Multi-static Passive Coherent Location In Single Frequency Network

In modern wars,traditional active radar faces severe challenges with the development of stealth technology,low altitude penetration,electromagnetic interference and anti radiation missiles.The multi-static passive coherent location(PCL)system uses the direct radiation signal and the target reflection echo signal of the transmitters to locate the target by coherent processing.The PCL system has many advantages,such as stealth,anti-stealth capabilities and anti interference ability,and attracts attentions in radar research field.When the DAB/DVB-T is used as illuminator of opportunity(transmitter),the transmitter work in the single frequency network(SFN).Although the SFN has advantages of frequency savings and high spectrum utilization in the field of communication,it brings new challenges for PCL.For example,the three dimensional(3-D)data association of “target-transmitter-measurement” must be taken into account.In order to solve this problem,this thesis focuses on the follows issues:1.In order to track multiple dim targets with a multi-static PCL system,an energy accumulation fast polar Hough transform(EA-FPHT)is proposed.The EA-FPHT method estimates the target state trajectory directly from the original measurement data space.Thus the target tracking problem is changed into a peak accumulation problem in the parameter space.The proposed method reduces the computational complexity and reduces the probability of false tracks acceptance.2.Furthermore,a maximum likelihood probabilistic multi-hypothesis algorithm is proposed to decrease the track fragmentary degree with the SFN.For a single target,the 3-D data association problem above mentioned is degraded into a 2-D one,in which only the association between the transmitters and the measurements are considered.In the proposed method,the log-likelihood function(LLR)is established by accumulating the multi-scan measurements.Then,the target state is estimated by optimizing the LLR.Simulation results show the effectiveness of the proposed method,also the fragmentation of track is reduced.3.In order to initialize tracks for dim targets(where target number is unknown)with passive coherent location system in SFN,a modified maximum likelihood probabilistic multi-hypothesis algorithm is proposed.First,based on the hypothetical target number,multiple super-targets and the corresponding LLR functions are established.Then,the quasi-Monte Carlo search algorithm is used to maximize all LLR functions to simultaneously estimate the number and the state of the targets.Finally,the posterior Cramér-Rao lower bound(PCRLB)in the presence of clutter is derived,the track validity test is carried out and the influence of error parameters on the performance of the algorithm is analyzed.