The High-Efficiency Lightweight Optimization Method for Analog-to-Digital Conversion System in IoT

To improve the performance and decrease the design cost, a high-efficiency lightweight optimization method for analog-to-digital conversion (ADC) system in Internet of Things (IoT) is proposed in this paper. According to the circuit structure and the proposed design specifications, the circuit is divided into several sub-module circuits. Then, the two-stage lightweight circuit model is trained based on local convolutional neural network (CNN) and the global -multilayer perceptron (MLP) to predict the circuit performance metrics, which can replace the circuit simulator to decrease simulation time. Based on the established models, the parameters of analog circuit are optimized by the constrained particle swarm optimization (PSO) algorithm. Three typical circuits in ADC system are optimized and verified on Cadence software. The fitting accuracy (R2) of the established models achieves 99%. Compared with advanced model-based optimization method, the dataset and simulation time can be decreased 91% and 95%, and the optimized power consumption can be decreased 17%. Meanwhile, compared with the advanced reinforcement learning (RL)-based optimization method, the optimization time can be decreased 89.6%, and the optimized power consumption can be decreased 29.5% under the same dataset. The results show that the proposed lightweight optimization method can greatly decrease the dataset and simulation time, and improve the optimization efficiency of ADC system in IoT.