Low power bulk-driven OTA design optimization using cuckoo search algorithm

The design of CMOS analog integrated circuits presents low level of automation and still depends on the experience of the designer. The sizing procedure for low power applications is challenging, since transistors must be biased in the sub-threshold region, whose modeling is complex and few intuitive if second order effects in sub-micron technologies are considered. This paper presents an automatic design methodology for a 0.25 V bulk-driven Miller OTA in 130 nm CMOS using bioinspired cuckoo search optimization technique implemented in the UCAF tool, which is able to explore efficiently the design space in all transistor operating regions, particularly weak inversion region. The methodology also includes the estimation of circuit performance under process parameter variation. The resulting sized circuit presented better performance if compared to a manual design, achieving a power consumption of 8.47 nW and a gain-bandwidth product of 1.61 kHz. We demonstrate in this work that an efficient search in the design space with an analog CAD tool can result in optimized circuits that demand transistors operating in sub-threshold region.