Robust Detection and Estimation with Compressive Sensing and Networked MIMO Radars

Current data acquisition systems include radars and general sensors first acquire samples of the signal or image, and then apply compression to reduce the amount of information. As the number of samples increases for higher detection resolution and more accurate information capturing, the processing cost for data compressing also increases. It is also wasteful of resources to acquire and store a large number of data samples which will be discarded later. In addition, the detection and estimation are made hard when there are clutters around the targets or deep fading in the paths of sensing signals especially in an urban environment with many obstacles. The loss and delay of sample data will further reduce the fusion and detection quality.

Two recent techniques, compressive sensing (CS) and multi-input and multi-output (MIMO) radar, have attracted a lot of attentions and show potential in addressing some of these issues.

Compressive sensing technique employs sparse sampling to reduce data redundancy for sparse scenes. Existing work often directly apply the concept of sparse sensing to reduce data used for

radar detection. On the other hand, MIMO radar technique employs multiple antennas to transmit different and often orthogonal waveforms simultaneously to increase degree of freedom for improved detection quality. In this work, we consider use of networked MIMO radars and CS techniques for an overall better detection and estimation performance in different domains.