Multi-target Localization With MIMO Radar Via Coupled Canonical Polyadic Decomposition

Multiple-input Multiple-output(MIMO)radars are widely used in target tracking and target localization.Among them,the multistatic MIMO radar can achieve more accurate target localization.The coupled tensor-based method is more suitable for multistatic MIMO radar multi-target localization than traditional methods because of its unique high-dimensional computing advantages.In this paper,the coupled tensor decomposition algorithm is applied to the multi-target location of multistatic MIMO radar,aiming to more effectively use the coupling relationship between multiple received signals,so as to calculate the targets position more accurately.This paper studies the multi-target location method of multistatic MIMO radar based on Coupled Canonical Polyadic Decomposition(C-CPD).A multistatic MIMO radar composed of a single transmitting array and multiple receiving arrays is considered,and it is assumed that the transmitting array periodically transmits signals in multiple consecutive pulse periods.This system is referred to as a Single-Transmit Multiple-Receive Multiple-Pulse(ST-MR-MP)MIMO radar system.The article divides the multi-target location problem of ST-MR-MP MIMO radar into the following three scenarios:Over-determined scenario: At least two conditions are met: a larger transmitting array,a larger receiving array,and targets moves slowly.That is,at least two of the number of antennas in the transmitting array,the number of antennas in each receiving array,and the number of pulse cycles of the transmitted signal are not less than the target number;Under-determined scenario I: The scale of the transmitting array is large,the scale of each receiving array is small,and the target moves fast.That is,the number of transmitting antennas is not less than the target number,and the number of receiving antennas and the number of pulse cycles of the transmitted signal are less than the target number;Under-determined scenario II: The scale of the transmitting array is small,and at the same time,the scale of each receiving array is small or the target moves faster.That is,the number of transmitting antennas is less than the target number,and one of the number of receiving antennas and the number of pulse cycles is less than the target number.The uniqueness condition and algorithm of the C-CPD-based MIMO radar multi-target localization method in three scenarios are studied separately.The method described in this article has many advantages when applied to MIMO radar target positioning.For example,the transmitted signals do not need to be orthogonal to each other;only DOA information is required;and the target positioning can be accurately completed in underdetermined scenarios.The simulation experiment proved the advantages of the method based on the C-CPD algorithm in the accuracy of ST-MR-MP MIMO radar target positioning.In the underdetermined scenario,compared with other algorithms,the advantages of the C-CPD algorithm are more obvious.Among them,the optimized C-CPD algorithm that uses the decomposition result of the algebraic C-CPD as the initial value has higher target position estimation accuracy.