CN104267413A - Lifting wavelet double-threshold denoising algorithm based on signal strength self-adaptive tabu search - Google Patents

Abstract

The invention relates to a lifting wavelet double-threshold denoising algorithm based on signal strength self-adaptive tabu search. The algorithm is mainly characterized in that weak global navigation satellite system signals are captured through a differential coherent accumulation algorithm; the weak global navigation satellite system signals are analyzed through decomposition of lifting wavelets; a signal strength self-adaptive tabu search algorithm is used for optimizing selection of the double thresholds in denoising of the lifting wavelets; the captured weak global navigation satellite system signals are subjected to post-processing through the optimal double thresholds. The method is reasonable in design, the weak GNSS signals are captured through the differential coherent accumulation algorithm, the captured weak GNSS signals are subjected to post-processing through the lifting wavelet double-threshold denoising method optimized by signal strength self-adaptive tabu search, low calculation time cost and high accuracy are achieved, and the signal to noise ratio of signal output is increased.

Description

Based on the Lifting Wavelet dual threshold value denoise algorithm of signal intensity self-adaptation tabu search

Technical field

The invention belongs to Weak Signal Processing technical field, especially a kind of Lifting Wavelet dual threshold value denoise algorithm based on signal intensity self-adaptation tabu search.

Background technology

In recent years, in order to obtain more accurate navigation information, such as height, speed and position, GLONASS (Global Navigation Satellite System) (GNSS) technology is widely used in military field, and all plays very important role at vehicle mounted guidance, personal hand-held terminal and multiple civil area.But under relatively rugged environment, such as rural area, forest, valley and indoor environment, because the reason of propagation loss and multipath fading makes the signal to noise ratio (S/N ratio) of the signal received very low, with open environment facies ratio, signal can produce the decay of 15dB-20dB.In addition, the power of Navsat feeble signal is far below the scope of application of general operation of receiver power.Therefore, traditional GNSS signal receiver can not catch gps signal under the environment of high request, says nothing of and follows the tracks of and located.In sum, need to pay close attention to catching of Weak GNSS signal more, and receiver catch amplitude lower and be embedded in and require enough sensitive in the feeble signal in noise, meanwhile, the lifting of accumulated time, higher signal to noise ratio (S/N ratio) and sensitivity is all the direction needing research and pay close attention to.

The method of Weak absorption is different from the detection of general signal, and traditional signal detecting method generally pays close attention to the reduction of noise and the raising of signal to noise ratio (S/N ratio).But when processing feeble signal, these conventional methods often lost efficacy, because conventional method can not improve the signal to noise ratio (S/N ratio) of the feeble signal of being corroded by noise effectively, and in the process of amplifying at traditional signal, feeble signal is torsional deformation very easily.In addition, the method (such as low frequency phase sensitive filter) that some traditional feeble signals obtain needs the specific information of some relevant signals, such as phase place, frequency etc.Therefore, need invent some new methods obtaining feeble signal and use it for solving practical problems.

Summary of the invention

The object of the invention is to overcome the deficiencies in the prior art, provide a kind of reasonable in design, precision is high and the Lifting Wavelet dual threshold value denoise algorithm based on signal intensity self-adaptation tabu search that time cost is low.

The present invention solves existing technical matters and takes following technical scheme to realize:

Based on a Lifting Wavelet dual threshold value denoise algorithm for signal intensity self-adaptation tabu search, comprise the following steps:

Step 1, differential coherence Cumulate algorithm is utilized to catch faint received global navigation satellite system signal;

Step 2, the decomposition of application Lifting Wavelet are analyzed faint received global navigation satellite system signal;

Step 3, application signal intensity self-adaptation tabu search algorithm are selected to be optimized to the dual threshold in Lifting Wavelet denoising;

Step 4, optimized dual threshold is utilized to carry out aftertreatment to the faint received global navigation satellite system signal of catching.

And the implementation method of described step 1 is: obtain feeble signal by differential coherence accumulation, then accumulation results is inputted Wavelet Denoising Method wave filter to find its code phase and Doppler shift, thus complete acquisition procedure.

And the concrete treatment step of described step 2 is:

(1) decomposition step, is decomposed into two subdivisions by original signal sequence Sj according to parity---even _j-1and odd _j-1:

(odd _j-1,even _j-1)＝Split(S _j)

(2) prediction steps, predict by following predicting machine:

d _j-1＝odd _j-1-P(even _j-1)

Wherein, P (even _j-1) be predict odd number value by even item, specifying information d _j-1pass through odd _j-1with prediction P (even _j-1) between error represent;

(3) step of updating, upgrades average by following formula:

S _j-1＝even _j-1+U(d _j-1)。

And the concrete treatment step of described step 3 is:

(1) upper threshold and lower limit---T ' is supposed _minwith T ' _max

(2) [T is set to by between the original area of threshold optimization _min, T _max], by upper limit T _maxbe set to T _max=3 σ, wherein σ is estimated as: σ ≈ Mid/0.6745, in formula, Mid be Decomposition order minimum time wavelet coefficient by the median after numerical ordering, lower limit T _minit is the absolute value of minimum coefficient than 0 in jth layer wavelet decomposition;

(3) at T ' _minwith T ' _maxbetween produce the subset of a candidate solution, then according to following fitness function, it is sorted:

$S=\mathrm{\α}\×\lg \left[\frac{\underset{n}{\mathrm{\Σ}}{f}_d^2\left(n\right)}{\underset{n}{\mathrm{\Σ}}{[f_d\left(n\right)-f\left(n\right)]}^2}\right]$

Wherein, α is the auto-adaptive parameter chosen according to the strength information of faint received global navigation satellite system signal, and fd is the reconstruction signal after denoising, and f is reference signal;

$\mathrm{\α}=\sqrt{{\mathrm{\σ}}_{\mathrm{\ω}}^2\left(t\right)+{\mathrm{\σ}}_n^2}$

Wherein, σ _ωt root-mean-square value that () is signal, for system average noise power;

(4) search for by signal intensity self-adaptation TS algorithm, carry out correction obtain feasible solution to current solution, if move to its contiguous solution be not better than current optimum solution, then adjacent solution will be accepted; Then introduce taboo list and avoid circulation;

(5) step (3) is returned until meet stopping criterion;

(6) following formulae discovery optimal threshold is utilized:

T′ _max＝y _1j×10 ⁰+y _2j×10 ^-1+y _3j×10 ^-2+y _4j×10 ^-3

T′ _min＝y _5j×10 ⁰+y _6j×10 ^-1+y _7j×10 ^-2+y _8j×10 ^-3。

And the concrete disposal route of described step 4 is: optimal double threshold value [T ' _min, T ' _max] determined after, a kth wavelet coefficient W in jth layer wavelet decomposition _j,kby following formula process:

$W_{j,k}^{\&prime;}=\left\{\begin{array}{c}0,\left|W_{j,k}\right|<T_{\min }^{\&prime;};\\ \frac{T_{\max }^{\&prime;}}{T_{\max }^{\&prime;}-T_{\min }^{\&prime;}}\left(\right|W_{j,k}|-T_{\min }^{\&prime;})\mathrm{sgn}\left(W_{j,k}\right),T_{\min }^{\&prime;}\&le;\left|W_{j,k}\right|\&le;T_{\max }^{\&prime;}.\\ W_{j,k},\left|W_{j,k}\right|>T_{\max }^{\&prime;}.\end{array}\right.$

Advantage of the present invention and good effect are:

1, the present invention is in the process of catching feeble signal, adopts differential coherence accumulation to obtain feeble signal, can reduce Squared Error Loss, reduce the amplification of noise, simultaneously larger to the raising of signal to noise ratio (S/N ratio).

2, feeble signal is caught the combination with Wavelet Denoising Method by the present invention, signal to noise ratio (S/N ratio) can be brought up to some specific degree, utilizes Wavelet noise-eliminating method to improve output signal-to-noise ratio.

3, the present invention adopts tabu search algorithm, can the optimization feature of simulating human memory function, and avoid roundabout search by local neighbor seaching mechanism and corresponding taboo criterion, some outstanding taboo states can be discharged to ensure the diversity of efficient search, the defect falling into the easy precocious of genetic algorithm can be suppressed, reach final optimization.

4, the present invention is reasonable in design, Weak GNSS signal is caught by differential coherence Cumulate algorithm, the Lifting Wavelet dual threshold value denoising method of application signal intensity self-adaptation tabu search optimization carries out aftertreatment to the Weak GNSS signal obtained, achieve cost and higher accuracy lower computing time, improve the signal to noise ratio (S/N ratio) that signal exports.

Accompanying drawing explanation

Fig. 1 is signal intensity self-adaptation tabu search process schematic of the present invention;

Fig. 2 is relation comparison diagram between the different wavelet basis wavelet decomposition number of plies and output signal-to-noise ratio;

Fig. 3 is the denoising effect comparison diagram of the present invention compared with original signal;

Fig. 4 is the output signal-to-noise ratio comparison diagram under the present invention and other three kinds of algorithms before and after noise reduction.

Embodiment

Below in conjunction with accompanying drawing, the embodiment of the present invention is further described.

Based on a Lifting Wavelet dual threshold value denoise algorithm for signal intensity self-adaptation tabu search, comprise the following steps:

Step 1, differential coherence Cumulate algorithm is utilized to catch faint received global navigation satellite system signal.

In the presence of a harsh environment, such as genuine and valley, prior art proposed different technological means to overcome a difficult problem for faint GLONASS (Global Navigation Satellite System) (GNSS) signal capture.In order to improve the sensitivity of receiver, and obtain Weak GNSS signal more accurately, traditional algorithm comprises coherent accumulation (COH), non-coherent accumulation (NCOH) and differential coherence accumulation (DFC).In Weak GNSS signal is caught, COH is directly a kind of and effective method, can improve the signal to noise ratio (S/N ratio) of the signal detected.But the navigation data of 50bps and frequency shift (FS) will double the time limiting coherent accumulation.Adopt the performance mean height 1.5dB of the Performance Ratio non-coherent accumulation (NCOH) of differential coherence accumulation (DFC) algorithm, DFC algorithm can obtain maximum processing gain.

Therefore, this step utilizes differential coherence to accumulate (DFC) and catches faint received global navigation satellite system signal.In the process of catching feeble signal, first implement differential coherence accumulation and obtain feeble signal, then accumulation results is inputted Wavelet Denoising Method wave filter to find its code phase and Doppler shift, caught subsequently.Differential coherence accumulation can reduce Squared Error Loss, and it is at adjacent coherent integrator intercropping conjugation multiplication.Therefore, the amplification of this method to noise is less, and larger to the raising of signal to noise ratio (S/N ratio).

The digital intermediate frequency signal of global navigation satellite system GNSS can be represented as:

$\begin{array}{c}{S}_{\mathrm{IF}}\left(t\right)=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}2P_i{D}_i(t-{\mathrm{\&tau;}}_i)C_i(t-{\mathrm{\&tau;}}_i)\&CenterDot;\mathrm{expj}(2\mathrm{\&pi;}(f_{\mathrm{IF}}+f_{\mathrm{di}})t+{\mathrm{\&phi;}}_i)\\ +\mathrm{\&xi;}\left(t\right)\end{array}$

Wherein, P _isignal amplitude, D _inavigation data, C _iit is the C/A spreading code when moment k with time delay τ.The correlativity had between the satellite-signal of same pseudo-random noise PRN and local C/A code can be expressed as:

$\left\{\begin{array}{c}{s}_I\left(t\right)=S_{\mathrm{IF}}\left(t\right)\hat{C}(t-\widehat{\mathrm{\&tau;}})\cos ({\mathrm{\&omega;}}_{\mathrm{IF}}t+{\widehat{\mathrm{\&omega;}}}_{d}t)\\ s_Q\left(t\right)=S_{\mathrm{IF}}\left(t\right)\hat{C}(t-\widehat{\mathrm{\&tau;}})\sin ({\mathrm{\&omega;}}_{\mathrm{IF}}t+{\widehat{\mathrm{\&omega;}}}_{d}t)\end{array}\right.$

Wherein, s _it () is in-phase component part, and s _qt () is quadrature component part.

If correlation time is N number of cycle, then correlated results is:

$S_I=\frac{1}{2}{N}_S\mathrm{ADR}\left(\mathrm{\&Delta;\&tau;}\right)\sin c\left(\mathrm{\&Delta;}{\mathrm{\&omega;}}_D{T}_C\right)\cos (\mathrm{\&Delta;}{\mathrm{\&omega;}}_D{T}_C+\mathrm{\&phi;})+n_{2,I}$

$S_Q=\frac{1}{2}{N}_S\mathrm{ADR}\left(\mathrm{\&Delta;\&tau;}\right)\sin c\left({\mathrm{\&Delta;\&omega;}}_D{T}_C\right)\sin ({\mathrm{\&Delta;\&omega;}}_D{T}_C+\mathrm{\&phi;})+n_{2,Q}$

Wherein, A represents signal amplitude, and R (τ) is autocorrelation function, noise n _{2, I}with n _{2, Q}gaussian distributed n () represents Gaussian distribution.

When adopting the accumulation of real differential, the differential accumulation of a signal is as follows:

$\begin{array}{c}{\mathrm{DF}}_k=S_{k,l}{S}_{k-a,I}+S_{k,Q}{S}_{k-a,Q}\&ap;\frac{1}{4}{N}_S^2{A}^2{R}^2\left(\mathrm{\&Delta;\&tau;}\right){\sin c}^2\left({\mathrm{\&Delta;\&omega;}}_d{N}_S{T}_C\right)\\ +n_{2,I,k}\&times;n_{2,I,k-1}+n_{2,Q,k}\&times;n_{2,Q,k-1}\end{array}$

If differential cumulative number is M step, then result is:

$\mathrm{DF}=\underset{k=1}{\overset{M}{\mathrm{\&Sigma;}}}{\mathrm{DF}}_k=\frac{1}{4}{\mathrm{MN}}_S^2{A}^2{R}^3\left(\mathrm{\&Delta;\&tau;}\right){\sin c}^2\left({\mathrm{\&Delta;\&omega;}}_d{N}_S{T}_C\right)+n_{3,I}+n_{3,Q}$

Wherein Δ τ _kwith Δ τ _k-1represent Delay Estima-tion error during moment k and k-1 respectively, Δ ω _d,kwith Δ ω _{d, k-1}represent the Doppler frequency estimation error of moment k and k-1 respectively, Δ ω _drepresentative is average Doppler frequency shift evaluated error from moment k-1 to moment k.Can prove there is Δ τ when code phase and Doppler shift complete matching _k=Δ τ _k-1=0 and Δ ω _d,k=Δ ω _{d, k-1}=0.

$\max \left\{\mathrm{DF}\right\}\&ap;\frac{1}{4}{\mathrm{MN}}_S^2{A}^2$

Wherein, noise n _{3, I}and n _{3, Q}gaussian distributed n () represents Gaussian distribution, γ be greater than zero constant coefficient, its value depends on integral time.

Step 2, the decomposition of application Lifting Wavelet are analyzed faint received global navigation satellite system signal.

The decomposition of this step application Lifting Wavelet is carried out analysis to Weak GNSS signal and is comprised following three steps:

(1) decomposition step:

In decomposition, original signal sequence Sj is broken down into two subdivisions according to parity---even _j-1and odd _j-1

(odd _j-1,even _j-1)＝Split(S _j)

This step is called the lazy wavelet transformation in lifting scheme.Signal decomposition is just two parts by this step simply, can not promotion signal.Next step lifting will identify that these two sequences are to reduce its correlativity.

(2) prediction steps:

Predicting machine is defined as

d _j-1＝odd _j-1-P(even _j-1)

Wherein, P (even _j-1) be predict odd number value by even item, and specifying information d _j-1pass through odd _j-1with prediction P (even _j-1) between error represent.This step is called the dual lifting in lifting scheme.When the correlativity of signal is very high, prediction is very effective.

(3) step of updating:

In order to make similarity signal S _j-1maintain original signal S _jsome characteristics, then press following formula upgrade average:

S _j-1＝even _j-1+U(d _j-1)

This step is called the original lifting in lifting scheme.

If three steps of above-mentioned decomposition are all for decomposed similarity signal S _j-1, then can obtain original signal S after the iterations of some _jmultilayer decompose.Can find, using lifting scheme to carry out one of sharpest edges of wavelet decomposition is that wavelet transformation can be decomposed into some basic steps and carries out, and their inverse transformation is more easily found.

For the setting identical with wavelet filter, lifting process can improve the speed of wavelet transformation doublely.The computation complexity of wavelet transformation is defined as calculating the multiplication number of times required for a pair coefficient and addition number of times when ground floor decomposes.For lifting scheme, suppose that the length of predicted operation coefficient and the length of renewal rewards theory coefficient divide other M and N.Can obtain:

$d\left(n\right)=\mathrm{odd}\left(n\right)-\underset{k=1}{\overset{M}{\mathrm{\&Sigma;}}}p\left(k\right)\mathrm{even}(n-M_d+k-1)$

$S\left(n\right)=\mathrm{even}\left(n\right)-\underset{k=1}{\overset{N}{\mathrm{\&Sigma;}}}u\left(k\right)d(n-N_d+k-2)$

Wherein, M _d=M/2-1, N _d=N/2-1.

Can draw from above formula, the computation complexity of lifting scheme is 2 (M+N), comprises the addition that (M+N) secondary multiplication is secondary with (M+N).In addition, the computation complexity of wavelet transform (DWT) is 4 (M+N)+2.

Therefore, for longer wave filter, the calculated amount of application lifting scheme is about the half of traditional algorithm.

Step 3, application signal intensity self-adaptation tabu search algorithm are selected to be optimized to the dual threshold in Lifting Wavelet denoising.

Feeble signal catches to combine with Wavelet Denoising Method by the present invention, and wherein Wavelet Denoising Method has been widely used in a lot of field.Signal to noise ratio (S/N ratio) can be brought up to some specific degree by Wavelet Denoising Method, therefore also becomes a study hotspot in satellite navigation application.When Wavelet Denoising Method proposes first for target be reduce white noise, be then applied to ultrasonic signal obtain and squelch aspect.As for field of satellite navigation, propose a kind of method based on wavelet transform (DWT) and be applied to catching of GNSS signal.Further, utilize Wavelet noise-eliminating method to improve output signal-to-noise ratio and be also used in the aftertreatment aspect that weak GPS signals obtains detection.

Tabu search algorithm is a kind of overall Neighborhood-region-search algorithm, and the method simulates the optimization feature of human mind function, and avoids roundabout search by local neighbor seaching mechanism and corresponding taboo criterion.Meanwhile, by the standard of breaking a taboo, it also can discharge some outstanding taboo states to ensure the diversity of efficient search.In addition, tabu search can suppress the defect falling into the easy precocious of genetic algorithm, reaches final optimization.

Wavelet Denoising Method mode comprises following two kinds of modes: hard-threshold mode and soft-threshold mode.All coefficients lower than bottom threshold are set to zero by the former, and other coefficient all remains unchanged; All coefficients lower than bottom threshold are set to zero by the latter, but other coefficient shrinks to zero.Determine scheme for threshold value, heuritic approach, after the iteration of carrying out certain step number, can choose optimum threshold value in theory.

High-frequency components and low frequency component is comprised in Weak GNSS signal.Ratio due to these two kinds of compositions is unfixed, if therefore easy choice hard-threshold mode or soft-threshold mode are by the information of inevitably losing certain frequency scope, because they delete high-frequency components all rambunctiously.Therefore, two kinds of algorithms above-mentioned all can not obtain gratifying performance in Wavelet Denoising Method process.

Apply signal intensity self-adaptation Tabu search algorithm in the present invention and carry out Wavelet Denoising Method.Because tabu search algorithm is a kind of heuritic approach, the heuritic approach that it is combined prior to being investigated algorithm by random algorithm and this locality.Signal intensity self-adaptation Tabu search algorithm process of the present invention is as follows:

(1) upper threshold and lower limit---T ' is supposed _minwith T ' _maxbe defined as four position effective digitals, wherein three be positioned at radix point after, as [y ₅, y ₆, y ₇, y ₈], [y ₁, y ₂, y ₃, y ₄].

(2) [T is set between the original area of threshold optimization _min, T _max], upper limit T _maxbe set to:

T _max＝3σ

In formula, σ is estimated as σ ≈ Mid/0.6745.

Mid be Decomposition order minimum time wavelet coefficient by the median after numerical ordering.

Lower limit T _minit is the absolute value of minimum coefficient than 0 in jth layer wavelet decomposition.

To retrain T ' as follows _minwith T ' _maxcarry out initialization:

T _min≤T′ _min＜T′ _max≤T _max

(3) at T ' _minwith T ' _maxbetween produce the subset of a candidate solution, then according to following fitness function, it is sorted:

$S=\mathrm{\&alpha;}\&times;\lg \left[\frac{\underset{n}{\mathrm{\&Sigma;}}{f}_d^2\left(n\right)}{\underset{n}{\mathrm{\&Sigma;}}{[f_d\left(n\right)-f\left(n\right)]}^2}\right]$

Wherein, α is the auto-adaptive parameter chosen according to the strength information of faint received global navigation satellite system signal, it according to the value of the actual information self-adaptative adjustment fitness function of the feeble signal of catching, can make fitness function can make for different feeble signals and judges more accurately; f _dfor the reconstruction signal after denoising, f is reference signal, and the value of s is larger, represents that fitness is higher;

$\mathrm{\&alpha;}=\sqrt{{\mathrm{\&sigma;}}_{\mathrm{\&omega;}}^2\left(t\right)+{\mathrm{\&sigma;}}_n^2}$

Wherein, σ _ωt root-mean-square value that () is signal, for system average noise power.

(4) process of signal intensity self-adaptation tabu search according to Fig. 1, carry out once simple correction to current solution and obtain feasible solution, such process is called a moved further.If move to its contiguous solution---candidate solution T* is not better than current optimum solution, and in order to avoid local optimum, candidate solution T* will be accepted no matter whether it is optimum solution.Then introduce taboo list and avoid circulation.All mobile steps that can not be accepted as current solution are stored in taboo list.The mobile step meeting taboo rule will be stored in taboo list.The use of taboo list reduces the possibility of circulation, because it is prevented that the solution returning in certain iterative steps and just accessed in the recent period.

(5) (3) are returned until meet stopping criterion.

(6) following formulae discovery optimal threshold is utilized:

T′ _max＝y _1j×10 ⁰+y _2j×10 ^-1+y _3j×10 ^-2+y _4j×10 ^-3

T′ _min＝y _5j×10 ⁰+y _6j×10 ^-1+y _7j×10 ^-2+y _8j×10 ^-3。

Step 4, optimized dual threshold is utilized to carry out aftertreatment to the faint received global navigation satellite system signal of catching.

Step 3 optimal double threshold value [T ' _min, T ' _max] determined after, by following formula by a kth wavelet coefficient W in jth layer wavelet decomposition _j,k:

$W_{j,k}^{\&prime;}=\left\{\begin{array}{c}0,\left|W_{j,k}\right|<T_{\min }^{\&prime;};\\ \frac{T_{\max }^{\&prime;}}{T_{\max }^{\&prime;}-T_{\min }^{\&prime;}}\left(\right|W_{j,k}|-T_{\min }^{\&prime;})\mathrm{sgn}\left(W_{j,k}\right),T_{\min }^{\&prime;}\&le;\left|W_{j,k}\right|\&le;T_{\max }^{\&prime;};\\ W_{j,k},\left|W_{j,k}\right|>T_{\max }^{\&prime;}.\end{array}\right.$

Achieved the Lifting Wavelet dual threshold value denoise algorithm optimized based on signal intensity self-adaptation tabu search being applied to Weak GNSS signal and catching by above four steps, significantly reduce time complexity and improve denoising accuracy.

In order to be described effect of the present invention, adopt the mode of Computer Simulation to carry out modeling to being applied to the present invention that Weak GNSS signal catches below, and achieving the simulation to real scene by assignment.Digital medium-frequency signal is as the input signal of emulation, and detailed process divides following four aspects to analyze:

(1) selection of the wavelet basis Sum decomposition number of plies

As shown in Figure 2, when decomposition level rises to 3, signal to noise ratio (S/N ratio) keeps stable even performance better.Haar wavelet transform base is put up the best performance in all candidate's wavelet basiss, and due to signal be square wave, the wavelet basis with minor fluctuations is more applicable, therefore selects Haar wavelet transform base as the wavelet basis applied in algorithm of the present invention.

In addition, also contains the discussion that Decomposition order is selected in fig. 2.If Decomposition order is too much, because shortage computing time can make efficiency reduce.And, lose the risk of effective information and can cause a poor signal to noise ratio (S/N ratio) result.But, when Decomposition order is not enough, can not optimum be obtained equally.Therefore, the present invention selects three layers of decomposition.

(2) denoising effect

The line be positioned in Fig. 3 below represents original signal, and another representative uses the Lifting Wavelet dual threshold value denoise algorithm (the present invention) optimized based on signal intensity self-adaptation tabu search to carry out the signal after denoising.Can obviously find out, along with the raising of input signal power, not only the output signal-to-noise ratio of original signal improves, and the output signal-to-noise ratio of denoised signal also improves thereupon.And the difference between them remains unchanged substantially, the system of this means can process the feeble signal of this input signal power between-176dBw to-168dBw, and signal to noise ratio (S/N ratio) can provide the gain of 8dB.

(3) contrast of the present invention and other algorithm

The comparison diagram of output signal-to-noise ratio in four kinds of situations is shown in Fig. 4.Line with round dot represents the denoising effect of the algorithm proposed in the present invention, namely based on the Lifting Wavelet dual threshold value denoise algorithm that signal intensity self-adaptation tabu search is optimized.Clearly, fact proved, the method that can obtain best output signal-to-noise ratio is the algorithm proposed in the present invention, and is used alone effect that Lifting Wavelet carries out denoising and is better than and is used alone conventional discrete wavelet transformation (DWT).Simultaneously, along with the increase of input signal power, be used alone DWT and be used alone the snr gain that Lifting Wavelet denoising can obtain and reduce gradually, on the contrary, the performance based on the Lifting Wavelet dual threshold value denoise algorithm of signal intensity self-adaptation tabu search optimization to be stablized and gratifying.

(4) wavelet decomposition lifting scheme and the contrast of DWT scheme on time complexity is applied

In the following table, compared for the computing time of two kinds of denoising methods, namely based on the DWT small echo dual threshold value denoise algorithm of signal intensity self-adaptation tabu search optimization and the Lifting Wavelet dual threshold value denoise algorithm based on the optimization of signal intensity self-adaptation tabu search.Should be noted that, specific computing time can not represent the real real time, but this contrast can be regarded as computation complexity and the reflection implementing difficulty.Therefore, can reach a conclusion, DWT needs to consume the time far above application wavelet decomposition lifting scheme.

It is emphasized that; embodiment of the present invention is illustrative; instead of it is determinate; therefore the present invention includes the embodiment be not limited to described in embodiment; every other embodiments drawn by those skilled in the art's technical scheme according to the present invention, belong to the scope of protection of the invention equally.

Claims (5)

1., based on a Lifting Wavelet dual threshold value denoise algorithm for signal intensity self-adaptation tabu search, it is characterized in that comprising the following steps:

Step 1, differential coherence Cumulate algorithm is utilized to catch faint received global navigation satellite system signal;

Step 2, the decomposition of application Lifting Wavelet are analyzed faint received global navigation satellite system signal;

Step 3, application signal intensity self-adaptation tabu search algorithm are selected to be optimized to the dual threshold in Lifting Wavelet denoising;

Step 4, optimized dual threshold is utilized to carry out aftertreatment to the faint received global navigation satellite system signal of catching.

2. the Lifting Wavelet dual threshold value denoise algorithm based on signal intensity self-adaptation tabu search according to claim 1, it is characterized in that: the implementation method of described step 1 is: obtain feeble signal by differential coherence accumulation, then accumulation results is inputted Wavelet Denoising Method wave filter to find its code phase and Doppler shift, thus complete acquisition procedure.

3. the Lifting Wavelet dual threshold value denoise algorithm based on signal intensity self-adaptation tabu search according to claim 1, is characterized in that: the concrete treatment step of described step 2 is:

(1) decomposition step, by original signal sequence S _jtwo subdivisions are decomposed into according to parity---even _j-1and odd _j-1:

(odd _j-1,even _j-1)＝Split(S _j)

(2) prediction steps, predict by following predicting machine:

d _j-1＝odd _j-1-P(even _j-1)

Wherein, P (even _j-1) be predict odd number value by even item, specifying information d _j-1pass through odd _j-1with prediction P (even _j-1) between error represent;

(3) step of updating, upgrades average by following formula:

S _j-1＝even _j-1+U(d _j-1)。

4. the Lifting Wavelet dual threshold value denoise algorithm based on signal intensity self-adaptation tabu search according to claim 1, is characterized in that: the concrete treatment step of described step 3 is:

(1) upper threshold and lower limit---T ' is supposed _minwith T ' _max

(2) [T is set to by between the original area of threshold optimization _min, T _max], by upper limit T _maxbe set to T _max=3 σ, wherein σ is estimated as: σ ≈ Mid/0.6745, in formula, Mid be Decomposition order minimum time wavelet coefficient by the median after numerical ordering, lower limit T _minit is the absolute value of minimum coefficient than 0 in jth layer wavelet decomposition;

(3) at T ' _minwith T ' _maxbetween produce the subset of a candidate solution, then according to following fitness function, it is sorted:

Wherein, α is the auto-adaptive parameter chosen according to the strength information of faint received global navigation satellite system signal, f _dfor the reconstruction signal after denoising, f is reference signal;

Wherein, σ _ωt root-mean-square value that () is signal, for system average noise power;

(4) search for by signal intensity self-adaptation TS algorithm, carry out correction obtain feasible solution to current solution, if move to its contiguous solution be not better than current optimum solution, then adjacent solution will be accepted; Then introduce taboo list and avoid circulation;

(5) step (3) is returned until meet stopping criterion;

(6) following formulae discovery optimal threshold is utilized:

T′ _max＝y _1j×10 ⁰+y _2j×10 ^-1+y _3j×10 ^-2+y _4j×10 ^-3

T′ _min＝y _5j×10 ⁰+y _6j×10 ^-1+y _7j×10 ^-2+y _8j×10 ^-3。

5. the Lifting Wavelet dual threshold value denoise algorithm based on signal intensity self-adaptation tabu search according to claim 1, is characterized in that: the concrete disposal route of described step 4 is: optimal double threshold value [T ' _min, T ' _max] determined after, a kth wavelet coefficient W in jth layer wavelet decomposition _j,kby following formula process: