The study of analog evolvable Chebyshev transconductance filters

An evolvable hardware structure and design method on an analogue evolvable trans-conductance filter was presented. Its own technical parameters could be changed with external environmentpsilas real-time changing. The filter has the very good adaptive ability and certain fault-tolerant ability. An improved parallel genetic algorithm with migration strategy based on parallel crossover migration was applied. Selection strategy used an improved expected value model. Crossover strategy selected AND/OR crossover logic operators. Mutation strategy selected dual exclusive NOR/exclusive OR logic operators. Probability of crossover and mutation used adaptive probability which could be adjusted dynamically by reflecting diversity degrees of a population. A multi-objective optimization fitness function for the filterpsilas parameters had been built. Speed of resource allocation and quality of solutions is improved greatly by improved parallel genetic algorithm. The in-system programmability programmable analog circuits (ispPAC10) were used. The four-order Chebyshev trans-conductance filter is built by cascade method, which is able to meet performance request in its stop-band, passing band and transition band. Its practicality had been simulated and verified. Values of evolutionary parameters are in line with theoretical value exceedingly. Simulation results are satisfying.