Performance-Driven Design of Two-Stage CMOS Op-Amp Using Bio-Inspired Dung Beetle Optimization Algorithm

Abstract

As designer the task of sizing analog circuits has become more important due to complex trade-offs in CMOS analog design process. Power use, silicon size, unity-gain bandwidth, slew rate, and open-loop gain are just a few of the performance factors that need to be evaluated together. The problem is obviously multi-objective because there are so many design criteria. Transistor-level equations are used in traditional sizing methods, but these methods often don't find the best global solution because they rely on simplifying assumptions. Researchers have turned to metaheuristic optimization techniques to get around these problems. These techniques work well in design spaces that are very nonlinear. This paper investigates the implementation of the Dung Beetle Optimization (DBO) algorithm, a new bio-inspired approach was implemented for optimizing the circuit of CMOS two-stage operational amplifier of 130nm technology within the SPICE environment. The results show that DBO can lower the area and power use of transistors while still meeting design requirements. DBO consistently achieves more favorable trade-offs than other well-known evolutionary algorithms like Particle Swarm Optimization (PSO), Cuckoo Search (CS), Whale Optimization Algorithm (WOA) and Artificial Bee Colony (ABC). These results show that DBO is not only fast at searching, but also good at balancing conflicting design goals. This makes it a strong candidate for improving analog VLSI design automation.