Research On Array Direction Finding Technique Using The Non-Circular Property Of Signals

Modulated signals with non-circular property are often used in modern radar and communication systems,whose pseudo covariance matrixes also contain information in the array direction finding fields.Therefore,if the non-circular property of signals is properly used to increase the effective information,the estimation performance of array direction finding algorithms can be enhanced.As a result,those array direction finding algorithms which characterized by making use of the non-circular property have been a research focus recently.The existing array direction finding algorithms mainly focused on the case which only non-circular signals of maximum non-circular rate exist.However when circular and non-circular signals coexist in the received waveforms,those algorithms can't estimate the number of signals accurately,nor can they estimate with high resolution or precision.In this paper,array direction finding algorithms making use of non-circular property of signals are studied thoroughly,and the main innovation points are listed below:1.The disadvantages and limitations of those existing source number estimation algorithms have been analyzed,and then an algorithm called CEP based on the projection of eigenspace is proposed which can estimation the number of sources on the background of colored noise.Then apply it to the virtual extended matrix,and further an algorithm based on means clustering,called MCEP,which can estimate the equivalent number of sources under the coexistence of circular and non-circular signals in the received waveforms is proposed.Compared to the existing algorithms,MCEP has better estimation performance in any SNR or snapshots.Moreover,the estimation by MCEP can still be effective even in the background of colored noise.2.Aimed at the problem of the estimation of DOA parameters,the reasons of poor resolution and pseudo peaks are analyzed under the coexistence of circular signals,non-circular signals of average non-circular rate and those of maximum non-circular rate,and then a resolution-enhancing algorithm called HRNC-MUSIC is proposed,which can estimate different kinds of signals in turns,thus decreasing their mutual influence and improving the resolution.Simulation results show that when the spatial angular distances between different kinds of incident signals are comparatively short,the resolution capacity and estimation accuracy of HRNC-MUSIC is apparently improved.3.Aimed at the problem of large amount of calculation and lack of the ability for array extension with regard to polarization sensitive array,DRL-MUSIC algorithm based on planar dual polarization sensitive array is proposed.Furthermore,direction-finding algorithms which can simultaneously make use of non-circular property and polarization information are studied.Then QNC-MUSIC algorithm based on quaternion and DRLNC-MUSIC algorithm based on dimension-reduced long vector are presented,which fills the blank of direction-finding algorithms suited for non-circular signals received in the polarization sensitive array.Simulation results indicate that the resolution of QNC-MUSIC is better than that of DRLNC-MUSIC on certain conditions,while estimation accuracy of DRLNC-MUSIC is superior to that of QNC-MUSIC and DRLNC-MUSIC can jointly estimate polarization parameters simultaneously with no extra calculation.