A multiple input-queued ATM switching fabric based on hopfield neural network cell scheduling

Abstract

A multiple input-queued ATM switching fabrics (ASF) for scheduling cell based on Hopfield neural network (HNN) is proposed. This scheme eliminates degeneration of performance due to head-of-line (HOL) blocking. Simulation results show that, compared with the single first in_first out and window input-queued ASF, the pr oposed approach can greatly improve the throughput and reduce the cell delay.