Integrating Low-Complexity and Flexible Sensing into Communication Systems

Abstract

Integrating sensing into standardized communication systems can potentially benefit many consumer applications that require both radio frequency functions. However, without an effective sensing method, such integration may not achieve the expected gains of cost and energy efficiency. Existing sensing methods, which use communication payload signals, either have limited sensing performance or suffer from high complexity. In this paper, we develop a novel and flexible sensing framework which has a complexity only dominated by a Fourier transform and also provides the flexibility in adapting for different sensing needs. We propose to segment a whole block of echo signal evenly into sub-blocks; adjacent ones are allowed to overlap. We design a virtual cyclic prefix (VCP) for each sub-block that allows us to employ two common ways of removing communication data symbols and generate two types of range-Doppler maps (RDMs) for sensing. We perform a comprehensive analysis of the signal components in the RDMs, proving that their interferenceplus-noise (IN) terms are approximately Gaussian distributed. The statistical properties of the distributions are derived, which leads to the analytical comparisons between the two RDMs as well as between the prior and our sensing methods. Moreover, the impact of the lengths of sub-block, VCP and overlapping signal on sensing performance is analyzed. Criteria for designing these lengths for better sensing performance are also provided. Extensive simulations validate the superiority of the proposed sensing framework over prior methods in terms of signal-to-IN ratios in RDMs, detecting performance and flexibility.

Integrated Sensing and Communication Signals Toward 5G-A and 6G: A Survey

“…The signal processing method based on correlation method adopts a virtual cyclic prefix (VCP) according to the sampling points [84]. The maximum detection distance is related to the length of VCP.…”

“…In this case, if a 2D FFT is applied, the sampling value at the RX contains noise. Direct division leads to noise amplification that affects the accuracy of distance and velocity estimation [84].…”

“…In 2011, Sturm et al propose a radar signal processing algorithm applying 2D FFT to the received modulation symbol matrix, which simplifies radar signal processing [83]. In order to improve the sensing accuracy and improve the anti-noise capability with low SNR, the cyclic correlation method and Bartlett's method are proposed respectively [84], [85]. The cyclic correlation method is also applicable to ISAC signals using OTFS.…”

OTFS-Based Joint Communication and Sensing for Future Industrial IoT

“…As PWP is performed in the frequency domain, taking the IDFT of the PWP result leads to the cyclic cross-correlation of the corresponding time-domain signals, similar to the matched filtering in conventional radar processing. The SINRs in the range-Doppler maps (RDM) generated by PWD and PWP have interesting relations, as revealed in [28]. In particular, in low SNR regions, the PWP-SINR is higher than PWD-SINR, while the relation reverses in high SNR regions.…”

“…In particular, in low SNR regions, the PWP-SINR is higher than PWD-SINR, while the relation reverses in high SNR regions. Interested readers are referred to [28] for more details.…”