DFT value of this part point, utilize interpolation DFT method that carrier frequency is estimated then, the renewal of carrier frequency in the line trace loop of going forward side by side.When navigation message recovers, utilized the output result of IFFT in the code tracking process to finish the demodulation of data.

As shown in Figure 3, this method may further comprise the steps:

Step 1: as shown in Figure 1, antenna receives the GNSS signal, forms the digital intermediate frequency signal after the mixing afterwards through preamplifier, RF/IF converter, A/D converter.

Output the digitlization intermediate-freuqncy signal as shown in the formula:

s(n)＝AD(nt _s-τ)C(nt _s-τ)cos[2π(f _IF+f _d)nt _s+φ ₀]+N′(n) (1)

Intermediate-freuqncy signal output valve when wherein, s (n) is n sampled point.A, f _IFAnd φ ₀Amplitude, frequency and the initial phase of representing carrier wave respectively, f _dBe Doppler frequency deviation, t _sBe the sampling interval, τ is a time delay.D (nt _s) and C (nt _s) represent navigation message data (D sign indicating number) and common ranging code, N ' expression noise signal respectively.

Step 2: digital baseband processor receives the digital intermediate frequency signal after the mixing, carries out signal capture, according to existing frequency domain quick capturing method, in Capture Circle, utilizes FFT and IFFT computational methods to carry out grouped calculation and simplifies time-domain related calculation.

If the sample frequency of A/D converter is a definite value, i.e. sampling interval t _sBe constant, then have: Z (m, f _d)=I+jQ=IFFT{FFT{s (n) exp[-j2 π (f _IF+ f _d) nt _s] FFT ^*[C (m)] } (2)

Wherein, n=(i-1) M+m represents m the sampled point in i ranging code cycle, and m ∈ [1,2 ..., M], M is the sampling number of each ranging code in the cycle.C (m) represents the local C/A sign indicating number that produces, ^*Conjugation is got in representative.Z (m, f _d) expression input signal and the local correlated results that reappears signal.Order

Signal capture is exactly to carry out two-dimensional search in code phase and Doppler frequency and detects.| Z (m, f _d) | if surpass threshold value V _Th, acquisition success then, otherwise catch failure.Certainly also can pass through here to judge | Z (m, f _d) | ²Realize.

Utilize FFT and IFFT computational methods to carry out grouped calculation and carry out above-mentioned computing.As shown in Figure 2, obtain complex signal: x (n)=I (n)+jQ (n) after the I of the quadrature that input signal and local carrier NCO produce respectively, the Q two branch road carrier multiplication, x (n) enters the FFT converter, carries out the FFT conversion.The sign indicating number sequence that simultaneously local common ranging code generator generates is through FFT converter and conjugation processor, the FFT that successively carries out result and x (n) after FFT conversion and conjugation are handled multiplies each other in multiplier through conversion result afterwards, through obtaining the correlated results of time domain after the IFFT conversion, promptly obtain correlation peak again corresponding to each code phase.These correlation peaks are through square converter, and its square result's peak value place is the code phase of input signal, and is entered in the logic controller.The carrier frequency that scan this moment is capture frequency.

Step 3: the signal after will catching is input to tracking module, carries out code tracking and carrier track.

Code tracking method flow process described in the invention as shown in Figure 4.The signal flow of the data relevant portion here is similar to acquisition procedure mentioned above.Difference is that the frequency renewal of carrier wave NCO is subjected to the resulting estimated frequency control of carrier tracking loop, and the carrier phase in simultaneously each update cycle should be continuous, and its reason will be described in detail in the analysis of demodulation module.

After carrying out related operation, obtain a plurality of correlation peaks corresponding to different code phases by the IFFT converter, the number of code phase is relevant with sample frequency here, and sample frequency is high more, and temporal resolution is high more.But be subjected to the restriction of data processor ability, can not infinitely improve sample frequency, in order further to improve the code phase estimated accuracy under limited sample rate, this method adopts the method for curve fit to come accurate estimated code phase place, with a kind of method of showing the triangle least square fitting of row.At first by the FFT/IFFT module in groups computing obtain correlation peak corresponding to each code phase.Correlated results input threshold decision device, if maximum correlation peak reaches decision threshold, tracing process is proceeded, otherwise the tracing process end is caught again.If tracing process continues, each correlation peak relatively in peak comparator then | the size of Z|, search and keep peak-peak and near peak point thereof, and peak information is input to the navigation message demodulation module.Next the comparative result in the peak comparator is input to the curve fit device, the curve fit device is that separation is divided into two groups with the peak point that keeps with the maximal peak point, use least square method every group of data point fitted to two straight lines, the abscissa of the intersection point of these two straight lines is the accurate estimated value of code phase.At last, common ranging code generator produces common ranging code according to the code phase estimated value that obtains, and feeds back to track loop.

To keep 9 peak point z ₁(x ₁, y ₁), z ₂(x ₂, y ₂) ..., z ₉(x ₉, y ₉) be example, concrete mathematic(al) representation is described below:

If the peak point that keeps in the peak comparator is:

{z ₁(x ₁，y ₁)，z ₂(x ₂，y ₂)，…，z ₉(x ₉，y ₉)}，(z ₁≤z ₂≤z ₃≤z ₄≤z ₅，z ₉≤z ₈≤z ₇≤z ₆≤z ₅) (3)

Two straight lines for the treatment of match are in the curve fit device:

$\left\{\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="l"\; filenames="H2007101758721E00051.png,H2007101758721E00052.png"\; state="\{\{state\}\}">l</figure-callout>}}_1=a_1+a_{2}x\\ l_2=b_1+b_{2}x\end{array}\right.---\left(4\right)$

Linear empirical equation by least square method can get:

$\left\{\begin{array}{c}{a}_2=\frac{\underset{i=1}{\overset{5}{\mathrm{\&Sigma;}}}\left(x_i{y}_i\right)-\frac{1}{5}\left(\underset{i=1}{\overset{5}{\mathrm{\&Sigma;}}}{x}_i\right)\left(\underset{i=1}{\overset{5}{\mathrm{\&Sigma;}}}{y}_i\right)}{\underset{i=1}{\overset{5}{\mathrm{\&Sigma;}}}{x}_i^2-\frac{1}{5}{\left(\underset{i=1}{\overset{5}{\mathrm{\&Sigma;}}}{x}_i\right)}^2}\\ a_1=\frac{1}{5}\underset{i=1}{\overset{5}{\mathrm{\&Sigma;}}}{y}_i-\frac{1}{5}{a}_2\underset{i=1}{\overset{5}{\mathrm{\&Sigma;}}}{x}_i\end{array}\right.---\left(5\right)$

$\left\{\begin{array}{c}{b}_2=\frac{\underset{i=5}{\overset{9}{\mathrm{\&Sigma;}}}\left(x_i{y}_i\right)-\frac{1}{5}\left(\underset{i=5}{\overset{9}{\mathrm{\&Sigma;}}}{x}_i\right)\left(\underset{i=5}{\overset{9}{\mathrm{\&Sigma;}}}{y}_i\right)}{\underset{i=5}{\overset{9}{\mathrm{\&Sigma;}}}{x}_i^2-\frac{1}{5}{\left(\underset{i=5}{\overset{9}{\mathrm{\&Sigma;}}}{x}_i\right)}^2}\\ {\mathrm{<figure-callout\; id="1"\; label="b"\; filenames="H2007101758721E00051.png,H2007101758721E00052.png"\; state="\{\{state\}\}">b</figure-callout>}}_1=\frac{1}{5}\underset{i=5}{\overset{9}{\mathrm{\&Sigma;}}}{y}_i-\frac{1}{5}{b}_2\underset{i=5}{\overset{9}{\mathrm{\&Sigma;}}}{x}_i\end{array}\right.---\left(6\right)$

With (5), (6) substitution (4), the intersection point abscissa that can get two straight lines is:

$x_0=\frac{b_1-a_1}{a_2-b_2}---\left(7\right)$

x ₀Value be the accurate estimated value of code phase.

Carrier wave tracing method flow process described in the invention is shown in Figure 5.

To feed back to navigation message (D sign indicating number) solving result of the demodulation gained of track loop by demodulation module, and import multiplier with the digital intermediate frequency signal of input and multiply each other, thereby remove the modulation of D sign indicating number.Common ranging code generator local common ranging code that produces and the digital intermediate frequency signal of removing after the D sign indicating number is modulated are multiplied each other through multiplier, finish despreading.Expression formula by s (n) can get, go to modulate and despreading after signal be:

s′(n)＝Acos[2π(f _IF+f _d)nt _s+φ ₀]+N′(n) (8)

If common ranging code Phase Tracking is accurate, the signal that then need carry out Frequency Estimation in the carrier tracking loop is approximately the simple signal s ' of plus noise.

Data segment to the simple signal of gained carries out the DFT computing through the DFT converter, analyzes its frequency spectrum by frequency estimator again, and then finishes Frequency Estimation.When carrying out estimating carrier frequencies in track loop, if N point Fourier transform value is calculated by FFT in the sliding window of DFT converter, calculated load is excessive, is difficult to real-time processing.Calculate Fourier's changing value that N is ordered even use the recurrence formula of multiple spot slip DFT algorithm, amount of calculation is still huge.By interpolation DFT algorithm as can be known, when carrying out Frequency Estimation, only need a maximum spectral line and near a spectral line value thereof, therefore can consider in the DFT converter, dwindle the scope of calculating DFT value point by catching the frequency information that obtains, and then the recurrence formula of application multiple spot slip DFT algorithm calculates Fourier's changing value that N is ordered.

If the frequency that is obtained by Capture Circle is f _Acq, the step value of Capture Circle medium frequency search is f _Step, then the carrier doppler frequency range should be:

f∈[f _acq-f _step，f _acq+f _step] (9)

By the DFT basic theories as can be known, when carrying out N point DFT computing, the index range of corresponding (9) formula is:

$N_{\mathrm{range}}\&Element;[\mathrm{int}\left(\frac{(f_{\mathrm{acq}}-f_{\mathrm{step}})}{f_s}N\right),\mathrm{int}\left(\frac{(f_{\mathrm{acq}}+f_{\mathrm{step}})}{f_s}N\right)]---\left(10\right)$

Int () expression rounds in the formula.

After having determined the data point scope, the DFT converter can be according to the DFT value of the multiple spot slip DFT algorithm computation each point in the existing document.As shown in Figure 6, suppose the P that is spaced apart between two windows, sliding window is long to be N.Multiple spot slip DFT algorithm purpose is according to recurrence relation and calculates DFT value in every window.If i constantly the sample variance value of time-domain signal waveform be s ' (n), Fourier transform is S _i(k), then be that the N point Fourier transform of original position is with i+p:

$S_{i+p}\left(k\right)=\underset{n=p}{\overset{N-1-p}{\mathrm{\&Sigma;}}}{s}^{\&prime;}\left(n\right)e^{-j\frac{2\mathrm{\&pi;}}{N}(n-p)}=\{S_i\left(k\right)+\underset{n=0}{\overset{p-1}{\mathrm{\&Sigma;}}}[s^{\&prime;}(n+N)-s^{\&prime;}\left(n\right)\left]e^{-j\frac{2\mathrm{\&pi;}}{N}\mathrm{nk}}\right\}{e}^{j\frac{2\mathrm{\&pi;}}{N}\mathrm{pk}}---\left(11\right)$

Once to calculate the Fourier transform value of being had a few different with FFT, (11) formula of application can directly be tried to achieve the Fourier transform value of single-point, the Fourier transform value that this just can directly calculate required point in (10) formula has so greatly reduced the computation complexity of DFT converter.

The purpose of carrier track is the accurate estimation that obtains carrier frequency, and Fig. 5 shows after obtaining the DFT value of required point, can finish Frequency Estimation by frequency estimator according to the DFT interpolation algorithm in the existing document.

Observe formula (8) signal, in the DFT converter to s ' _iAfter carrying out DFT, the amplitude item of frequency spectrum is:

$S\left(k\right)=\frac{A\sin \left[\mathrm{\&pi;}(k-f_{\mathrm{IF}}T)\right]}{\sin [\mathrm{\&pi;}(k-f_{\mathrm{IF}}T)/N]}---\left(12\right)$

T=Nt in the formula _s, the estimated value expression formula that then provides carrier frequency is:

$f=f_{\mathrm{IF}}+f_d=\frac{1}{{\mathrm{Nt}}_s}[{\mathrm{<figure-callout\; id="0"\; label="k"\; filenames="H2007101758721E00051.png,H2007101758721E00052.png"\; state="\{\{state\}\}">k</figure-callout>}}_0+\frac{r\left|S({\mathrm{<figure-callout\; id="0"\; label="k"\; filenames="H2007101758721E00051.png,H2007101758721E00052.png"\; state="\{\{state\}\}">k</figure-callout>}}_0+r)\right|}{\left|S\left(k_0\right)\right|+\left|S({\mathrm{<figure-callout\; id="0"\; label="k"\; filenames="H2007101758721E00051.png,H2007101758721E00052.png"\; state="\{\{state\}\}">k</figure-callout>}}_0+r)\right|}]---\left(13\right)$

Wherein, N is that DFT counts k ₀It is the position of s ' DFT maximum spectral line (n).R=± 1, when | S (k ₀+ 1) |≤| S (k ₀-1) | the time, r=-1; When | S (k ₀+ 1) | 〉=| S (k ₀-1) | the time, r=1.

Therefore, in frequency estimator, DFT value substitution (13) formula of the data point that will be obtained by the DFT converter just can obtain the estimated value of carrier frequency.The carrier frequency valuation that obtains is input to the code tracking loop, if the maximal correlation peak value of code tracking loop reaches decision threshold, then chooses data segment again, tracing process is proceeded, otherwise tracing process finishes.

The frequency resolution of DFT is 1/T, T=Nt _sBe the accumulated time of signal, when Δ t was definite value, the frequency resolution of DFT was subjected to the restriction of sliding window length N.If frequency resolution is low excessively, even adopt the method for interpolation DFT also to be difficult to improve the estimated accuracy of frequency.So the length that should accumulate certain carrier only data segment is used for spectrum analysis.When the GNSS signal was extremely faint, sliding window length should suitably increase, the cumulative data segment length when promptly increasing each carrier frequency and upgrading.

Step 4: judge the upset of navigation message data according to the carrier phase difference of modulation navigation message.In the navigation message demodulation module, output information according to track loop, by judging the real part situation of change of code tracking loop IFFT converter operation result at peak-peak place in two adjacent code phase place update cycles, judge the variation of navigation message data b it.Thereby carry out the demodulation of navigation message data.Simultaneously, the result after the demodulation is fed back to track loop.

Described in the invention is the frequency domain method for processing baseband signal, so the demodulation mode of data is different from the demodulation method in traditional GNSS receiver.The frequency that frequency domain carrier track process is only followed the tracks of carrier wave, thus can not accurately estimate carrier phase, so the present invention judges the upset of navigation message data according to the carrier phase difference of modulation navigation message, and then finish demodulation.In the code tracking loop, the real part of the IFFT converter operation result at peak-peak place has reflected the bit height of navigation data, therefore the navigation message demodulation module is judged the variation of navigation message data b it by judging the real part situation of change of the IFFT converter operation result at peak-peak place in two adjacent update cycles.Because be the upset of judging demodulating data by the variation of judging phase place, so, should be continuous in the carrier phase of code tracking loop in each update cycle.This demodulation mode may make data bits the situation of opposite in sign occur.The ambiguity of this demodulating data can be resolved in frame synchronizing process.

Got by (2) formula, the signal terms in i accumulation result of I branch road accumulator is summarized as:

$I_i\&ap;\frac{A_r{A}_i{d}_i}{2}{N}_{i}R\left(\mathrm{\&Delta;}{\mathrm{\&tau;}}_i\right)\mathrm{Sa}\left(\mathrm{\&pi;\&Delta;}{f}_i{T}_i\right)\cos \left(\mathrm{\&Delta;}{\mathrm{\&phi;}}_i\right)---\left(14\right)$

A in the formula _rThe amplitude of the local reference carrier that produces for carrier generator, T ' _iBe the time span that adds up by the end of i moment accumulator, N _iBe the sampling number that adds up altogether constantly by the end of i, T _i=N _it _s, Δ f _iBe the estimated frequency error of i moment reference carrier, Δ φ _i=φ ₀-φ ₀Be the local reference carrier of i carrier generator generation constantly and the phase difference of received signal, Δ τ _iBe i moment ranging code phase estimation error.R () is that correlation function, Sa () are sampling function.When the code phase time-delay is very accurate with estimating carrier frequencies,

I _i≈Kd _icos(Δφ _i) (15)

Wherein, K is a proportionality coefficient.

And the phase place of the local reference carrier that produces of carrier generator is continuous, Δ φ _iThe ≈ Δ _I-1So,,

I _iI _i-1≈K ²cos ²(Δφ _i)d _id _i-1∝d _id _i-1 (16)

Therefore, the navigation message demodulation module among the present invention is judged the variation of navigation message data b it by judging the real part situation of change of code tracking loop IFFT converter operation result at peak-peak place in two adjacent update cycles.This demodulation mode may make data bits the situation of opposite in sign occur.The ambiguity of this demodulating data can be resolved in frame synchronizing process.

Computer simulation program produces the L1 signal of GPS, is transformed to by RF on the intermediate frequency of 1.25MHz, and the sample frequency of receiving terminal is set to 5MHz, so in the C/A sign indicating number cycle of each GPS 5000 sampled points is arranged.It is 504 sampled points that the chip time-delay is set, and Doppler frequency deviation is 1907Hz, and the signal to noise ratio that arrives the intermediate-freuqncy signal of receiver is-10dB.

Shown in Figure 7, be the ranging code phase capturing result schematic diagram that receives in this method step 2, as can be seen from the figure, the position of maximum correlation peak is near 500, and then the code phase delay valuation is near the actual delay of 504 chips.By shown in Figure 8, for keeping peak-peak and near 8 correlation peak points thereof in this method step 3 during code tracking, obtain the accurate estimated value of code phase by the triangle least-squares estimation, publish picture as seen, estimate to such an extent that the abscissa of triangular apex is 504 chips, consistent with the emulation set point.Fig. 9 shows the carrier track process in 20 navigation message data segments.As shown in the figure, catching the carrier frequency that obtains is 1252kHz, after the carrier track processing, converge near the actual carrier doppler frequency 1251.907kHz, and it is very fast to the convergence rate of tracking frequency by capture frequency, the fluctuating range of the frequency estimation in each update cycle is less, can accurate tracking.

Figure 10, Figure 11 have contrasted the transmitting terminal navigation message and receiving terminal is separated the text result who is transferred to, demodulation output waveform as can be seen and the transmission waveform basically identical of making a start, can obtain correct demodulation result, verify the correctness of utilizing the inventive method to carry out demodulation.

This area professional can finish the common ranging code signal processing of base band of GNSS receiver according to description of the invention, and application software or hardware platform, particularly large scale integrated circuit are realized method of the present invention.Than the time domain method for processing baseband signal of traditional GNSS receiver, the present invention is at acquisition speed, and the dynamic range of tracking bandwidth and receiver all has significantly to be improved.

Claims (1)

1. GNSS receiver baseband signal processing method is characterized in that this method may further comprise the steps:

Step 1: antenna receives the GNSS signal, through forming digital intermediate frequency signal after preposition amplification, RF/IF conversion, the A/D conversion;

Step 2: digital baseband processor receives the digital intermediate frequency signal after the mixing, carries out signal capture, according to the frequency domain quick capturing method based on FFT, utilizes FFT and IFFT computational methods to carry out grouped calculation and simplifies time-domain related calculation;

Step 3: the signal after will catching is input to tracking module, carries out code tracking and carrier track;

When carrying out the ranging code tracking, at first utilize FFT and IFFT computing to find out a plurality of relevant peaks position, utilize least square method to carry out the triangle match then, finish the accurate estimation of theoretical peak-peak position, and then finish the accurate estimation of code phase according to the position of triangular apex; When carrying out carrier track, at first adopt multiple spot slip DFT algorithm, utilize the Fourier transform result of calculation of signal sampling point in the last window, go to calculate the Fourier transform value of sampled point in this window, utilize interpolation DFT algorithm to carry out spectrum analysis then, finish the accurate estimation of carrier frequency; When adopting the Fourier transform value of sampled point in this window of multiple spot slip DFT algorithm computation, at first, obtain frequency search step-length and rough estimating carrier frequencies value by Capture Circle, determine to need to calculate the scope of DFT value point then, utilize the DFT value of this part point of multiple spot slip DFT algorithm computation again;

Step 4: judge the upset of navigation message data according to the carrier phase difference of modulation navigation message; According to the output information of track loop,, judge the variation of navigation message data b it by judging the real part situation of change of code tracking loop IFFT computing at peak-peak place in two adjacent code phase place update cycles; Thereby carry out the demodulation of navigation message data; Simultaneously, the result after the demodulation is fed back to tracking module.

Images