Online Phase-Noise Estimation in FMCW Radar Transceivers Using an Artificial On-Chip Target

Abstract

The presence of phase noise (PN) severely deteriorates the sensitivity and range of frequency-modulated continuous wave radar systems. Thus, characterization and continuous monitoring of the PN is indispensable, and its estimation is often employed directly on chip. Although many contributions investigate methods for PN estimation, almost all of them consider a continuous-wave (CW) input signal. Differently, in this paper, we aim to estimate the PN from a linear frequency modulated continuous wave (FMCW) signal. For that, we propose two methods utilizing a so-called artificial on-chip target and further digital signal processing. These methods are evaluated in real-time and are the first-known solutions to determine PN estimates during the operation of an FMCW radar transceiver. We prove our methods with both simulation and measurement results from a hardware prototype. Further, we present techniques to efficiently realize the concepts in digital hardware. Finally, we compare and trade off the proposed methods against computational complexity and performance. The novel techniques may be applied to arbitrary input frequencies and bandwidths, and do not require a reference-clock input.