An Evolutionary Synthesis Algorithm to Design Optimum Performance CMOS RFSSCAs

An automated synthesis procedure, based on genetic algorithms, to design optimum performance Si-integrated radio-frequency and microwave binary-weighted single-ended switched capacitor arrays (RFSSCAs), is presented in this paper. The method determines the RFSSCA circuit size and the circuit components values from the top-level system specifications. The genetic synthesis tool optimizes a fitness function based on user-specified performance constraints. Besides that, the algorithm uses a sharing scheme in order to achieve a set of optimal solutions. The solutions set, obtained by the genetic algorithm, give to the designer the liberty of choosing several possible implementations maintaining the performance objective. To confirm the circuit solutions obtained by the algorithm, two RFSSCAs are synthesized in a 2-poly 4-metal 0.35 ¿m standard CMOS technology. The circuit is then simulated in SpectreRF. Since the synthesized results are in very good agreement with the simulated outcomes, it can be stated that the proposed genetic synthesis method is a very useful tool to design optimum performance RFSSCAs.