Multistage Adaptive Pulse Compression

Abstract

Abstract-Many modern radar systems employ pulse compression to maximize the energy on target while maintaining high range resolution. For a solitary point target in white noise, employing a matched filter on receive will maximize the target signal-to-noise ratio (SNR) at the output of the receiver. The matched filter itself is a time-reversed version of the transmitted waveform which is convolved with the received time series to pulse compress the data. A drawback to the matched filter receiver is the range sidelobes which extend on either side of the point target and may mask another weaker target. To reduce range sidelobes after pulse compression, novel adaptive pulse compression techniques have been developed.One such technique is the Reiterative Minimum Mean Square Error Adaptive Pulse Compression (RMMSE-APC) algorithm. This algorithm employs an optimal compression filter at each range bin and significantly reduces the range sidelobes in the vicinity of large targets. In this paper, a pulse compression filter with output identical to the RMMSE filter is derived by employing a multi-stage decomposition of the Wiener filter. A reduced rank version of the Multi-Stage Wiener Filter (MSWF) with lower computational complexity can be created by pruning the number of stages in the decomposition. I. BACKGROUNDThe signal received at the radar can be represented as a convolution of the transmitted signal and the impulse response of the range profile illuminated by the radar [1,2]. Let the vector s denote N discrete samples of the transmitted waveformT , and letT represent a length N portion of the illuminated range profile for l = 0, 1, …, L-1, where l is a discrete delay index and L is the number of range cells of interest. Then the signal received at the radar iswhere v(l) is an N-by-1 vector of additive noise samples. The noise is assumed to be temporally and spatially white and uncorrelated with the transmitted signal samples s(k). The matrix A is written as ( )