Design and modeling of a continuous-time delta-sigma modulator for biopotential signal acquisition: Simulink vs. Verilog-AMS perspective

Abstract

In the current trend of short time-to-market and complex circuits and systems containing billions of nanoscale transistor, fast and accurate time-domain simulations are crucial for analog and mixed-signal (AMS) design and verification. This will ensure reduction in the non-recurrent cost and make electronics cheaper. In this paper, in order to investigate the options for fast and accurate simulations, two popular modeling tools and languages (Simulink and Verilog-AMS) capable of constructing behavioral models are evaluated. A delta-sigma modulator design with biomedical applications is used as a case study. The system-level design of a third-order, feedforward continuous-time (CT) delta-sigma modulator (DSM) with a signal-to-noise ratio (SNR) of 87.3 dB and 20 kHz input bandwidth is presented. This CT DSM is to be employed in an analog-to-digital converter (ADC) targeting several portable biomedical applications which require a 10 kHz signal bandwidth and higher than 10-bit resolution. Simulink and Verilog-AMS were used throughout the design. The efficiency are compared in terms of modeling effort, simulation performance, and accuracy.