Near Field Active Imaging Using Sparse Arrays

Sensor arrays designed for far field operation may experience performance degradation when imaging near field objects. Specifically, sparse active arrays utilizing the additional degrees of freedom provided by the sum co-array are susceptible to these effects, as the co-array depends on both the range and direction of scatterers close to the array. Consequently, a uniform far field sum co-array may become non-uniform in the near field. As a result, co-array processing algorithms, such as image addition, are subject to undesired grating lobes in the presence of near field scatterers. This paper proposes an extension to image addition for mitigating such undesired distortions. The method compensates for near field effects by computing spatially varying transmit and receive element weights. These weights minimize the discrepancy between the desired and achieved near field point spread function, while using as few image addition components as possible. Given a desired point spread function and a set of calibration measurements of the near field array steering vectors, a regularized convex optimization problem is then solved for each pixel of the image.