Research On Multi-Target Detection And Tracking Of The Sky-Wave OTHR

The sky-wave over-the-horizon radar(OTHR), using the electromagnetic wave propagation in the ionosphere to realize target detection, has a special way of working and special frequency band, owning the following characteristics, such as the detection distance is far and the cover of range is very wide, effective ways of detecting stealth aircraft, and strong anti-electromagnetic interference capacity. Anyhow, the sky-wave OTHR is the most economical and effective means of realization of distant early warning, which has important strategic warning value.The multi-target detection and tracking of sky-wave OTHR is mainly aimed at strategic weapons. However, in practice, the sky-wave OTHR suffers the ionospheric contamination, the presence of multipath effects and the strong sea clutter, making target detection face a bad environment with weak goals, multi-target, low signal-to-noise ratio and so on. Now target detection is facing more and more great difficulties and challenge.First, the paper used constant false alarm rate(CFAR) detection technology to detect radar target on the Doppler frequency domain, which can ensure detection performance under the condition of unknown targets. But conventional CFAR parametric model need to master the distribution characteristics of background environment, so before target detection we need to analyze the distribution of the measured data with the Log-normal or Weibull distribution, the paper addressed the iterative-elimination log-t CFAR detection algorithm. The algorithm can detect targets effectively used in non-uniform and sea clutter is dominant on the Doppler velocity spectrum. Appling this algorithm to VC++, the paper designed a program interface, which is efficient on detecting and outputting all the test results, demonstrating the proposed algorithm in a large number of measured data.Next, the received echo data of sky-wave OTHR is spread through the ionosphere medium, namely when the state of the ionosphere changes fast, the propagation path may change either, which can produce fast phase modulation, causing first-order Bragg peak split into multiple distinguishable peaks, we could not make a distinction whether it is target or not. When testing it will output a large number of measured values, and make the true target detection confused. The paper put forward to using the high-order ambiguity function(HAF) on the mechanism of different structures between multi-objective pollution and phase modulation spectrum, make it as the discrimination authenticity of a multi-objective.Finally, the sky-wave radar exist multipath effect, there will be a lot of uncertainty relationship between the multiple targets state and measurements, so the traditional data association algorithm will lead to large amount of calculation, complicated solving problem. Applying the PHD algorithm to OTHR’s multi-target tracking,combined the target motion equation with sky-wave radar measurement equation, the PHD algorithm simulation in the framework of particle filtering verified the tracking simulation realization.