PaperlayerView exact revision on arXiv

Source-linked AI summary

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

Zhiqing Wei, Hanyang Qu, Yuan Wang, Xin Yuan, Huici Wu, Ying Du, Kaifeng Han, Ning Zhang, Zhiyong Feng

arXiv:2301.03857v3cs.ITeess.SP

TL;DR

ISAC signals are foundational to integrating communication and radar sensing, but their design, processing, and optimization for 5G-A and 6G have not been thoroughly reviewed. This article synthesizes ISAC signal designs, processing algorithms, and optimization methods from the perspective of mobile communication systems, providing guidelines for future research.

  • Problem

    ISAC signals directly affect sensing and communication, yet their design, processing, and optimization from the 5G-A and 6G mobile-communication perspective have not been thoroughly reviewed.

  • Method

    The article systematically reviews ISAC signals across 5G, 5G-A, and 6G, covering signal design, radar signal processing algorithms, and signal optimization.

  • Results

    The review organizes ISAC signal research around modulation designs, channel information matrix, spectrum lines estimator and super resolution methods, and PAPR, interference-management and adaptive optimization techniques.

  • Takeaways & Limitations

    The review may provide guidelines for researchers designing flexible and reconfigurable ISAC signals for varied 5G-A and 6G scenarios.

Abstract

from arXiv · show

Integrated sensing and communication (ISAC) has the advantages of efficient spectrum utilization and low hardware cost. It is promising to be implemented in the fifth-generation-advanced (5G-A) and sixth-generation (6G) mobile communication systems, having the potential to be applied in intelligent applications requiring both communication and high-accurate sensing capabilities. As the fundamental technology of ISAC, ISAC signal directly impacts the performance of sensing and communication. This article systematically reviews the literature on ISAC signals from the perspective of mobile communication systems, including ISAC signal design, ISAC signal processing algorithms and ISAC signal optimization. We first review the ISAC signal design based on 5G, 5G-A and 6G mobile communication systems. Then, radar signal processing methods are reviewed for ISAC signals, mainly including the channel information matrix method, spectrum lines estimator method and super resolution method. In terms of signal optimization, we summarize peak-to-average power ratio (PAPR) optimization, interference management, and adaptive signal optimization for ISAC signals. This article may provide the guidelines for the research of ISAC signals in 5G-A and 6G mobile communication systems.

I. INTRODUCTION

ISAC unifies wireless communication and radar sensing to improve spectrum utilization and reduce hardware cost, addressing emerging 5G-A and 6G applications. This article reviews ISAC signals through their design, processing, and optimization in mobile communication systems.

  • Emerging 5G-A and 6G applications require high-rate communication and high-accuracy sensing, while separated systems cannot efficiently provide both under scarce spectrum resources.
  • ISAC integrates wireless communication and radar sensing, improving spectrum utilization and reducing equipment size and energy consumption.
  • ISAC signal design: ISAC signal design must balance communication requirements, including spectrum efficiency and anti-interference capability, with radar requirements such as bandwidth, autocorrelation, dynamic range, and Doppler shift.
  • ISAC signal processing and optimization: ISAC signal processing requires high-accuracy, low-complexity algorithms under limited computation resources, while signal optimization must adapt to diverse sensing and communication scenarios.
  • ISAC signal design: The article reviews ISAC signal design for 5G, 5G-A, and 6G systems, including OFDM, FBMC, GFDM, DFT-s-OFDM, and OTFS signals.
  • ISAC signal processing and optimization: The review covers radar signal processing methods, including channel information matrix, spectrum lines estimator, and super resolution methods, followed by signal optimization methods.

A. Performance Metrics of ISAC Signals

ISAC signal evaluation uses sensing, communication, and implementation-oriented metrics to characterize performance across application scenarios. OFDM-based signals provide flexible orthogonal subcarriers and efficient communication, but introduce high-PAPR and interference challenges.

  • Resolution measures an ISAC signal’s ability to distinguish multiple targets.Smaller resolution values indicate stronger target-separation capability.
  • Ambiguity function captures time-frequency composite autocorrelation and supports deriving sensing metrics such as resolution, accuracy, and clutter suppression.
  • Doppler sensitivity measures sensing and communication output errors caused by Doppler frequency shifts.Doppler shifts can affect sensing outputs and received-signal BER.
  • PAPR is the ratio of peak instantaneous power to average power, and high PAPR requires higher-performance power amplifiers.
  • OFDM-based signal characteristics: OFDM’s orthogonal, flexibly spaced subcarriers reduce inter-carrier interference, improve spectrum efficiency, and support synchronization and equalization.The subcarrier number and spacing can be adjusted for different scenarios.
  • OFDM-based signal combinations: Combining OFDM with LFM, phase coding, or spread spectrum can improve sensing resolution, ambiguity properties, anti-interference capability, or PAPR.OFDM-LFM increases time-bandwidth product for long-distance target resolution, while phase coding reduces PAPR.

TLF M

The survey describes carrier- and non-carrier-modulated OFDM-LFM designs alongside phase-coded OFDM approaches. These combinations target high-resolution sensing, improved Doppler behavior, communication-rate enhancement, and sidelobe control.

  • TLF M: LFM’s increasing instantaneous frequency raises time-bandwidth product for high-resolution and long-distance sensing.OFDM-LFM also improves Doppler sensitivity and reduces velocity-estimation error.
  • TLF M: OFDM-LFM signals use carrier or non-carrier modulation, respectively mapping symbols onto LFM signals or altering sideband and subcarrier arrangements.
  • TLF M: Higher-order FSK, PSK, and QAM schemes improve the low data rate of earlier LFM modulation methods.
  • TLF M: Sideband multiplexing can avoid communication-radar interference, with DIR proportional to the number of communication subcarriers.
  • TLF M: Generalized OFDM-LFM designs use shifted, orthogonal sequences to obtain large time-bandwidth products and high-resolution distance and velocity estimation.
  • Phase coding: Phase-coded OFDM optimizes ambiguity-function PSLR, while high-autocorrelation sequences reduce sidelobe levels.Phase coding may be direct sequence modulation or PMCW, and sequence choice affects radar-sensing performance.
  • Phase coding: A phase-coded OFDM design supports high-resolution distance and velocity estimation with high DIR.

2) Phase modulated continuous waveform (PMCW):

PMCW and spread-spectrum-based ISAC signals are reviewed as approaches for sharper sensing characteristics, lower ambiguity, reduced PAPR, and improved interference behavior. Their sequence and multiplexing choices shape sensing and communication performance.

  • 2) Phase modulated continuous waveform (PMCW):: PMCW provides high-resolution sensing through a sharp pushpin ambiguity function that reduces distance-Doppler coupling.PMCW is also described as easily implemented.
  • 2) Phase modulated continuous waveform (PMCW):: MC-PMCW reduces PMCW modulation complexity and improves OFDM’s anti-noise and anti-interference performance.
  • 2) Phase modulated continuous waveform (PMCW):: Embedding communication symbols into PMCW reduces distance-Doppler ambiguity in MTC scenarios.
  • Spread spectrum: Combining OFDM with spread spectrum reduces OFDM PAPR and provides confidentiality, low power spectral density, and anti-interference capability.
  • Spread spectrum: SS-OFDM spreads transmitted communication data before IFFT and recovers it through FFT, demodulation, and despreading.A despreading module is added at the communication receiver.
  • Spread spectrum: Spread factor affects ranging-ambiguity sidelobes, while Gold sequences reduce the peak of the cross-ambiguity function.
  • Spread spectrum: Multi-carrier direct-sequence designs use scrambling and orthogonal variable-spread-factor codes to reduce subcarrier correlation, interference, and PAPR.

F. ISAC Signals Using the Candidate Signals of 5G-A/6G

The survey reviews FBMC, GFDM, DFT-s-OFDM, and OTFS as candidate ISAC signals for 5G-A/6G, highlighting trade-offs among spectral efficiency, sensing performance, implementation, and mobility robustness.

  • 1) FBMC:: FBMC avoids cyclic-prefix overhead and improves spectral efficiency, distance estimation, and Doppler performance, but filtering complicates demodulation and creates long time-domain tails.Its distance resolution is equivalent to OFDM, while its larger Doppler bandwidth improves distance and velocity estimation.
  • 2) GFDM:: GFDM uses grouped-subcarrier filters with shorter time-domain tails and flexible frames, while reducing out-of-band emissions and mutual interference compared with OFDM.GFDM requires complex reception to mitigate inter-symbol and inter-carrier interference, but achieves higher distance resolution than OFDM.
  • 3) DFT-s-OFDM:: DFT-s-OFDM provides lower PAPR and simple implementation while retaining OFDM-based multi-user processing, but high-order QAM can still produce high PAPR.Maintaining low PAPR alongside higher spectral efficiency remains a design challenge for 5G-A and 6G ISAC.
  • 4) OTFS:: OTFS maps symbols into the delay-Doppler domain, where it supports reliable high-mobility communication through nearly constant fading and sparse channel representation.This sparsity reduces pilot requirements and improves channel-estimation efficiency under large Doppler shifts.
  • 4) OTFS:: OTFS and OFDM match FMCW sensing performance while transmitting at high DIR; OTFS additionally uses shorter CP, longer sensing distance, faster tracking, and ICI-free high-velocity estimation.An OTFS-based ISAC design can use the entire frame for communication without channel-estimation pilots.

III. ISAC SIGNAL PROCESSING

ISAC signal processing methods are organized into channel information matrix, spectrum lines estimator, and super-resolution approaches, with computational complexity and sensing quality as key evaluation concerns.

  • Method categories: The survey groups ISAC processing into channel information matrix, spectrum lines estimator, and super-resolution methods.The channel information matrix category includes 2D FFT, cyclic correlation, and Bartlett methods, while Prony’s method represents spectrum-line estimation.
  • Performance metrics: Processing performance is evaluated using computational complexity, accuracy, RMSE, and CRLB, alongside sensing measures such as resolution and Doppler sensitivity.RMSE reflects estimation error and, because SNR affects sensing results, also reflects anti-noise performance.
  • Channel information matrix method: 2D FFT directly processes transmitted and received modulation symbols, exploiting OFDM’s time-frequency structure for low-complexity sensing.Distance and Doppler are extracted by IDFT across channel-matrix columns and DFT across rows, respectively.
  • Channel information matrix method: The channel information matrix is formed after eliminating known subcarrier phase shifts, with the channel matrix linking transmitted and received communication symbols.The Kronecker product represents the structured distance- and Doppler-dependent components.

2) Cyclic correlation method:

The section describes cyclic correlation and Bartlett processing alongside Prony’s spectrum-line method, emphasizing practical estimation procedures and noise-related limitations.

  • 2) Cyclic correlation method:: Cyclic correlation uses a virtual cyclic prefix whose length determines maximum detection distance, removing the direct limitation imposed by the communication CP.It forms a channel information matrix by correlating continuous time-domain echo and transmitted-signal sub-blocks.
  • 2) Cyclic correlation method:: Cyclic correlation is better suited than 2D FFT to OTFS signals when communication symbols may be zero, because direct division can amplify received noise.Distance and velocity are estimated from the resulting channel information matrix using ML and DFT methods.
  • 3) Bartlett’s method:: Bartlett processing groups channel-matrix vectors, averages their power spectra, and searches the resulting periodogram peak for distance and velocity.The method provides stable, accurate sensing and mitigates noise effects, whereas 2D FFT accuracy varies with FFT-point count.
  • C. Spectrum Lines Estimator Method: Prony’s method estimates sensing information from sampled values by solving difference equations and deriving Doppler shift and delay from characteristic roots.Its performance is affected by sampling interval and noise; TLS-SVD reduces computation and mitigates noise impact.

D. Super Resolution Method

Super-resolution methods overcome sampling-rate limits and improve sensing accuracy and resolution, with MUSIC offering the strongest accuracy at high computational complexity. MUSIC estimates DoA, distance, and velocity through subspace decomposition and spatial-spectrum peak searches.

  • Super-resolution methods overcome the sampling-rate limitations of 2D FFT and CC methods, improving estimation accuracy and resolution.
  • MUSIC provides the best super-resolution accuracy but high computational complexity; ESPRIT has medium accuracy and complexity, while MVDR has low resolution and accuracy.
  • MUSIC decomposes signal and noise subspaces to estimate the number of sources, DoA, distance, velocity, signal strength, and interference statistics.
  • MUSIC processing forms the received antenna-array vector, computes its autocorrelation matrix, constructs a spatial spectrum, and searches its peaks for sensing parameters.
  • For distance and velocity, MUSIC applies subcarrier and OFDM-symbol phase differences through channel-information-matrix row and column vectors.
  • MUSIC estimates multiple targets’ DoAs more accurately than ML and maximum entropy methods, with parameter estimates close to the CRLB.

3) MVDR method:

MVDR processes ISAC channel information to estimate target delay and Doppler while reducing noise and interference effects. In OFDM-based processing, 2D FFT offers low-complexity distance and velocity estimation.

  • MVDR constructs a channel information matrix, forms delay- and Doppler-domain frequency-response vectors, and uses autocorrelation matrices to build spatial spectra.
  • 2D FFT estimates target distance and velocity with low complexity and is widely used for OFDM-based ISAC signals.
  • OFDM reception removes the cyclic prefix and applies FFT to recover modulation symbols and process the received echo signal.
  • The channel response, Doppler frequency shift, and propagation delay characterize the received OFDM echo signal.

3) ISAC signal based on spread spectrum:

This section reviews spread-spectrum and related ISAC signal processing and optimization approaches, including OTFS conversion, PAPR reduction, and interference-aware waveform designs. The methods trade implementation overhead, spectrum use, and signal performance in different ways.

  • For SS-OFDM, dividing the received matrix by the transmitted matrix produces a channel information matrix used by 2D FFT to estimate target distance and velocity.
  • OTFS reception transforms received time-domain signals into the delay-Doppler domain using a Wigner transform, receiving window, and SFFT.
  • ISAC signal optimization covers PAPR reduction, interference management, and adaptive signal optimization because high PAPR can distort signals through power-amplifier nonlinearity.
  • Coding method: Using the Golay sequence reduces PAPR to 3 dB while providing strong error-correction capability and high sensing performance.
  • Probability method: PTS and SLM select or phase-adjust candidate sequences to reduce PAPR; PTS improves with more sub-blocks, while SLM avoids transmitted-signal distortion.
  • Probability method: PTS and SLM require side information for demodulation, making communication and sensing performance depend on side-information accuracy.
  • Probability method: A hybrid SLM-PTS method conditionally applies PTS after SLM, reducing PAPR with low complexity without degrading communication or sensing performance.
  • Tone preserving method: Tone reservation reduces PAPR by reserving subcarriers for peak-canceling signals, but this wastes spectrum resources.

B. Interference Management

ISAC interference management addresses mutual interference among users and self-interference between transmission and reception. The survey covers cancellation and avoidance methods, each with different processing requirements and operating constraints.

  • Mutual interference arises in multipath and multi-user scenarios, while self-interference occurs when echoes return before transmission is complete.
  • Mutual-interference cancellation: Serial and selective interference cancellation reconstruct interference from received signals, with selective cancellation processing only the strongest identified interferer.
  • Mutual-interference cancellation: Cancellation performance depends on frequency-offset estimation; small errors can produce incorrect reconstruction, while residual offsets accumulate during serial subtraction.
  • Self-interference cancellation: Self-interference can greatly exceed the echo because transmit and receive antennas are close, while the random channel complicates cancellation despite known transmitted signals.
  • Self-interference cancellation: Adaptive cancellation generates an opposite-phase, equal-amplitude signal, whereas channel estimation can eliminate self-interference for full-duplex moving-target detection.
  • Interference avoidance uses duplex or multiple-access schemes to avoid interference without complex cancellation processing.
  • Mutual-interference avoidance: TDMA, CDMA, OFDMA, and CSMA separate users through time slots, orthogonal codes, subcarriers, or distributed access to avoid mutual interference.

C. Adaptive Signal Optimization

Adaptive ISAC signal optimization adjusts parameters across space, time, and frequency to meet scenario-specific sensing and communication requirements. Pilot structure and resource allocation are also optimized to improve sensing performance while preserving communication functions.

  • Adaptive signal optimization: Optimization variables span space, time, and frequency, with objectives and constraints selected for metrics such as sensing MI, DIR, PAPR, and transmit power.Spatial methods include beamforming and precoding, while time-frequency methods optimize signal resources and power.
  • Adaptive signal optimization: Sensing MI is maximized under DIR and subcarrier-power constraints, or jointly with communication MI under a total-power constraint.The joint mutual-information formulation yields a closed-form optimal power allocation in the cited studies.
  • Adaptive signal optimization: Other objectives include radar detection probability and power allocation constrained by pulse-compression sidelobes and communication data rate.One cited method outperforms traditional windowing and waterfilling techniques under its stated conditions.
  • Pilot-based signal design: Pilot signals are known modulation-domain sequences used for synchronization and channel estimation, with high power, sensing performance, and anti-interference capability.Common OFDM pilot structures include block pilots, which suit frequency-selective fading, and comb pilots, which suit time-selective fading.
  • Pilot-based signal design: A comb-pilot system achieved 0.1 m distance resolution and 0.58 m/s velocity resolution using Barker coding, matched filtering, and a 2D Fourier transform.Superimposed pilots combine advantages of comb pilots for velocity estimation and block pilots for distance estimation.
  • Future trends: Future ISAC signals should be flexible and reconfigurable across scenarios, with sensing enhanced through pilot and data components and multidomain optimization.Resource allocation can optimize sensing accuracy and channel capacity, while dynamic pilot-subcarrier counts support short-, medium-, and long-range radar detection.

A. ISAC Signal Design

ISAC signal design for 5G-A and 6G spans waveform, frame-structure, and transmission-mode choices, while sensing algorithms must balance accuracy with computational complexity. The surveyed literature identifies flexible multidomain and cooperative designs as continuing research needs.

  • Signal design: ISAC signal design covers waveform, frame-structure, and transmission-mode design, with waveform selection requiring performance evaluation for sensing and communication.Duplex transmission remains an open design problem.
  • Waveform design: OFDM remains mainstream, while FBMC, GFDM, and OTFS are proposed for next-generation systems; THz-band signals are noted for high DIR and sensing accuracy.The paper also identifies low sidelobe, high reliability, multi-user interference, and communication-sensing balance as design requirements.
  • Frame-structure design: Fixed standardized frame structures cannot perfectly adapt to differentiated scenarios, motivating flexible and reconfigurable designs with pilots supporting sensing without affecting communication.The cited discussion emphasizes 3GPP 5G NR and IEEE 802.11 structures as current references.
  • Signal processing: ISAC signal processing pursues high sensing accuracy and low computational complexity, creating a trade-off between these performance indices.Multiple reference signals and accurate processing for future mmWave and THz systems remain open issues.
  • Cooperative sensing: Multi-BS cooperative sensing can improve sensing accuracy and range, but its algorithms remain challenging because existing methods mainly target pulse or coherent radar.The paper identifies cooperative signal processing as requiring further study for large-scale, high-accuracy sensing.
  • Future research: Comprehensive space-time-frequency optimization and flexible frame-structure integration remain limited, especially for multi-BS cooperative sensing.The surveyed conclusion presents multidomain optimization as a future trend for differentiated communication and sensing requirements.
Loading 2301.03857v3…