Radar target detection is determined by the energy received from the target and compared with the energy of background noise. The radar range equation accounts for the signal-to-noise ratio (SNR) due to transmitter, return path, receiver, integration, losses, and radar cross sections of targets. Frequency Modulated Continuous Wave (FMCW) radars are effective in distinguishing between moving targets and clutter. However, a weak target in the presence strong clutter can be easily overwhelmed, especially when the target is slow moving. In addition, a slow moving target can be undetected in the presence of wind-blown foliage. Wind-blown foliage can contribute to Doppler shifts caused by movements of branches and leaves, which can be challenging in target detection. In this paper, we will discuss clutter mitigation caused by wind-blown foliage and clutter mitigation with slow-moving targets. Traditional approaches, such a pulse canceler essentially is a low-pass filter is designed to remove slow moving clutter and is not effective in mitigating foliage clutter during windy conditions. In this paper, we introduce a method to pre-process radar returns with a wavelet transform. The wavelet transform produces subband channels that are progressively smaller which can reduce the order of operations. The subband channels will further be processed with coherent integration and coherent subtraction to mitigate strong clutter introduced by stationary objects that close to a target. We will also investigate the mitigation of clutter due to wind-blown foliage using subband channels to estimate covariance, and extract singular value decompositions. A detection of wind-blown clutter is kept track in temporal bookkeeping, called a clutter map. The entry in the clutter map is deleted when clutter is not present or the expiration of the entry is reached.
