Optimum synchronous phase detection and its application in smart sensor interfaces

Abstract

Synchronous phase detection is widely used in sensors and telecommunication systems. The resulting SNR depends on the shape of both the input signal and the phase detector's reference. This paper presents a theoretical analysis of optimum synchronous phase detection in terms of SNR. In contrast to a matched filter, the optimum phase reference in the presence of additive white noise is found to be proportional to the derivative of the signal, rather than to the signal itself. Two case studies are discussed: the readout of a thermal-diffusivity sensor (TD sensor) and that of a resistor-based sensor embedded in a Wien bridge (WB sensor). Simulations show that compared to the standard square-wave phase reference, the use of the optimum synchronous phase reference improves the SNR of the TD sensor by 4.2 dB and that of the WB sensor by 6.3 dB. A three-level waveform is proposed as a practical alternative to the optimum phase reference. Its use degrades the optimum SNR by only 1.1 dB and 1.3 dB for the TD and WB sensors, respectively.