Readout Interfaces For MEMS Capacitive Microphones

In this thesis,two chip design schemes are proposed for the readout interface circuit of MEMS Capacitive Microphone(MCM)sensor: analog microphone system and digital microphone system.At present,smart devices based on voice for human-computer interaction,such as traditional smart phones and emerging smart audio,are rapidly developing.The demand for voice pick-up chips is also growing exponentially.As a new type of acoustic sensor,MCM has the advantages of miniaturization,high reliability,and easy integration,making it an irreversible replacement for the electret microphone(ECM).Therefore,it becomes important to design an application-specific integrated circuit(ASIC)for it.The basic characteristics of MCM are analyzed and studied in this thesis.The MCM sensor senses the sound pressure and causes a weak capacitance change to form the output signal,the output of the MCM sensor is converted into an electrical signal(analog or digital)by readout interface circuit.Because the signal is very weak,to achieve a high signal-to-noise ratio,high requirements are placed on the noise processing in readout circuit design.In this thesis,the specific circuits in the two design schemes are introduced in detail,and a detailed analysis of noise processing and low noise implementation in the circuit is given.The analog microphone system captures the weak electrical signal of the MCM sensor and amplified by the preamplifier,and then outputs the processed analog signal.This thesis gives a detailed introduction and simulation analysis of the main sub-modules,which involves implementation of low-voltage structure,realization of interface bias scheme,noise processing of signal chain circuit.The chip has been taped out using Dongbu Hi Tek 0.18μm BCD processing and tested.The test results:(1)in the condition of sound pressure 1pa frequency 1KHz,signal-noise ratio is typically 62 d B,(2)The maximum value of the chip's output impedance is less than 50Ω,(3)sensitivity range for preamplifier gain of 20 d B at 1KHz:-20-24 d B,(4)Total harmonic distortion plus noise is less than 0.5%,(5)supply voltage range(1.5-5V),(6)the power consumption is 210 μA.The digital microphone system uses a fully differential structure amplifier to read out the weak electrical signal of the MCM sensor and amplify it;the Sigma-Delta ADC is then used to convert the analog signal into a pulse density modulation(PDM)data stream.In this thesis,the bias method of the full-differential readout interface,low-noise preamplifier,and fourth-order high-accuracy sigma-delta ADC are detailed introduced and simulated.