Sensing-Aided Distortion Estimation for OFDM Radar With Nonlinear Transmitter

Abstract

This article presents a method to estimate nonlinear distortion of an orthogonal frequency-division multiplexing (OFDM) radar signal by using detected target parameters. Since high transmitting power is desirable for OFDM radar to have a long detection range, the transmitter (TX) is preferred to work in a nonlinear region for high power efficiency. This causes strong distortions of the OFDM radar signals, which have a high peak-to-average power ratio (PAPR). Conventionally, this distortion can be compensated by utilizing equalization at the receiver (RX) or digital predistortion (DPD) at the TX. However, both approaches require information on the transmitted signals obtained through an additional feedback path, which increases the hardware complexity of the radar system. To address this issue, a sensing-aided distortion estimation (SADE) is proposed to estimate the distorted OFDM signals. Initially, radar processing is performed on the received signals with the prior known undistorted symbols. This allows detection of some initial targets in the range-Doppler (RD) domain. Once the target parameters are detected, the distorted symbols can be estimated through division of the received signals by the calculated target signals. This approach leverages the initial target sensing as a feedback loop between the TX and RX. This allows estimation of the distorted OFDM symbols without any additional hardware. The radar processing for subsequent targets demodulates the received signals by using the estimated distorted symbols.