Bistatic ULA-MIMO Radar Target Parameters Estimation Methods Based On Quaternion Theory

It has been more than ten years since the concept of MIMO radar was putforward.Due to it’s inherent characteristic such as virtual aperture, spatial diversity,waveform diversity, it is more superior than conventional phased array radar inperformance of the target detection and localization. Therefore, the MIMO radar hasattracted wide attention of scholars both at home and abroad. As an important branchof the centralized MIMO radar, Bistatic MIMO radar is playing an increasinglyimportant role. And it has been a hot topic in the research field of MIMO radar.Theparameter estimation is not only an important goal for the field of bistatic MIMOradar, but also the realization of the key technologies of radar precise targetlocalization. The echo signals of bistatic radar contain receiving Angle informationand launch Angle information, which makes the target parameters of bistatic MIMOradar has a high estimation precision. Therefore it has a broad applicationprospects.This paper researches on the bistatic MIMO radar whose transmiting andreceiving array both are uniform linear arrays.In recent years Quaternion theory has got application in the field of rigid bodymechanics, image processing, pattern recognition, array signal processing. Since thequaternion orthogonal vector has stronger orthogonal constraints than plural, so theparameter estimation algorithm based on quaternion theory has better model errorfault tolerance. By means of Hilbert transform this article construct four orthogonaldata as a quaternion four dimensional orthogonal component based on the model ofthe traditional bistatic ULA-MIMO radar data to discribe the bistatic ULA-MIMOradar quaternion model visually. Then based on the quaternion model, studies thetarget parameter estimation of the bistatic ULA-MIMO radar.The paper studies the MUSIC algorithm for the bistatic ULA-MIMO radar quaternion model. Based on the bistatic ULA-MIMO radar quaternion model,quaternion data matrix is constructed and the MUSIC algorithm based on thequaternion matrix singular value decomposition and the MUSIC algorithm based onthe educing matrix of quaternion matrix singular value decomposition are put forward.The results of simulation esperiment show the effectiveness of the proposed twoalgorithms. Based on the quaternion model self-conjugate quaternion data matrix isconstructed and the MUSIC algorithm based on the self-conjugate quaternion righteigenvalue decomposition and the MUSIC algorithm based on the educing matrix ofthe self-conjugate quaternion eigenvalue decomposition are put forward. The resultsof simulation experiment show the effectiveness of the proposed two algorithms.Finally, performance contrast simulation experiments are done with the proposed fourquaternion MUSIC algorithms. The experimental results show that the MUSICalgorithm based on the educing matrix of quaternion matrix singular valuedecomposition is the most excellent.The paper studies the ESPRIT algorithm for the bistatic ULA-MIMO radarquaternion model. Based on the quaternion data matrix, the ESPRIT algorithm basedon the educing matrix of quaternion matrix singular value decomposition is putforward. The results of simulation experiment show the effectiveness of the proposedalgorithm. Based on the self-conjugate quaternion data matrix, the ESPRIT algorithmbased on the educing matrix of the self-conjugate quaternion matrix eigenvaluedecomposition is put forward. The results of simulation experiment show theeffectiveness of the proposed algorithm. Finally, performance contrast simulationexperiments are done with the proposed two quaternion ESPRIT algorithms. Theexperimental results show that the ESPRIT algorithm based on the educing matrix ofquaternion matrix singular value decomposition is the most excellent.