Active Sensing Based on 5G-NR Single-Numerology Filtered OFDM Signals

Abstract

Integrated sensing and communication (ISAC) has been actively researched for the next generation mobile network, given the plethora of new applications involving sensing functionalities in, such as, Internet of Things (IoT), vehicle-to-everything (V2X) communications, autonomous driving, human activity sensing, and unmanned aerial vehicle (UAV) networks. Given that filtered orthogonal frequency division multiplexing (F-OFDM) is a waveform in 5G New Radio (NR) and 6G systems for enhancing spectral efficiency by applying subband filtering to reduce out-of-band emissions, in this paper we analyze the ISAC performance when sensing using F-OFDM signals. Firstly, we detail the F-OFDM signaling used for an active sensing scenario. Secondly, to achieve more accurate estimation of velocity and range of target, we introduce a Time Domain (TD) matched filter sensing method, where cyclic prefix (CP) is utilized in the correlation operation. Thirdly, to further improve the Frequency Domain (FD) sensing resolution, we propose to invoke Multiple Classification (MUSIC) algorithm for fractional delay-Doppler (DD) indices detection. Finally, to reduce the high computational complexity of the TD matched filter sensing, we propose a hybrid Fast Fourier Transform (FFT)-assisted matched filter sensing algorithm. Simulation results show that the TD sensing based on the matched filter algorithm, including our hybrid FFT-assisted matched filter sensing algorithm outperforms the FD sensing in low Signal-to-Noise Ratio (SNR) region. On the other hand, in FD sensing, the MUSIC algorithm outperforms conventional FFT based method. Our studies and results demonstrate that the F-OFDM signals have the potential for the implementation of active communication-centric ISAC, which employ the flexibility for sensing implementation in either TD or FD, or in the joint frequency-time domains.