A learnable self-feedback ratio-memory cellular nonlinear network (SRMCNN) for associative memory applications

A self-feedback ratio-memory cellular nonlinear network (SRMCNN) with the B template and modified Hebbian learning algorithm to learn and recognize image patterns is proposed and analyzed. In this SRMCNN, the coefficients of space-variant B templates are determined from the exemplar patterns during the learning period. The weights are the ratio of the absolute summation of its neighborhood weights in the B templates stored in the associative memory. The SRMCNN can recognize the learned patterns with distinct white-black noise and output the correct patterns. Matlab and HSPICE software have simulated the operation of the proposed SRMCNN. It is shown that the 18/spl times/18 SRMCNN can successfully learn and recognize 8 incompletely noisy patterns. As compared to other learnable CNN as associative memories, the proposed SRMCNN could improve pattern learning and recognition capability. The architecture can be implemented in nano-CMOS technology for a giga-scale learning system in real-time applications.