Adaptive filters realized with nano-scale VLSI circuit technology

This paper investigates design strategies for achieving reliable performance in VLSI adaptive filters that are prone to transient errors due to increasingly smaller feature dimensions and supply voltages of the CMOS circuits. First it will be shown that stochastic search algorithms (e.g. Particle Swarm Optimization) have a natural resistance to transient errors. It is then shown how modular hardware based on residue number system (RNS) coding can be designed to more effectively manage transient errors in adaptive filters with stochastic search algorithms.