Abstract
Sparse iterative covariance-based estimation (SPICE) method is a popular sparse reconstruction approach with many advantages, such as, stable statistical property and hyper-parameter free. In this paper, we propose a maximum entropy method (MEM) to optimize the weight function which is the kernel of the SPICE algorithm. In the proposed MEM, the weight in SPICE is regarded as the inverse of the estimated energy spectrum. Thus, the beamforming technique is exploited. Using the optimized weight, sources with small interval can also be successfully resolved. Compared with existing SPICE algorithms, the proposed MEM has a better angular resolution capability. Numerical simulations verify the superiority of the proposed MEM.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
