Automation in nominal design of analog circuits considerably reduces the overall time to market of mixed-mode ICs. Application of meta-heuristic optimization algorithms with simulator level fitness evaluation for circuit sizing is a popular automation approach. In this work, with the same approach, we investigate the performance of Harmony Search (HS) algorithm. HS is a relatively new meta-heuristic inspired from musical process, in which population represents memory of a group of musicians, that searches for a perfect state of harmony. Performance of HS is compared with Differential Evolution (DE) algorithm for design of a 180nm CMOS process based Folded Cascode Operational Trans-conductance Amplifier (FCOTA) with continuous time common mode feedback. Results show that, at initial period of iterations, HS converges quickly but fails to converge to the global optimum as compared to DE algorithm.
展开▼
