CN105589061A - Signal processing algorithm for shore-based radar - Google Patents

Abstract

The invention discloses a signal processing algorithm for a shore-based radar. According to the signal processing algorithm, sampling points within the repetition frequency of each pulse of an echo signal is stored in the range direction, and accumulative points are stored in the azimuth direction so as to form a two-dimensional matrix in the form of single-frame data. If a multi-frame signal is to be acquired, the two-dimensional matrix is extended to be a three-dimensional matrix. After that, the echo signal is subjected to the pulse pressure treatment, and then a processed signal is obtained through the video integration process by means of a double-pole filter. The target distance is determined, and then the constant false alarm rate (CFAR) is detected according to a fixed threshold firstly and then a floating threshold. Based on the signal processing algorithm, the echo signal is effectively treated and then the target location is successfully detected. Therefore, the requirements of pulse radars on the detection of inshore ships can be met.

Description

A kind of signal processing algorithm of land-based radar

Technical field

The invention belongs to radar monitoring field, relate to a kind of signal processing algorithm of land-based radar.

Background technology

In the development process of land-based radar, signal processing algorithm is the core technology of radar monitoring target. ForMeet radar and can carry out to inshore marine site the monitoring of round-the-clock, the sea Ship Target that monitoring is invaded is alsoThe main task of location, needs suitable signal processing algorithm. The algorithm of existing airborne radar can not be expiredThe requirement of foot land-based radar.

Summary of the invention

The object of this invention is to provide a kind of signal processing algorithm of land-based radar, land-based radar can be completedInshore marine site is carried out to the monitoring of round-the-clock, the task of the sea Ship Target that monitoring is invaded location.

Technical scheme of the present invention is as follows:

A signal processing algorithm for land-based radar, comprises the steps: to obtain after echo-signal, by echoSampled point in the each pulse recurrence frequency of signal be stored in distance to, accumulation count be stored in orientation to, fromAnd a two-dimensional matrix of formation; Described two-dimensional matrix is frame data, if sampling multiframe signal,Described two-dimensional matrix extends to three-dimensional matrice; Then echo-signal is carried out to pulse pressure processing, afterwards by bipolarPoint wave filter, carries out video integration to obtain processing signals, determines target range, then carries out the permanent void of CFARAlarm probability detects, and first passes through by fixed threshold the thresholding that floats again.

Useful technique effect of the present invention is:

Coast defence search radar adopts pulse system X-band radar signal, and it is assorted that the prompt change technology of proportion suppresses seaRipple, realizes the Search/Track to sea-surface target by technological means such as non-coherent accumulation and multi-objectives.Paired pulses echo carries out pulse pressure processing, subsequently by two-pole filter, carries out distance to non-coherent accumulation meter,Obtain range-to-go. In this project, do not consider target velocity, so do not carry out the FFT place of Doppler's dimensionReason.

The function of accumulation is that the amplitude of same distance door in adjacent several main pulses is done to sliding window accumulation, carries with thisHigh s/n ratio. Coherent accumulation is cumulative in strict accordance with phase relation, is N, if participate in the range gate number of accumulationSignal to noise ratio is improved N doubly. Be different from coherent accumulation and process, random during due to the initial phase of pulse echo,Therefore adopt non-coherent accumulation. Because the amplitude of echo is inhomogeneous, therefore target echo is being accumulatedTo be weighted processing by certain rule before.

The video integration of radar is carried out between the different repetition periods of same distance unit, therefore in distanceIn door, there are how many range units will carry out how many times accumulation. Although in the accumulation computing of each range unitAmount is little, but all the total operand of range unit video integration is sizable, therefore adopts recursion type accumulationDevice is realized optimum weighting to reduce operand as two-pole filter is similar to, the present invention has adopted bipolar justThe weighting scheme of some filtering is weighted and processes the distance that detects target echo.

By signal processing algorithm of the present invention, realize effective processing of echo-signal and successfully detectedTarget location, meets pulse radar and detects the requirement of the targets such as inshore naval vessel. Effectively detect distanceCanoe is 11.2km, and large ship is 12.73km, all meets the technical requirement of 10 kilometers.

Brief description of the drawings

Fig. 1 is signal processing algorithm flow chart.

Fig. 2 is echo-signal storage schematic diagram.

Fig. 3 is the time domain block diagram of duopole filtering.

Detailed description of the invention

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

Algorithm flow of the present invention as shown in Figure 1, first builds echo-signal, and echo-signal is carried out to pulse pressureProcess, by two-pole filter, carry out video integration and obtain processing signals afterwards, determine target range,Owing to not considering target velocity parameter, do not carry out Doppler and tie up FFT processing. Then carry out CFAR CFARProbability Detection, first passes through by fixed threshold the thresholding that floats again.

1) echo-signal builds

As shown in Figure 2, after obtaining echo-signal, (pulse repeats each PRF the storage mode of signalFrequency) in sampled point be stored in distance to, thereby orientation to for accumulation count form a two-dimensional matrix.This two-dimensional matrix is frame data, if sampling multiframe signal, then this two-dimensional matrix would be extended to three-dimensionalMatrix.

2) pulse pressure processing

Pulse compression, wide while transmitting in radar transmitter, the pulse that peak power is lower, passes throughCarrier frequency is encoded to increase to the bandwidth of transmitted waveform, and then in receiver, echo is implemented to pulse and pressContracting, thus reach burst pulse, resolution capability and spy that the simple pulse radar of high-peak power hasSurvey ability.

3) duopole filtering

Two-pole filter is a second order iir filter, can be similar to and realize optimum weighting, duopole filterThere are two limits in ripple system function, namely shock response can be obtained two and solve wave filter, is oneThe individual ranking operation device that has two feedbacks.

The time domain block diagram of duopole filtering as shown in Figure 3. Can obtain thus its transfer function:

$H\left(z\right)=\frac{z^{-1}}{1-k_1{z}^{-1}+k_2{z}^{-2}}$

Wherein

$K_1=2\exp (-{\mathrm{\ξ\ω}}_d\mathrm{\τ}/\sqrt{1-{\mathrm{\ξ}}^2})cos\left({\mathrm{\ω}}_d\mathrm{\τ}\right)$

$K_2=\exp (-2{\mathrm{\ξ\ω}}_d\mathrm{\τ}/\sqrt{1-{\mathrm{\ξ}}^2})$

The state equation that can draw system according to block diagram is as follows:

y(n)＝x(n)-k₂z(n-1)

z(n)＝y(n-1)+k₁z(n-1)

4) video integration

When the inswept target of radar beam, radar receives a string amplitude and believed by the echo of antenna unwrapping wire figure modulationNumber, in order to make full use of backward energy, improve the detectability of radar, conventionally will be to receiving to obtain multiple arteries and veinsRush in row accumulation. The accumulation completing after detection is video integration. Because the amplitude of echo is inhomogeneous,Therefore before being accumulated, target echo to be weighted processing by certain rule. The video integration of radarBetween the different repetition periods of same distance unit, carry out, therefore in range gate, have how many distances singleUnit will carry out how many times accumulation. Although the accumulation operand at each range unit is little, all distancesIt is sizable that unit video accumulates total operand, therefore adopts recursion type integrator as next in two-pole filterBe similar to and realize optimum weighting to reduce operand.

The formula of video integration is as follows can be obtained fom the above equation:

z(n)＝x(n-1)+k₁z(n-1)-k₂z(n-2)

Echo-signal obtains processing signals by after wave filter, visible processing signals in simulation result afterwardsCan tell accurately target, this algorithm reasonable.

5) CFAR target detection

It is one of important content of Radar Signal Processing that CFAR (invariable false alerting) processes, in automatically detecting,Constant false alarm rate processing can make terminal computer be unlikely to transship because disturbing by force. For clutter map effect is enteredRow improves, and adopts the method for cell-average in this scheme, the data of each clutter unit is averagedProcess:

$X_n(i,j)=\frac{1}{m}\underset{i=1}{\overset{m}{\Σ}}{X}_n(i^{\′},j^{\′})$

In order to set the required threshold level of detecting unit, the interference power level of same unit must be known.During due to interference power level, change, must obtain by data estimation. This method of estimation is based on twoImportant hypothesis:

The statistical property of contained clutter of closing on unit is consistent with unit to be checked

Closing on unit and do not comprise any target, only there is interference noise in it

Base area clutter changes in spatial distribution, reference unit (sliding window size) the general district of numerical value n is at 4-16Between. What in this programme, consider is extra large clutter, in space is conventionally to distribute in flakes, distributes ratio and mix in sheetRipple is wanted evenly, so that reference unit can be obtained is larger, being generally no more than the 50. thresholding factors is oneIndividual scalar, size is determined according to the size of required invariable false alerting. Rayleigh Clutter places an order target detection canReference formula:

1/Pfa＝(1+ratio)^n。

Above-described is only the preferred embodiment of the present invention, the invention is not restricted to above embodiment. CanUnderstand, those skilled in the art directly derive or association without departing from the spirit and concept in the present inventionThe oher improvements and changes that arrive, within all should thinking and being included in protection scope of the present invention.

Claims (1)

1. a signal processing algorithm for land-based radar, is characterized in that, comprises the steps: to obtain backAfter ripple signal, by the sampled point in each echo-signal pulse recurrence frequency be stored in distance to, accumulation is countedBe stored in orientation to, thereby form a two-dimensional matrix; Described two-dimensional matrix is frame data, if adoptedSample multiframe signal, extends to three-dimensional matrice described two-dimensional matrix; Then echo-signal is carried out to pulse pressure placeReason, by two-pole filter, carries out video integration and obtains processing signals afterwards, determines target range, connectsAnd carry out the detection of CFAR invariable false alerting, first pass through again by fixed threshold the thresholding that floats.