CN105044681A - Mismatch filter optimization method of high-code-rate low-distance resolution phase coding signals - Google Patents

Abstract

The invention belongs to the field of a radar technology, and discloses a mismatch filter optimization method of high-code-rate low-distance resolution phase coding signals. The optimization method comprises the following steps: determining phase coding signals s; according to the phase coding signals s and the length Nh of a mismatch filter h, constructing a target function of the mismatch filter h; constructing a constraint condition of the target function about the mismatch filter h; according to the constraint condition of the target function, solving the target function of the mismatch filter h; and determining the mismatch filter. The method provided by the invention can reduce the distance sidelobe level of the phase coding signals after pulse compression.

Description

The mismatched filter optimization method of the low range resolution phase-coded signal of high code check

Technical field

The invention belongs to Radar Technology field, specifically the mismatched filter optimization method of the low range resolution phase-coded signal of a kind of high code check, for reducing the distance side lobe level after phase-coded signal mismatch filter.

Background technology

Pulse compression technique solves the contradiction between the operating distance of radar and range resolution, is widely used in field of radar.But pulse compression signal has higher distance side lobe usually, and higher distance side lobe is unfavorable for that radar effectively detects target, Faint target detection particularly under multiple goal or strong clutter background, the main lobe of weak signal target is very easily flooded by the distance side lobe of strong target echo and causes false dismissal, affects the detection performance of radar.Such as, the distance side lobe of high-amplitude may trigger false-alarm, and the distance side lobe of high-power target echo can flood the miniwatt target at neighbouring range unit place.

In order to reduce the distance side lobe of phase-coded signal, the pulse compression signal of design low sidelobe is usually adopted to realize.In existing multiple phase-coded signal, Barker code signal is optimum biphase coded signal, and it has minimum peak sidelobe, but the Baud Length of Barker code signal is the longest is 13, the application requirement on discontented full border.Compared with biphase coded signal, heterogeneous Barker code signal has lower peak sidelobe.Frank code, P1 code, P2 code, P3 code and P4 code are several typical polyphase code signals, and they are all the polyphase code signals of being derived by linear FM signal.P (n, k) coded signal is the polyphase code signal of being derived by NLFM signal, and compared with the polyphase code signal of being derived by linear FM signal, P (n, k) coded signal has lower peak sidelobe.

Existing phase-coded signal, under many circumstances, the application requirement on its peak sidelobe still discontented full border, and the Timed automata of existing phase-coded signal equals the Baud Length of phase-coded signal, inverse wide when the bandwidth of phase-coded signal equals a code element.Time wide fixing phase-coded signal can reduce distance side lobe further by increasing its Baud Length.But for a wideband radar system, its signal transacting bandwidth is fixing, now cannot reduces distance side lobe by simply increasing Baud Length, also not proposing the method for designing of the mismatched filter based on phase-coded signal for this problem at present.

Summary of the invention

For above-mentioned existing methods shortcoming, the object of the invention is to the optimization method of the mismatched filter proposing a kind of phase-coded signal, the distance side lobe level after phase-coded signal pulse compression can be reduced.

The technical thought realizing the object of the invention is: the phase-coded signal of the low range resolution of given high code check, when the bandwidth of phase-coded signal is constant, take snr loss as constraint condition, to minimize peak sidelobe that this phase-coded signal exported by mismatched filter and to approach the main lobe shape of expectation for objective function, optimize the mismatched filter obtaining actual needs.

For achieving the above object, the embodiment of the present invention adopts following technical scheme to be achieved.

A mismatched filter optimization method for the low range resolution phase-coded signal of high code check, is characterized in that, comprise the following steps:

(1) phase-coded signal s is determined;

(2) the length N of mismatched filter h is set _h, and according to the length N of described phase-coded signal s and mismatched filter h _h, build the objective function about described mismatched filter h;

(3) bound for objective function about described mismatched filter h is built;

(4) according to described bound for objective function, the objective function of described mismatched filter h is solved;

(5) described mismatched filter h is determined.

The feature of technical solution of the present invention and being improved to further:

Step (1) specifically comprises following sub-step:

(1a) first phase coded signal s is set ₁baud Length N ₁;

(1b) according to described first phase coding s ₁baud Length N ₁, determine the Baud Length N of described phase-coded signal s _s, wherein, N _s=b × N ₁, b is the increase multiple of the Baud Length of phase-coded signal s, and b value is integer;

(1c) the main lobe width controlled quentity controlled variable M=fix (δ × b) of described phase-coded signal s is set, δ is experience factor, value is in the scope of (0.5 ~ 1.0), and b is the increase multiple of the Baud Length of phase-coded signal s, and b value is integer;

(1d) according to the Baud Length N of described phase-coded signal s _s, the main lobe width controlled quentity controlled variable M of described phase-coded signal s, builds the objective function about described phase-coded signal s and constraint condition;

(1e) according to described objective function and described constraint condition, described phase-coded signal s is solved.

Step (2) specifically comprises following sub-step:

(2a) the pulse compression vector of phase-coded signal s after mismatched filter h is ρ _c, wherein, ${\mathrm{\ρ}}_c=s\⊗h/N_s={[{\mathrm{\ρ}}_1,{\mathrm{\ρ}}_2,...,{\mathrm{\ρ}}_i,...,{\mathrm{\ρ}}_{N_s+N_h-1}]}^T,i\∈[1,N_s+N_h-1];$

(2b) at the pulse compression vector ρ of phase-coded signal s after mismatched filter h _cin, to each distance side lobe delivery value respectively, obtain distance side lobe modulus value vector ρ:

$\mathrm{\ρ}={\left[\right|{\mathrm{\ρ}}_1|,...,|{\mathrm{\ρ}}_k|,...|{\mathrm{\ρ}}_{(N_s+N_h)/2-M-1}|,|{\mathrm{\ρ}}_{(N_s+N_h)/2+M+1}|...,|{\mathrm{\ρ}}_l|,...|{\mathrm{\ρ}}_{N_s+N_h-1}\left|\right]}^T$

Wherein, k ∈ [1, (N _s+ N _h)/2-M-1], l ∈ [(N _s+ N _h)/2+M+1, N _s+ N _h-1];

(2c) at the pulse compression vector ρ of phase-coded signal s after mismatched filter h _cin, the actual main lobe B of output _mfor $B_m={[{\mathrm{\ρ}}_{(N_s+N_h)/2-M},{\mathrm{\ρ}}_{(N_s+N_h)/2-M+1},....,{\mathrm{\ρ}}_t,...,{\mathrm{\ρ}}_{(N_s+N_h)/2+M}]}^T,$

Wherein, t ∈ [(N _s+ N _h)/2-M, (N _s+ N _h)/2+M];

(2d) the expectation main lobe b that phase-coded signal s exports after mismatched filter h is determined _m, wherein, the main lobe expected can be rectangle, the main lobe of sinc function, or the main lobe not increasing the phase-coded signal before he number.Main lobe b is expected described in this example _mshape be the main lobe of described first phase coded signal, by described first phase coded signal s ₁functional value interval corresponding to the main lobe part discrete column vector turning to 2 × M+1 dimension equably, and using described column vector as expectation main lobe b _mvalue;

(2e) according to described distance side lobe modulus value vector ρ, described actual main lobe B _mwith described expectation main lobe b _m, establishing target function $\begin{array}{cc}\underset{h}{\min }& {\left|\right|\mathrm{\ρ}\left|\right|}_{\∞}\end{array}+\mathrm{\α}\·{\left|\right|B_m-b_m\left|\right|}_{\∞}.$

Step (3) specifically comprises following sub-step:

(3a) by the symmetrical zero padding of phase-coded signal s head and the tail, being extended to length is N _hphase-coded signal

(3b) be N by length _hphase-coded signal spin upside down and get conjugation and obtain matched filter

Wherein, () ^*represent the conjugation of ().

(3c) according to described mismatched filter h, described matched filter the bound for objective function obtained about described mismatched filter h is

Further, step (1d) specifically comprises step quickly:

A () calculates the kth of phase-coded signal s ₁individual distance side lobe ρ _k1:

Wherein, displacement k1=M+1, M+2 ..., (N _s-1), M is the main lobe width controlled quentity controlled variable of phase-coded signal s, J _k1for slip matrix, slip matrix J _k1form be:

$J_{k1}=J_{-k1}^T=\left[\begin{array}{cc}{0}_{(N_s-k1)\×k1}& I_{N_s-k1}\\ 0_{k1\×k1}& 0_{k1\×(N_s-k1)}\end{array}\right],$

In formula, () ^trepresent transposition, 0 represents full null matrix, I representation unit matrix, 0 and the dimension of subscript representing matrix of I;

B () is according to the distance side lobe ρ of phase-coded signal s _k1, obtain the peak sidelobe PSL of distance side lobe _s, PSL _s=max| ρ _k1|; According to the peak sidelobe PSL of distance side lobe _s, the objective function obtaining phase-coded signal s is: pSL _s, wherein, p is the phase vectors of phase-coded signal s;

C () determines the expectation main lobe b of phase-coded signal s _n, wherein, the main lobe expected can be rectangle, the main lobe of sinc function, or the main lobe not increasing the phase-coded signal before he number.Main lobe b is expected described in this example _nshape be the main lobe of described first phase coded signal, by described first phase coded signal s ₁functional value interval corresponding to the main lobe part discrete column vector turning to 2 × M+1 dimension equably, get M element wherein to the 1st element as expecting main lobe b _nvalue;

D () is according to described expectation main lobe b _n, the constraint condition obtaining phase-coded signal s is:

$\begin{array}{cc}s.t.& |b_n\left(k2\right)-\frac{s^H{J}_{k2}s}{N_s}|\≤\mathrm{\γ},k2=\mathrm{1,2},...,M\end{array}$

0≤p(k)≤2π，k＝1，2，…，N _s

Wherein, b _n(k2) expectation main lobe b is represented _nkth 2 elements, J _k2for slip matrix, displacement k2=1,2 ..., M, M are the main lobe width controlled quentity controlled variable of phase-coded signal s, and p (k) represents a kth element of phase vectors p, k=1,2 ..., N _s, γ is constant, and γ represents that the main lobe of phase-coded signal s approaches and expects main lobe b _ndegree, value is 0.01 ~ 0.2.

The present invention compared with prior art tool has the following advantages.The mismatched filter of the present invention's design is while guarantee phase-coded signal bandwidth is constant, add phase-coded signal Baud Length, in the process of therefore optimal design, there is larger degree of freedom, thus reduce further the distance side lobe level of phase-coded signal.

Accompanying drawing explanation

Fig. 1 is the schematic flow sheet of the mismatched filter optimization method of the low range resolution phase-coded signal of a kind of high code check that the embodiment of the present invention provides;

To be the mismatched filter that provides of the embodiment of the present invention carry out the main lobe after mismatch filter and phase-coded signal to phase-coded signal to Fig. 2 carries out the main lobe after matched filtering and contrast schematic diagram, horizontal ordinate represents relative time delay, unit is us, and ordinate represents amplitude, and unit is dB;

Fig. 3 is that mismatched filter that the embodiment of the present invention provides carries out the result after mismatch filter to phase-coded signal and phase-coded signal carries out the Comparative result schematic diagram after matched filtering, horizontal ordinate represents relative time delay, unit is us, and ordinate represents amplitude, and unit is dB;

To be the mismatched filter that provides of the embodiment of the present invention carry out to phase-coded signal the contrast schematic diagram that the main lobe phase after mismatch filter and phase-coded signal carry out the main lobe phase after matched filtering to Fig. 4, horizontal ordinate represents relative time delay, unit is us, and ordinate represents phase place, and unit is degree.

Embodiment

With reference to Fig. 1, performing step of the present invention is as follows.

Step 1, determines phase-coded signal s.

Determine that phase-coded signal s specifically comprises following sub-step:

(1a) first phase coded signal s is set ₁baud Length N ₁.

(1b) according to first phase coding s ₁baud Length N ₁, determine the Baud Length N of phase-coded signal s _s.

Wherein, N _s=b × N ₁, b is the increase multiple of the Baud Length of phase-coded signal s, and b value is integer, and N _h-N _sfor even number, N _h+ N _salso be even number.

(1c) the main lobe width controlled quentity controlled variable M=fix (δ × b) of phase-coded signal s is set.

Wherein, δ is experience factor, and general value is in the scope of (0.5-1.0), and usual value is 0.8; B represents the multiple that the transmission of symbols speed of phase-coded signal s increases, and is also the increase multiple of phase-coded signal s Baud Length; Fix () represents downward round numbers.

Here determine that the value formula M=fix (δ × b) of the main lobe width controlled quentity controlled variable M of phase-coded signal s is an experimental formula.In practice, the remarkable situation changing the bandwidth of phase-coded signal s may be there is according to the M value that this experimental formula obtains, now need artificially to adjust the value of phase-coded signal s main lobe width controlled quentity controlled variable M, the process of M value adjustment follows following rule: the value increasing phase-coded signal s main lobe width controlled quentity controlled variable M, and the bandwidth of phase-coded signal s reduces; Reduce the value of phase-coded signal s main lobe width controlled quentity controlled variable M, the bandwidth of phase-coded signal s increases.Therefore, compromise is needed to consider that main lobe approaches the size expecting the degree of main lobe and the bandwidth of phase-coded signal s when determining the value of main lobe width controlled quentity controlled variable M.

(1d) according to the Baud Length N of phase-coded signal s _s, the main lobe width controlled quentity controlled variable M of phase-coded signal s, builds the objective function about phase-coded signal s and constraint condition.

Build the objective function about phase-coded signal s and constraint condition, specifically comprise following sub-step:

A () calculates kth 1 distance side lobe ρ of phase-coded signal s _k1,

${\mathrm{\ρ}}_{k1}=\frac{s^H{J}_{k1}s}{N_s}$

Wherein, displacement k1=M+1, M+2 ..., (N _s-1), M is the main lobe width controlled quentity controlled variable of phase-coded signal s, N _srepresent the Baud Length of phase-coded signal s, J _k1for slip matrix, slip matrix J _k1concrete form be:

$J_{k1}=J_{-k1}^T=\left[\begin{array}{cc}{0}_{(N_s-k1)\×k1}& I_{N_s-k1}\\ 0_{k1\×k1}& 0_{k1\×(N_s-k1)}\end{array}\right],$

In formula, () ^trepresent transposition, 0 represents full null matrix, I representation unit matrix, 0 and the dimension of subscript representing matrix of I.

B () is according to the distance side lobe ρ of phase-coded signal s _k1, obtain the peak sidelobe PSL of distance side lobe _s, PSL _s=max| ρ _k1|; According to the peak sidelobe PSL of distance side lobe _s, the objective function obtaining phase-coded signal s is: pSL _s, wherein, p is the phase vectors of phase-coded signal s; Wherein, max represents and gets maximal value, and min represents and gets minimum value, || represent delivery value _s

C () determines the expectation main lobe b of phase-coded signal s _n.

Wherein, the main lobe expected can be rectangle, the main lobe of sinc function, or the main lobe not increasing the phase-coded signal before he number.Main lobe b is expected described in this example _nshape be the main lobe of described first phase coded signal, by described first phase coded signal s ₁functional value interval corresponding to the main lobe part discrete column vector turning to 2 × M+1 dimension equably, get M element wherein to the 1st element as expecting main lobe b _nvalue.

D () is according to expectation main lobe b _n, the constraint condition obtaining phase-coded signal s is:

$\begin{array}{cc}s.t.& |b_n\left(k2\right)-\frac{s^H{J}_{k2}s}{N_s}|\≤\mathrm{\γ},k2=\mathrm{1,2},...,M\end{array}$

0≤p(k)≤2π，k＝1，2，…，N _s，

Wherein, s.t. represents constraint condition, || represent delivery value, b _n(k2) expectation main lobe b is represented _nkth 2 elements, J _k2for slip matrix, displacement k2=1,2 ..., M, M are the main lobe width controlled quentity controlled variable of phase-coded signal s, N _srepresent the Baud Length of phase-coded signal s, p (k) represents a kth element of the phase vectors p of phase-coded signal s, k=1,2 ..., N _s, γ is constant, and γ represents that the main lobe of phase-coded signal s approaches and expects main lobe b _ndegree, value is 0.01 ~ 0.2.

In theory, phase restriction 0≤p (k)≤2 π, k=1,2 ..., N _scan ignore, because exp function cycle is 2 π, but in the solution procedure of reality, phase restriction can ensure that optimized algorithm is better restrained, and therefore, in constraint condition, gives phase restriction.

Constant γ is an empirical value, finds in emulation, and when the shape expecting main lobe is the shape of sinc function main lobe, the main lobe shape after γ value 0.1 just can make phase-coded signal s pulse compression well approaches the shape expecting main lobe.If reduce the value of constant γ, the main lobe after phase-coded signal s pulse compression better can approach the shape expecting main lobe, but can raise the peak sidelobe of distance side lobe.If increase the value of constant γ, the peak sidelobe of distance side lobe can reduce, but the main lobe after phase-coded signal s pulse compression is to expecting that the Approximation effect of main lobe is deteriorated.Therefore, compromise is needed to consider when determining the value of constant γ.

(1e) according to described objective function and described constraint condition, described phase-coded signal s is solved.

Under constraint condition, use p norm optimization Algorithm for Solving objective function, the phase-coded signal s after being optimized.

Step 2, the length N of setting mismatched filter h _h, and according to the length N of phase-coded signal s and mismatched filter h _h, build the objective function about mismatched filter h.

Mismatched filter h is the mismatched filter of phase-coded signal s.

The length N of setting mismatched filter h _h, and according to the Baud Length N of phase-coded signal s _swith main lobe width controlled quentity controlled variable M, and the length N of mismatched filter h _h, build the objective function solving mismatched filter h, specifically comprise following sub-step:

(2a) the pulse compression vector of phase-coded signal s after mismatched filter h is ρ _c, ${\mathrm{\ρ}}_c=s\⊗h/N_s={[{\mathrm{\ρ}}_1,{\mathrm{\ρ}}_2,...,{\mathrm{\ρ}}_i,...,{\mathrm{\ρ}}_{N_s+N_h-1}]}^T,i\∈[1,N_s+N_h-1].$

Wherein, represent convolution, [.] ^trepresent transposition, i ∈ [1, N _s+ N _h-1], N _srepresent the Baud Length of phase-coded signal s, N _hrepresent the length of mismatched filter h.

(2b) at the pulse compression vector ρ of phase-coded signal s after mismatched filter h _cin, to each distance side lobe delivery value respectively, obtain distance side lobe modulus value vector ρ:

$\mathrm{\ρ}={\left[\right|{\mathrm{\ρ}}_1|,...,|{\mathrm{\ρ}}_k|,...|{\mathrm{\ρ}}_{(N_s+N_h)/2-M-1}|,|{\mathrm{\ρ}}_{(N_s+N_h)/2+M+1}|...,|{\mathrm{\ρ}}_l|,...|{\mathrm{\ρ}}_{N_s+N_h-1}\left|\right]}^T$

Wherein, k ∈ [1, (N _s+ N _h)/2-M-1], l ∈ [(N _s+ N _h)/2+M+1, N _s+ N _h-1];

At the pulse compression vector ρ of phase-coded signal s after mismatched filter h _cin, the left comer code vector L of distance side lobe extracting the main lobe left side be L=[1 ..., m ..., (N _s+ N _h)/2-M-1] and main lobe on the right of the right corner code vector R of distance side lobe be R=[(N _s+ N _h)/2+M+1 ..., n ..., N _s+ N _h-1], wherein, m ∈ [1, (N _s+ N _h)/2-M-1], n ∈ [(N _s+ N _h)/2+M+1, N _s+ N _h, and m and n is positive integer-1]; After the distance side lobe delivery value that note exports, obtain secondary lobe modulus value vector:

$\mathrm{\ρ}={\left[\right|{\mathrm{\ρ}}_1|,...,|{\mathrm{\ρ}}_k|,...|{\mathrm{\ρ}}_{(N_s+N_h)/2-M-1}|,|{\mathrm{\ρ}}_{(N_s+N_h)/2+M+1}|...,|{\mathrm{\ρ}}_l|,...|{\mathrm{\ρ}}_{N_s+N_h-1}\left|\right]}^T$

Wherein, || represent delivery value, k ∈ [1, (N _s+ N _h)/2-M-1], l ∈ [(N _s+ N _h)/2+M+1, N _s+ N _h-1].

(2c) at the pulse compression vector ρ of phase-coded signal s after mismatched filter h _cin, the actual main lobe B of output _mfor $B_m={[{\mathrm{\ρ}}_{(N_s+N_h)/2-M},{\mathrm{\ρ}}_{(N_s+N_h)/2-M+1},....,{\mathrm{\ρ}}_t,...,{\mathrm{\ρ}}_{(N_s+N_h)/2+M}]}^T,$

Wherein, t ∈ [(N _s+ N _h)/2-M, (N _s+ N _h)/2+M].

(2d) the expectation main lobe b that phase-coded signal s exports after mismatched filter h is determined _m, wherein, the main lobe expected can be rectangle, the main lobe of sinc function, or the main lobe not increasing the phase-coded signal before he number.Main lobe b is expected in this example _mshape be first phase coded signal s ₁main lobe, by first phase coded signal s ₁functional value interval corresponding to the main lobe part discrete column vector turning to 2 × M+1 dimension equably, and using column vector as expectation main lobe b _mvalue.

(2e) according to distance side lobe modulus value vector ρ, actual main lobe B _mwith expectation main lobe b _m, establishing target function $\begin{array}{cc}\underset{h}{\min }& {\left|\right|\mathrm{\ρ}\left|\right|}_{\∞}\end{array}+\mathrm{\α}\·{\left|\right|B_m-b_m\left|\right|}_{\∞}.$

Wherein, h is length is N _hmismatched filter, || || _∞represent Infinite Norm, α is the arithmetic number of artificial setting, for Sidelobe Suppression effect and the main lobe approximation ratio of compromising.If the value of α is larger, the main lobe of phase-coded signal s after mismatched filter h better can approach the shape expecting main lobe, but the peak sidelobe of distance side lobe can be raised, if the value of α is less, the peak sidelobe of distance side lobe can reduce, but the main lobe of phase-coded signal s after mismatched filter h is to expecting that the Approximation effect of main lobe is deteriorated, therefore, compromise is needed to consider when determining the value of arithmetic number α.

Step 3, builds the bound for objective function about mismatched filter h.

(3a) by the symmetrical zero padding of phase-coded signal s head and the tail, being extended to length is N _hphase-coded signal

(3b) be N by length _hphase-coded signal spin upside down and get conjugation and obtain matched filter

Wherein, () ^*represent the conjugation of ().

(3c) according to mismatched filter h, matched filter the bound for objective function obtained about mismatched filter h is:

$\left[\begin{array}{cc}s.t.& {\left|\right|h-\stackrel{\&OverBar;}{s}\left|\right|}_2<\mathrm{\&epsiv;}\end{array}\right]$

Wherein, s.t. represents constraint condition, || || ₂represent 2 norms, ε is length is N _hphase-coded signal mismatching, if the value of mismatching ε is larger, the peak sidelobe of phase-coded signal s after mismatched filter h can reduce, but snr loss can raise, if the value of ε is less, snr loss can diminish, but the peak sidelobe of phase-coded signal s after mismatched filter h can raise; Can snr loss according to actual needs adjust, usual value is 0 < ε < 10.

Step 4, according to bound for objective function, solves the objective function of mismatched filter h: $\begin{array}{cccc}\underset{h}{\min }& {\left|\right|\mathrm{\&rho;}\left|\right|}_{\&infin;}+\mathrm{\&alpha;}\&CenterDot;{\left|\right|B_m-b_m\left|\right|}_{\&infin;},& s.t.& {\left|\right|h-\stackrel{\&OverBar;}{s}\left|\right|}_2<\mathrm{\&epsiv;}.\end{array}$

The value of energy weight coefficient α belongs in the scope of [0,1].

Step 5, determines mismatched filter h.

In constraint condition under, use the convex optimization tool bag cvx of Matlab to solve objective function $\begin{array}{cc}\underset{h}{\min }& {\left|\right|\mathrm{\&rho;}\left|\right|}_{\&infin;}\end{array}+\mathrm{\&alpha;}{\left|\right|B_m-b_m\left|\right|}_{\&infin;},$ Obtain mismatched filter h.

Effect of the present invention can further illustrate in conjunction with emulation experiment.

(1) simulation parameter

First phase coded signal s ₁baud Length N ₁=64, transmission of symbols speed increases multiple b=4, then the Baud Length N of phase-coded signal s _s=256; The length N of mismatched filter _hthe main lobe width controlled quentity controlled variable M=4 of=512, phase-coded signal s, expects that the shape of main lobe is Baud Length N _smain lobe shape after the phase-coded signal s pulse compression of=256, weight coefficient α=0.3, mismatching ε=2.25.

(2) content is emulated

Emulation 1, designed phase coded signal s.

According to the Baud Length N of phase-coded signal s _swith main lobe width controlled quentity controlled variable M, by expecting that the shape of main lobe is for criterion with the peak sidelobe minimizing this phase-coded signal s with approaching, p norm optimization algorithm is used to obtain the phase-coded signal s of low distance side lobe.

Emulation 2, carries out matched filtering and mismatch filter respectively to phase-coded signal s.

In simulations, Baud Length N is set ₁the phase-coded signal s of=64 ₁with Baud Length N _sthe phase-coded signal s of=256 has identical time width 15.36us, and sample frequency is 100MHz.

The simulation and analysis of a main lobe characteristic that () phase-coded signal s outputs signal after carrying out matched filtering and mismatch filter respectively.

Respectively matched filtering and mismatch filter are carried out to phase-coded signal s, X-Y scheme is drawn as by after the main lobe result delivery value of output signal, as shown in Figure 2, the main lobe characteristic schematic diagram outputed signal after its center line-o-represents phase-coded signal s matched filtering, the main lobe characteristic schematic diagram that line-*-expression phase-coded signal s outputs signal after mismatched filter h.

As shown in Figure 2, the main lobe outputed signal after phase-coded signal s matched filtering ,-o-the line namely in figure, with the main lobe outputed signal after the coded signal s mismatch filter of position ,-the * namely in figure-line, overlaps substantially; Range value is-3dB place in the drawings, and the width of the main lobe outputed signal after phase-coded signal s matched filtering and mismatch filter is equal.Therefore, the bandwidth that the mismatched filter that the inventive method design obtains outputs signal after mismatch filter phase-coded signal s is constant.

The simulation and analysis of b amplitude characteristic that () phase-coded signal s outputs signal after carrying out matched filtering and mismatch filter respectively.

Respectively matched filtering and mismatch filter are carried out to phase-coded signal s, X-Y scheme is drawn as by after the result delivery value of output signal, as shown in Figure 3, the wherein amplitude characteristic figure that outputs signal after representing phase-coded signal s matched filtering of dotted line, solid line represents the amplitude characteristic figure that phase-coded signal s outputs signal after mismatched filter h.

As shown in Figure 3, the maximum amplitude value outputed signal after phase-coded signal s matched filtering is-36.98dB, the maximum amplitude value that phase-coded signal s outputs signal after mismatched filter h is-42.92dB, compared with the distance side lobe after matched filtering, reduce 5.94dB, snr loss-0.085dB; Therefore, the present invention designs the mismatched filter h that obtains while guarantee phase-coded signal s bandwidth is constant, by increasing the Baud Length of phase-coded signal s, reduce further the distance side lobe level of phase-coded signal s.

The simulation and analysis of c main lobe phase characteristic that () phase-coded signal s outputs signal after carrying out matched filtering and mismatch filter respectively.

Take out the phase place of the main lobe part outputed signal after phase-coded signal s carries out matched filtering and mismatch filter respectively, change into angle value, be drawn as X-Y scheme, as shown in Figure 4, the wherein phase propetry figure of main lobe that outputs signal after representing phase-coded signal s matched filtering of dotted line, solid line represents the phase propetry figure of the main lobe that phase-coded signal s outputs signal after mismatched filter h.

As shown in Figure 4, the main lobe phase variation range outputed signal after phase-coded signal s matched filtering is [-0.164 °, 0.164 °], solid line represents that the variation range of the main lobe phase that phase-coded signal s outputs signal after mismatched filter h is [-7.20710 ^-9° ,-7.20710 ^-9°], and the main lobe phase that phase-coded signal s outputs signal after mismatched filter h is compared with the main lobe phase outputed signal after phase-coded signal s matched filtering, and maximum difference is negligible.Therefore, the main lobe phase that outputs signal after the mismatched filter h that the inventive method design obtains can ensure pulse compression of phase-coded signal s is substantially constant.

Claims (5)

1. the mismatched filter optimization method of the low range resolution phase-coded signal of high code check, is characterized in that, comprise the following steps:

Step 1, determines phase-coded signal s;

Step 2, the length N of setting mismatched filter h _h, and according to the length N of described phase-coded signal s and described mismatched filter h _h, build the objective function about described mismatched filter h;

Step 3, builds the bound for objective function about described mismatched filter h;

Step 4, according to described bound for objective function, solves the objective function of described mismatched filter h;

Step 5, determines described mismatched filter h.

2. the mismatched filter optimization method of the low range resolution phase-coded signal of high code check according to claim 1, it is characterized in that, step 1 specifically comprises following sub-step:

(1a) first phase coded signal s is set ₁baud Length N ₁;

(1b) according to described first phase coding s ₁baud Length N ₁, determine the Baud Length N of described phase-coded signal s _s, wherein, N _s=b × N ₁, b is the increase multiple of the Baud Length of phase-coded signal s, and b value is integer;

(1c) set the main lobe width controlled quentity controlled variable M=fix (δ × b) of described phase-coded signal s, δ value is in the scope of (0.5 ~ 1.0), and b is the increase multiple of the Baud Length of phase-coded signal s, and b value is integer;

(1d) according to the Baud Length N of described phase-coded signal s _s, the main lobe width controlled quentity controlled variable M of described phase-coded signal s, builds the objective function about described phase-coded signal s and constraint condition;

(1e) according to objective function and the constraint condition of described phase-coded signal s, described phase-coded signal s is solved.

3. the mismatched filter optimization method of the low range resolution phase-coded signal of high code check according to claim 1, it is characterized in that, step 2 specifically comprises following sub-step:

(2a) the pulse compression vector of phase-coded signal s after mismatched filter h is ρ _c, ${\mathrm{\&rho;}}_c=s\&CircleTimes;h/N_s={[{\mathrm{\&rho;}}_1,{\mathrm{\&rho;}}_2,...,{\mathrm{\&rho;}}_i,...,{\mathrm{\&rho;}}_{N_s+N_h-1}]}^T,i\&Element;[1,N_s+N_h-1];$

(2b) at the pulse compression vector ρ of phase-coded signal s after mismatched filter h _cin, to each distance side lobe delivery value respectively, obtain distance side lobe modulus value vector ρ:

$\mathrm{\&rho;}=\left[\right|{\mathrm{\&rho;}}_1|,...,|{\mathrm{\&rho;}}_k|,...|{\mathrm{\&rho;}}_{(N_s+N_h)/2-M-1}|,|{\mathrm{\&rho;}}_{(N_s+N_h)/2+M+1}|...,|{\mathrm{\&rho;}}_l|,...|{\mathrm{\&rho;}}_{N_s+N_h-1}|{]}^T$

Wherein, k ∈ [1, (N _s+ N _h)/2-M-1], l ∈ [(N _s+ N _h)/2+M+1, N _s+ N _h-1];

(2c) at the pulse compression vector ρ of phase-coded signal s after mismatched filter h _cin, the actual main lobe B of output _mfor $B_m=[{\mathrm{\&rho;}}_{(N_s+N_h)/2-M},{\mathrm{\&rho;}}_{(N_s+N_h)/2-M+1},...,{\mathrm{\&rho;}}_t,...,{\mathrm{\&rho;}}_{(N_s+N_h)/2+M}{]}^T,$

Wherein, t ∈ [(N _s+ N _h)/2-M, (N _s+ N _h)/2+M];

(2d) the expectation main lobe b that phase-coded signal s exports after mismatched filter h is determined _m, wherein, described expectation main lobe b _mshape be the main lobe of described phase-coded signal s, by functional value interval corresponding for the main lobe part of the described phase-coded signal s discrete column vector turning to 2 × M+1 dimension equably, and using described column vector as expectation main lobe b _mvalue;

(2e) according to described distance side lobe modulus value vector ρ, described actual main lobe B _mwith described expectation main lobe b _m, build the objective function about described mismatched filter h $\begin{array}{cc}\underset{h}{\min }& {\left|\right|\mathrm{\&rho;}\left|\right|}_{\&infin;}+\mathrm{\&alpha;}\&CenterDot;{\left|\right|B_m-b_m\left|\right|}_{\&infin;},\end{array}$ The value of energy weight coefficient α belongs in the scope of [0,1].

4. the mismatched filter optimization method of the low range resolution phase-coded signal of high code check according to claim 1, it is characterized in that, step 3 specifically comprises following sub-step:

(3a) by the symmetrical zero padding of phase-coded signal s head and the tail, being extended to length is N _hphase-coded signal

(3b) be N by length _hphase-coded signal spin upside down and get conjugation and obtain matched filter

Wherein, () ^*represent the conjugation of ().

(3c) according to described mismatched filter h, described matched filter the bound for objective function obtained about described mismatched filter h is $\begin{array}{cc}s.t.& {\left|\right|h-\stackrel{\&OverBar;}{s}\left|\right|}_2<\mathrm{\&epsiv;}.\end{array}$

5. the mismatched filter optimization method of the low range resolution phase-coded signal of high code check according to claim 2, is characterized in that, sub-step (1d) also specifically comprises following sub-step:

A () calculates kth 1 distance side lobe ρ of phase-coded signal s _k1:

Wherein, be shifted k ₁=M+1, M+2 ..., (N _s-1), M is the main lobe width controlled quentity controlled variable of phase-coded signal s, J _k1for slip matrix, slip matrix J _k1form be:

$J_{k1}=J_{-k1}^T=\left[\begin{array}{cc}{0}_{(N_s-k1)\&times;k1}& I_{N_s-k1}\\ 0_{k1\&times;k1}& 0_{k1\&times;(N_s-k1)}\end{array}\right],$

In formula, () ^trepresent transposition, 0 represents full null matrix, I representation unit matrix, 0 and the dimension of subscript representing matrix of I;

B () is according to the distance side lobe ρ of phase-coded signal s _k1, obtain the peak sidelobe PSL of distance side lobe _s, PSL _s=max| ρ _k1|; According to the peak sidelobe PSL of distance side lobe _s, the objective function obtaining phase-coded signal s is: $\begin{array}{cc}\underset{p}{\min }& {\mathrm{PSL}}_s\end{array},$ Wherein, p is the phase vectors of phase-coded signal s;

C () determines the expectation main lobe b of phase-coded signal s _n, wherein, described expectation main lobe b _nshape be the shape of sinc function main lobe, by functional value interval corresponding for the sinc function main lobe part discrete column vector turning to 2 × M+1 dimension equably, get M element wherein to the 1st element as expecting main lobe b _nvalue;

D () is according to described expectation main lobe b _n, the constraint condition obtaining phase-coded signal s is:

$\begin{array}{cc}s.t.& |b_n\left(k2\right)-\frac{s^H{J}_{k2}s}{N_s}|\&le;\mathrm{\&gamma;},k2=\mathrm{1,2},...,M\end{array}$

0≤p(k)≤2π，k＝1，2，…，N _s，

Wherein, b _n(k2) expectation main lobe b is represented _nkth 2 elements, J _k2for slip matrix, displacement k2=1,2 ..., M, M are the main lobe width controlled quentity controlled variable of phase-coded signal s, and p (k) represents a kth element of phase vectors p, k=1,2 ..., N _s, γ is constant, and γ represents that the main lobe of phase-coded signal s approaches and expects main lobe b _ndegree, value is 0.01 ~ 0.2.