The Study And Realization Of Radar Signal Detection And Range Ambiguity Technology

PD radar, with a coherent processing system, occupies an important position in modern radar system. It uses a higher accumulation of the pulses to improve the Signal to Noise Ratio and performance of detection. The signals received by radar include echoes from targets, noise and clutter. And the detection’s task is to distinguish the target from the clutter and measure the range, radial velocity of the targets. The radar detection system is composed of MTI, MTD, CFAR detection, Pulse Accumulation, Range Ambiguity.CFAR detection can provide an adaptive threshold with the changing of background clutter power to control the false alarm rate in a constant range. According to the different grazing angle and resolution, the Rayleigh distribution, log-normal distribution, Weibull distribution, K distribution etc. have been used to imitate the probability density of echoes from grass, trees, buildings, rain, snow, waves and other background clutter. Many kinds of CFAR detection methods has been proposed in rayleigh distribution background, including Mean-Level, Order Statistics and other CFAR processors, of which performance and detection loss on clutter edge or in multiple-target situations are studied in the paper.In PD radar system, range ambiguity resulted by shorter pulse repetition period can be resolved by using different pulse repetition period. First, we can get Congruence equations with the measured range cell on different PRTs, then we can yield the only correct unambiguous range where the ambiguous range for each PRT are irreducible from each other. Resolutions are usually set about from three aspects which are Chinese Remainder Theorem, ordering all the possibilities and assuming and tracking the targets. One-dimensional set algorithm, residues’ difference look-up table method etc. and their calculating amounts and redundancy are analyzed in the paper. Given the processing features of FPGA, one new look-up table of average square error is proposed, which is demonstrated to be of less calculating amount and more redundancy of measure error.In the last part, a detection system adopted by an actual radar system is described. CFAR detector realized on FPGA saves a lot of time and can process data in a flow. Without sequencing data and with the look-up table of averaging square error, the mending one-dimensional set method improves the detection efficiency greatly. The finer thresholds are adopted according to the analysis of actual echo data and superior detection effect is achieved.