LETTER Communicated by K. K. Tan Analysis and Design of Associative Memories Based on Recurrent Neural Networks with Linear Saturation Activation Functions and Time-Varying Delays

In this letter, some sufficient conditions are obtained to guarantee re-current neural networks with linear saturation activation functions, and time-varying delays havemultiequilibria located in the saturation region and the boundaries of the saturation region. These results on pattern characterization are used to analyze and design autoassociative memo-ries, which are directly based on the parameters of the neural networks. Moreover, a formula for the numbers of spurious equilibria is also de-rived. Four design procedures for recurrent neural networks with linear saturation activation functions and time-varying delays are developed based on stability results. Two of these procedures allow the neural net-work to be capable of learning and forgetting. Finally, simulation results demonstrate the validity and characteristics of the proposed approach.