MEMS microphones offer a significant scope to improve miniaturization, integration and cost of acoustic systems, poised to be the preferred microphone option for consumer electronics and medical advancements. A MEMS microphone needs a readout interface to convert the microphone’s output to a digital code for further processing, while its poor driving ability poses a challenge on the design of readout ADCs.

In this thesis, the theory and implementation of a high input impedance continuous-time sigma-delta modulator (SDM) for a MEMS microphone readout is presented. A pseudo-virtual ground feedforward structure is used to eliminate the internal feedback DAC and contribute to enhanced linearization. To meet the requirement of high input impedance, a Gm-C first integrator is employed, featuring a resistive source degeneration structure and a local Gm-boosting loop to enhance the linearity of the first stage. For the second stage, a VCO-based integrator and quantizer are employed, offering advantages including inherent multilevel quantization and intrinsic clock-level averaging (CLA). The second order SDM consumes an estimated power of 57μW, achieving an 83dB SNR and a 79dB SNDR in simulation, reflecting its efficiency in audio applications.