A Coordinate Descent Framework to Joint Design of MPSK Sequences and Receive Filter Weights in MIMO Radar Systems

Abstract

In this paper, we aim at a joint design of Mary Phase Shift Keying (PSK) -or MPSK -transmit waveform and receive space-time filter to maximize the Signal to Interference plus Noise Ratio (SINR) in colocated Multiple Input Multiple Output (MIMO) radar systems. The design problem is formulated into a maximization of the SINR, including unimodular discrete phase constraint on the transmit waveforms. The resulting problem is non-convex, whereby we devise an iterative algorithm based on the Coordinate Descent (CD) framework to tackle it iteratively. To reduce the computational complexity in designing the transmit waveforms, we exploit the properties of Fast Fourier Transform (FFT) in this paper. Numerical examples illustrate the superior performance of the proposed algorithm in comparison with the state-ofthe-art counterparts.

Joint Waveform/Receiver Design for Vital-Sign Detection in Signal-Dependent Interference

“…In this section, we deal with the joint design of the radar code and the corresponding receive filter under a practical constraint on the phase of the transmit code sequences. Notice this is a multi-variable non-convex constrained optimization, and the technique we adopt is based on a sequential optimization procedure, iteratively optimizing the SINR [32]. Specifically, starting from an admissible radar code s (0) , we design the receive filter w (0) which maximizes the SINR corresponding to the transmitted sequence s (0) .…”

Joint Design of Colocated MIMO Radar Constant Envelope Waveform and Receive Filter to Reduce SINR Loss

“…Since the objective function is usually a quadratic function, and the constraints imposed on the waveform are usually non-convex [ 1 , 2 , 3 , 5 , 7 , 11 , 16 , 17 , 24 , 30 , 31 , 32 , 33 , 34 ], it is difficult to solve the optimization problem [ 31 ]. Various methods [ 3 , 11 , 14 , 16 , 23 , 31 , 35 , 36 , 37 ] are used to solve the non-convex optimization problem. Feraidooni and Gharavian developed in [ 31 ] an algorithm for the joint design of a continuous/discrete [ 33 , 34 , 36 ] phase sequence and space-time receive filter to improve SINR, using the coordinate descent framework to deal with the constrained non-convex problem.…”

“…Various methods [ 3 , 11 , 14 , 16 , 23 , 31 , 35 , 36 , 37 ] are used to solve the non-convex optimization problem. Feraidooni and Gharavian developed in [ 31 ] an algorithm for the joint design of a continuous/discrete [ 33 , 34 , 36 ] phase sequence and space-time receive filter to improve SINR, using the coordinate descent framework to deal with the constrained non-convex problem. Also, another joint design method is designed in [ 3 ].…”

“…However, in [ 14 , 19 , 20 , 23 , 24 , 29 ], the waveform optimization problem established adopts energy constraint, which cannot give full play to the amplifier’s maximum performance and reduce the output SINR. In [ 1 , 14 , 15 , 17 , 20 , 21 , 23 , 25 , 27 , 33 , 34 , 36 ], etc., pulse compression and ambiguity characteristics of the waveform are not taken into account, and the solution of [ 14 , 19 , 23 ] cannot guarantee global optimization. Therefore, the performance of the system can be improved through the design of the joint transmit waveform and receive filter based on semidefinite programming.…”

“…Moreover, the designed waveform usually imposes some constraints on the waveform in actual engineering applications, for example, constant envelope constraint [ 3 , 8 , 9 , 10 , 11 , 12 , 16 , 18 , 21 , 27 , 30 , 32 , 34 , 35 , 36 , 38 , 50 ], PAR constraint [ 8 , 9 , 19 , 35 , 47 ], similarity constraint [ 11 , 16 , 29 , 43 , 50 , 51 ] or sidelobe constraint [ 15 , 21 , 22 , 30 , 32 , 33 , 38 , 47 , 50 ]. The constant envelope constraint is more stringent than the PAR constraint because most radar transmitters use nonlinear power amplifiers, which usually work in a saturated state, and the amplitude of the waveform is required to be constant.…”