Optimized Waveform Design for OFDM-based ISAC Systems Under Limited Resource Occupancy

TL;DR

This paper proposes an optimized OFDM waveform design for 6G ISAC systems that enhances sensing accuracy by efficient resource allocation and a novel matrix completion interpolation technique, outperforming existing methods.

Contribution

It introduces a new interpolation method using Schatten p quasi-normal approximation for improved sidelobe suppression in OFDM-based ISAC systems.

Findings

Enhanced sensing performance with limited resource occupancy

Effective sidelobe suppression via Schatten p matrix completion

Outperforms traditional resource scheduling and interpolation methods

Abstract

The sixth generation (6G) of wireless networks introduces integrated sensing and communication (ISAC), a technology in which communication and sensing functionalities are inextricably linked, sharing resources across time, frequency, space, and energy. Despite its popularity in communication, the orthogonal frequency division multiplexing (OFDM) waveform, while advantageous for communication, has limitations in sensing performance within an ISAC network. This paper delves into OFDM waveform design through optimal resource allocation over time, frequency, and energy, maximizing sensing performance while preserving communication quality. During quasi-normal operation, the Base Station (BS) does not utilize all available time-frequency resources, resulting in high sidelobes in the OFDM waveform's ambiguity function, as well as decreased sensing accuracy. To address these latter issues, the paper proposes a novel interpolation technique using matrix completion through the Schatten p quasi-normal approximation, which requires fewer samples than the traditional nuclear norm for effective matrix completion and interpolation. This approach effectively suppresses the sidelobes, enhancing the sensing performance. Numerical simulations confirm that the proposed method outperforms state-of-the-art frameworks, such as standard complaint resource scheduling and interpolation, particularly in scenarios with limited resource occupancy.