Optimal design of low power high gain and high speed CMOS circuits using fish swarm optimization algorithm

This work presents a swarm intelligent optimization methodology based on swarming principle and collective behaviour of natural species termed as fish swarm optimization algorithm (FSOA). It is applied for the optimization for the design of a Complementary Metal Oxide semiconductor (CMOS) two-stage comparator and a folded cascode operational trans-conductance (FCOTA). The basic idea of FSOA is to model the traditional behavior of fish such as preying, swarming followed by local fish of individual optimization for global search. The suitably chosen control parameters of FSOA balance the exploration and exploitation of search space. Results obtained by SPICE for the two analog circuits, justify the capability of producing a desired result and the superiority of FSOA over other evolutionary optimization algorithms like HS (harmonic search), PSO (particle swarm optimization) and DE (differential evolution) in terms of convergence speed and design performances. The results obtained using FSOA show improved performances for the designed analog circuits. The circuits designed using FSOA take lesser areas for MOS transistors; both dissipate low powers and provide high gains. The performances of FSOA based designed circuits are significantly better than those of reported works. Simulation results obtained from SPICE also prove that FSOA is the best in comparison with the previously reported techniques in terms of the area occupied by CMOS transistors, gain, power etc. Cadence version 5.10.41 is used to perform the simulation using TSMC 0.35μm and TSMC 1.25μm technology parameters for two-stage comparator and FCOTA, respectively.