A dynamic K-winners-take-all analog neural circuit

Continuous time mathematical model of K-winners-take-all (KWTA) neural circuit which can quickly identify the K-winning from N neurons, where 1 les, K <N, whose input signals are larger than of remaining N - K neurons, are proposed. A functional block diagram of corresponding analog neural circuit is presented. For N competitors, such circuit is composed of N feedforward and one feedback hardlimiting neurons, which is used to determine the dynamic shift of input signals. There has been shown that the proposed circuit is simpler and presents more high speed of signal processing than existing analogs.