Research On Several Key Technologies Of Target Detection In Distributed MIMO Radars

In this paper,the signal-fused target detection algorithms for distributed radar systems are researched.A target may have varied spatial scattering characteristics from different observation angles,which can be exploited for a distributed multiple-input multiple-output(MIMO)radar system to obtain the spatial diversity gain and improve the target detection performance.In the applications,there are two important problems to be resolved.One is to reduce the effect of the unfavorable factors on the distributed radar systems,another is to decrease the implementation cost on condition of the stable detection performance.Thus,three key problems for distributed radar systems are emphasized to improve the radar system performance in this paper.Firstly,the effect of the signal-to-noise ratio(SNR)on the performance of the signal-fused detection is analyzed.Then,the signal-fused detection based on the double-threshold is considered to reduce the large amount of transmission data with stable detection performance.Lastly,taking into account the unavoidable registration errors,the effect of the registration errors on the detection performance for distributed radar systems is analyzed and corresponding robust algorithms for the registration errors are proposed.The detailed research work of this paper is as follows:Firstly,the conventional signal-fused detection algorithms for distributed MIMO radars are simply introduced.Firstly,the classical system configuration,signal model and signal processing are given.Based on knowing the prior SNR information or not,conventional likelihood ratio test(LRT)detector and non-coherent integration(NCI)detector are derived,respectively.Compared to the mono-static radar,the improvement of detection performance caused by spatial diversity gain can be illustrated.Then,the detection performance between the LRT detector and the NCI detector with different SNRs are compared and the effect of prior SNR on the detection performance is analyzed.Lastly,the detection performance of the distributed radar systems is assessed in detail by simulation results.Secondly,the effect of the SNR diversity on the signal-fused detection performance of the distributed MIMO radar is studied.For the LRT detector,the perfect detection performance can be obtained due to the prior SNR.In this paper,we focus on the NCI detector.It is straightforward that the detection performance may decrease when the high SNR channels and the low SNR channels are fused together.From simulation results,it can be noted that adding the same low SNR channels may lead to different performance effect under different conditions.Introducing the concept of the cumulative contribution rate(CCR),a robust signal-fused detection algorithm for the diversity is proposed to decrease the effect of the low SNR channels without the knowledge of SNR.Note that the test statistics reflect the SNR.Thus,given the CCR threshold,figuring out the CCR of the local test statistics can increase the accumulation of the high SNR channels.Simulation results show that the proposed method can improve the detection performance of distributed radar systems with the SNR diversity.Thirdly,the effect of the data transmission rate on the signal-fused detection performance is researched.It is essential for the conventional detectors to transfer all the observation data,which leads to a huge communication data and system load.In the applications,most of the radar echo data are noises,and the target signals are sparse.What's more,the target signals have a larger power than the noise.Depending on the above features,the double threshold signal-fused detection algorithms are studied.Firstly,the first thresholds are determined according to the data transmission rate of the spatial diversity channels.Then,the data over the first threshold are transferred to the fusion center.Lastly,the final test statistics are yielded and compared to the second threshold to obtain the detection result.After the process of the first threshold,most of data with low power is eliminated and the data transmission rate can be decreased.Simulation results show that the data transmission rate can be decreased dramatically with a little loss of detection performance for the double threshold detectors which have a great value for engineering application.Lastly,the signal-fused detection algorithm with spatial registration errors is researched.Firstly,the effect of the registration errors on the detection performance of the conventional LRT detector is analyzed,and the explicit expression for the probability of false alarm and the probability of detection are derived.In order to improve efficiently the detection performance with registration errors,a signal-fused detection algorithm based on uncertainty area is proposed.In each spatial diversity channel,the position of the target signal in the uncertainty area is estimated first.Then,the estimated target signals are fused together.Knowing the prior distribution of registration errors or not,the generalized likelihood ratio test detector and the Bayesian likelihood ratio test detector are proposed,respectively.Simulation results illustrate that the registration errors have a great degradation the detection performance,and the proposed methods are robust for the registration errors.