Pulse Modulation VLSI Implementation of Clustering Algorithm Based on Stochastic Association Model

The clustering algorithm employing “stochastic association”, which we have already proposed, offers a simple and efficient soft-max adaptation rule. The adaptation process is the same as the on-line K-means clustering method except for adding random fluctuation in the distortion error evaluation process. This paper describes VLSI implementation of this new clustering algorithm based on a pulse modulation circuit architecture. Random fluctuation is added by a chaos generator circuit that we previously designed using a pulse-width/pulse-phase modulation approach. Experimental results using a fabricated chip have demonstrated that successful “stochastic association” clustering is achieved on a time scale of milliseconds.