PaperlayerView exact revision on arXiv

Source-linked AI summary

Interference Management for Integrated Sensing and Communication Systems: A Survey

Yangyang Niu, Zhiqing Wei, Lin Wang, Huici Wu, Zhiyong Feng

arXiv:2403.16189v1cs.NI

TL;DR

ISAC systems face interference within a base station and across coordinated base stations, complicating communication and sensing. This article surveys interference-management approaches for these settings. It organizes methods across SI, MI, clutter, and CLI and covers suppression, avoidance, and exploitation.

Abstract

from arXiv · show

Emerging applications such as autonomous driving and Internet of things (IoT) services put forward the demand for simutaneous sensing and communication functions in the same system. Integrated sensing and communication (ISAC) has the potential to meet the demands of ubiquitous communication and high-precision sensing due to the advantages of spectrum and hardware resource sharing, as well as the mutual enhancement of sensing and communication. However, ISAC system faces severe interference requiring effective interference suppression, avoidance, and exploitation techniques. This article provides a comprehensive survey on the interference management techniques in ISAC systems, involving network architecture, system design, signal processing, and resource allocation. We first review the channel modeling and performance metrics of the ISAC system. Then, the methods for managing self-interference (SI), mutual interference (MI), and clutter in a single base station (BS) system are summarized, including interference suppression, interference avoidance and interference exploitation methods. Furthermore, cooperative interference management methods are studied to address the cross-link interference (CLI) in a coordinated multipoint ISAC (CoMP-ISAC) system. Finally, future trends are revealed. This article may provide a reference for the study of interference management in ISAC systems.

I. INTRODUCTION

ISAC combines sensing and communication through shared hardware, spectrum, and signal processing, but faces interference within and across base stations. This survey reviews interference-management methods and identifies gaps in prior surveys, including CoMP-ISAC.

A. Background and Motivation

  • ISAC integrates sensing and communication by sharing hardware, spectrum resources, and signal processing, while applications such as smart cities demand both functions.The paper also notes that ISAC systems face interference from sensing and communication and more complex inter-base-station crosstalk.Source: A. Background and Motivation · p. 1

1) Self-Interference (SI) :

  • Sensing echoes encounter self-interference because they arrive at the sensing receiver before ISAC signal transmission is complete.Source: 1) Self-Interference (SI) · p. 1

2) Mutual-Interference (MI) :

  • In downlink, stronger sensing signals affect communication reliability; in uplink, interference between communication signals and echoes challenges target recognition and localization.Source: 2) Mutual-Interference (MI) · p. 1

3) Clutter :

  • Industrial and urban environments expose ISAC systems to both stationary and dynamic clutter from the environment and target.Source: 3) Clutter · p. 1

4) Cross-Link Interference (CLI):

  • Coordinated multi-point ISAC can leverage diversity from distributed sensing nodes, but coordinating transmission and sensing processing across areas presents complex cross-link interference among base stations.Source: 4) Cross-Link Interference (CLI) · p. 1

B. Existing Surveys and Tutorials

  • Prior ISAC surveys emphasized system design and signal processing, while comprehensive coverage of interference management and CoMP-ISAC interference remained lacking.Source: B. Existing Surveys and Tutorials · p. 2

C. Contributions and Structure

1) SI Management :

The article reviews interference-management methods across several interference types and also surveys ISAC channel modeling and performance metrics. It organizes management reviews by interference type and, for CLI, by BS operational modes, uplink assistance, and network resource allocation.

1) SI Management :

  • SI cancellation and avoidance methods are reviewed in the digital, analog, and propagation domains.Source: 1) SI Management · p. 2

2) MI Management :

  • MI management methods are summarized separately for uplink and downlink transmission, covering suppression, avoidance, and exploitation.Source: 2) MI Management · p. 2

4) CLI Management :

  • CLI management methods are organized by BS operational modes, uplink assistance, and network resource allocation.Source: 4) CLI Management · p. 2

II. SYSTEM MODEL

  • The system-model section reviews ISAC channel modeling and performance metrics as foundations for understanding interference and evaluating system design.Source: §II · p. 2

1) Related Works:

  • Prior channel-modeling work includes a cluster-based model using shared scatterers and a model considering forward and backward scattering components.The cluster-based model is described as robust and efficient for scenarios with stacked or extended antenna configurations.Source: 1) Related Works · p. 3

2) Sensing Channel Modeling:

  • The sensing-channel model considers a broadband downlink with a BS, UE, target, and planar array, and specifies the phase difference relative to the reference antenna element.The UPA has size P × Q with uniform element intervals; the phase expression depends on azimuth and elevation angles, wavelength, and element indices.Source: 2) Sensing Channel Modeling · p. 4
  • Sensing-channel modeling expresses the channel impulse response and identifies target Doppler shift, delay, RCS, distance, radial velocity, and array steering parameters.Source: 2) Sensing Channel Modeling · p. 4

3) Communication Channel Modeling:

The communication-channel model represents one line-of-sight path and multiple non-line-of-sight paths, with path and array parameters described in the text and Table IV.

3) Communication Channel Modeling:

  • The model assumes one LoS path and L −1 NLoS paths arising from refraction, scattering, and other propagation effects.The paths are indexed from 0 through L.Source: 3) Communication Channel Modeling · pp. 4–5
  • Table IV is identified as the table of communication-channel modeling parameters, while Table III covers sensing-channel modeling parameters.Source: Table IV · p. 5, Table III · p. 5
  • Its parameters include large-scale fading, the Rician factor, path Doppler shifts and delays, distances, radial velocities, and a transmit/receive steering-vector product.Table IV gives the key communication-channel parameter definitions.Source: 3) Communication Channel Modeling · p. 5, Table IV · p. 5

B. Performance Metric

The section distinguishes communication and sensing performance measures, then frames self-interference as a threat to receiver operation and reviews analog suppression approaches.

B. Performance Metric

  • Communication is assessed by rate, outage probability, and mutual information, while sensing uses detection probability, resolution, accuracy, ambiguity function, and sensing mutual information.Table V summarizes these metrics and their definitions.Source: B. Performance Metric · p. 5, Table V · p. 6

III. SINGLE BS SI MANAGEMENT

  • Large SI can saturate the receiver ADC, distort or lose communication data, and constrain sensing of distant targets; suppression aims to reduce SI to the local-channel noise floor.The survey distinguishes analog, digital, and spatial-domain suppression.Source: §III · p. 5

1) Analog SI Mitigation:

  • Hardware optimization uses LNA output as a reference for tap weights, achieving over 30 dB suppression at lower bandwidths; hardware-aware tap-delay resolution can reach up to 45 dB.The first result is limited to lower bandwidths; the second characterizes channel bandwidth, carrier frequency, and bit precision.Source: 1) Analog SI Mitigation · p. 6
  • Analog cancellation reconstructs and subtracts SI in the RF domain to avoid ADC saturation, but nonlinear circuit effects and complex hardware tuning remain challenges.A two-stage design uses dual taps for direct and reflected leakage, then a nonlinear model for residual SI.Source: 1) Analog SI Mitigation · pp. 6–7

2) Digital Interference Mitigation:

Digital SI mitigation acts after ADC conversion through signal, antenna, algorithm, and receiver-filter designs; SI avoidance instead limits coupling through hardware, antennas, or beamforming.

2) Digital Interference Mitigation:

  • An auxiliary receiver chain supplies a digital copy of transmitted SI and can share an oscillator to reduce phase-noise effects; ICA-based algorithms exploit independence between echo and SI signals.The ICA algorithms are reported to reduce computational complexity compared with least squares.Source: 2) Digital Interference Mitigation · p. 7
  • Antenna selection and transmit-receive beamforming can suppress SI using channel characteristics, while digital beamforming can focus interference toward its arrival direction.These approaches can require frequent calibration as antenna arrays and environments change.Source: 2) Digital Interference Mitigation · p. 7
  • Pulse-based transmission at different times can completely cancel SI but underuses spectrum; a full-duplex design instead sends communication signals during radar waiting time to support sensing.Golay-sequence calibration and inserted radar or reference subcarriers are also described for short-range and distant-target sensing.Source: 2) Digital Interference Mitigation · pp. 7–8
  • A neural network followed by a linear filter achieves over 20 dB of interference cancellation, while linear prediction improves SIR by 12 −16 dB.Sparse recovery methods are also used to reduce sensing-SNR loss during interference separation.Source: 2) Digital Interference Mitigation · p. 8
  • Matched filters reduce the ALMS loop's peak response and residual-SI frequency component; sampling echoes at zero SI amplitude preserves desired signals but requires SI knowledge.The ALMS cancellation level depends on loop gain, tap delay, number of taps, and roll-off factor.Source: 2) Digital Interference Mitigation · p. 8

B. SI Avoidance

  • Inductors, shields, circulators, and couplers can reduce SI generation, but improving linearity may increase power consumption.A circulator-and-coupler design is reported to improve communication rates without increasing power consumption.Source: B. SI Avoidance · p. 8
  • A symmetric antenna structure with a π phase shifter uses destructive interference to improve transmitter-receiver isolation.Source: B. SI Avoidance · p. 8
  • Transmit-beamforming SIC places signals in the SI-channel null space, but noise, nonlinearity, estimation defects, and hardware limits can impair suppression as complexity raises computational cost.The survey notes simulation work examining channel-estimation errors.Source: B. SI Avoidance · p. 8

IV. SINGLE BS MI MANAGEMENT

Overlapping sensing and communication signals create mutual interference (MI), reducing communication throughput and sensing dynamic range. The section reviews downlink suppression and multiple-access approaches for managing MI.

IV. SINGLE BS MI MANAGEMENT

  • MI arises when sensing and communication signals overlap in time or frequency, diminishing communication throughput and sensing dynamic range.The survey treats suppression, avoidance, and exploitation separately for uplink and downlink transmission.Source: §IV · p. 8

A. Downlink MI Suppression

1) Direct Interference Suppression:

  • Direct suppression reconstructs MI from known interference structure and subtracts it, including through threshold filtering and pilot-based methods.Pilot symbols can support sensing-echo detection and communication-signal extraction for interference reconstruction.Source: 1) Direct Interference Suppression · pp. 8–9

2) Indirect Interference Suppression:

  • Indirect suppression anticipates or alters interference propagation, but its effectiveness depends on information such as channel models and pilot structures.A recursive neural network autoencoder supports blind detection when prior channel information is unavailable; space-time alignment can reduce antenna requirements for complete elimination.Source: 2) Indirect Interference Suppression · p. 9

1) Multiple Access (MA) Techniques:

  • TDMA, FDMA, NOMA, SDMA, and RSMA offer different ways to allocate shared resources or coordinate signals to avoid MI.The approaches include time-frequency allocation, user and sensing signal coordination, spatially distinct beams, and common/private communication streams.Source: 1) Multiple Access (MA) Techniques · pp. 9–11
  • In the mmWave frequency band, this approach reduced target detection error by 18.5%.Separately, under the specified NR frame structure, communication overhead can be reduced by 43.24%.Source: 1) Multiple Access (MA) Techniques · pp. 9–10

2) Beamforming Design:

The section reviews beamforming, transmitted-signal design, and RIS assistance as approaches to managing interference in downlink ISAC. These designs coordinate sensing and communication through spatial beams, waveform choices, and controllable reflections.

2) Beamforming Design:

  • Separate sensing and communication beams, or differentiated main- and side-lobe functions, are used to manage interference while balancing their effects on system performance.The reviewed designs include spatially directed beams and beamforming optimization to limit multi-user interference or mismatch with desired beams.Source: 2) Beamforming Design · p. 11

3) Transmited Signal Design:

  • Sensing-centric designs embed communication in radar signals, while communication-centric designs reuse communication signals for sensing and balance preamble length against sensing accuracy.Weighted designs impose constraints such as constant modulus or similarity, adding implementation complexity while accommodating sensing requirements.Source: 3) Transmited Signal Design · pp. 11–12

4) RIS Assisted:

  • Passive RIS controls reflected-signal phases to avoid MI and coherently combine desired signals, while active RIS amplifies reflections to compensate for path loss.Passive-RIS designs include joint waveform optimization and transmit-beamforming/reflection-coefficient optimization; an active-RIS example combines RIS with delay alignment modulation.Source: 4) RIS Assisted · p. 12

C. Downlink MI Exploitation

The section describes exploiting constructive downlink MI to improve signal quality or reduce transmit power, alongside uplink methods for suppressing or avoiding interference between communication signals and sensing echoes.

1) Power Exploitation:

  • Constructive downlink MI can enhance received SINR or reduce expected transmit power, including through precoding that retains beneficial multi-user interference.An IoT energy-harvesting approach uses direct and RIS-reflected links and is reported to increase network throughput while reducing interference to access points.Source: 1) Power Exploitation · p. 12

2) Reduce SER:

  • Downlink interference can be used to move received signals into favorable decision regions, reducing symbol error rate or improving target detection probability and energy efficiency.The reviewed methods include waveform optimization that pushes noiseless received signals away from the decision threshold.Source: 2) Reduce SER · p. 12

1) Interference Suppression:

  • Uplink suppression includes detecting and subtracting communication signals to recover sensing echoes, while NOMA-assisted SIC processes communication and radar signals in sequence.The reviewed NOMA-SIC approaches report higher degrees of freedom than frequency-division ISAC at the same communication rate.Source: 1) Interference Suppression · pp. 12–13

2) Interference Avoidance:

  • Uplink interference avoidance uses sensing-aided channel estimation, concurrent downlink-uplink processing, and flexible beam management to limit interference effects.An extended-KF beam-alignment method is described as reducing beam-tracking overhead in high-speed mobility without additional sensing devices.Source: 2) Interference Avoidance · p. 13

3) Interference Exploitation:

The section reviews uplink information as an aid to downlink channel estimation, then surveys clutter management through physical methods, filtering, and waveform design. It notes that clutter suppression can depend on prior target and clutter information.

3) Interference Exploitation:

  • Uplink observations can support downlink CSI acquisition and reduce feedback or training overhead, including recovery of channel parameters in high-mobility scenarios.The survey also describes RIS-assisted statistical channel estimation with over 100% gain in fair SNR allocation among users using 40 RIS elements.Source: 3) Interference Exploitation · p. 13

1) Physical Methods:

  • Physical clutter removal can be limited by unpredictable clutter statistics and coupling between clutter and useful signals, making physical methods alone challenging for high sensing demands.Raising base-station height or adding clutter-rejecting structures also entails cost or can weaken reception of useful signals.Source: 1) Physical Methods · pp. 13–14

2) Filter Design:

  • Clutter filters operate in time, frequency, or space-time domains, with time-domain removal risking loss of valuable echoes arriving within the filter duration.Frequency-domain methods distinguish moving targets from clutter by spectral differences, while STAP relies on accurate clutter covariance matrix estimation.Source: 2) Filter Design · p. 14

1) Waveform Design:

  • Waveform designs use joint optimization, symbol-level precoding, or neighboring-range-unit clutter mitigation to improve radar SINR and communication performance.The surveyed clutter suppression methods can rely on prior target and clutter information; inaccurate environmental parameters or frequency offsets can degrade detection and estimation.Source: 1) Waveform Design · pp. 14–15

2) Flexible Antenna Structures:

This section surveys flexible RIS and RHS antenna designs for clutter avoidance, then describes clutter’s use in environmental applications and cooperative strategies for managing CLI across ISAC base stations.

2) Flexible Antenna Structures:

  • Passive RIS redirects clutter and supports multi-target detection, active RIS adds amplification or hybrid control, and RHS uses amplitude-controlled beamforming with lower radar power but added processing complexity.RHS can be combined with RIS to reduce environmental clutter in broadband DFRC designs.Source: 2) Flexible Antenna Structures · p. 15

C. Clutter Exploitation

  • Clutter supports weather monitoring, 3D environmental reconstruction and obstacle boundaries, and search-and-rescue detection using mobile robots or harmonic radar.The examples include NEXRAD precipitation tracking, UAV mapping, and Baltic Sea tests tagging life jackets and rescue boats.Source: C. Clutter Exploitation · pp. 15–16

VI. CLI MANAGEMENT IN COMP-ISAC SYSTEM

  • CoMP-ISAC CLI arises among communication links, sensing links, and sensing-communication links across multiple base stations.The survey organizes CLI management around cooperative sensing, uplink/downlink cooperation, and network resource allocation.Source: §VI · p. 16

A. Collaborative Sensing Approach

  • Coordinating base-station transmit and receive modes can use distributed macro-diversity and suppress interference, with self-transmit/self-receive, multi-transmit/self-receive, and multi-transmit/multi-receive modes.Source: A. Collaborative Sensing Approach · p. 16

1) Self-Transmit Self-Receive:

  • Clustering base stations by mutual coupling loss and aligning their link directions can prevent strong CLI between neighboring stations.The reviewed strategy groups pairs whose mutual coupling loss falls below a threshold.Source: 1) Self-Transmit Self-Receive · p. 16

2) Multi-Transmit Self-Receive:

The section reviews cooperative sensing configurations, uplink/downlink cooperation, and resource-allocation approaches for managing interference across base stations.

2) Multi-Transmit Self-Receive:

  • A selected sensing receiver can coordinate with transmitting base stations through joint beamforming optimization; dynamic receiver selection can reduce base-station power and improve detection probability.Communication-user selection can also minimize transmission power to suppress inter-user interference.Source: 2) Multi-Transmit Self-Receive · p. 16

3) Multi-Transmit Multi-Receive:

  • Multi-transmit/multi-receive designs coordinate base-station power or fuse multi-view sensing, while cell-free networks add degrees of freedom but increase scheduling and processing demands.One design’s sensing performance declines when targets lie between two base stations; cell-free schemes also motivate low-complexity sensing and receiver-cancellation algorithms.Source: 3) Multi-Transmit Multi-Receive · pp. 16–17

B. Cooperation of Uplink and Downlink

  • Uplink-aided sensing uses cross-antenna cross-correlation to mitigate timing and carrier-frequency offsets, while joint time-space-frequency processing improves angle-of-arrival estimation over spatial samples alone.The section also reports that uplink-downlink collaboration can raise sensing rates while maintaining the same communication rate.Source: B. Cooperation of Uplink and Downlink · p. 17

C. Network Resource Allocation

  • CoMP-ISAC power control must balance interference reduction against desired-signal weakening and energy efficiency, motivating allocation that strengthens desired signals with minimal power.Source: C. Network Resource Allocation · p. 17

1) Network Architecture Design :

  • Network architecture options include dedicated sensing receivers, passive monitoring terminals, sensing-assisted neighbor discovery, and congestion-aware multipath routing.Passive target-monitoring terminals can receive BS transmissions and avoid self-interference; sensing-based neighbor discovery can also avoid CLI.Source: 1) Network Architecture Design · p. 17

2) Power Allocation:

  • Power-allocation methods combine water-filling with precoding, jointly optimize beamforming and receiver filters, or coordinate uplink and downlink power to suppress interference.The reviewed approaches also include CoMP with successive interference cancellation for higher throughput in multi-user NOMA clusters.Source: 2) Power Allocation · pp. 17–18

VII. FUTURE TRENDS

Future directions emphasize signal transmission, sensing algorithms, and data fusion to address accuracy, robustness, and complexity requirements.

1) Signal Transmission:

  • Future signal-transmission research should improve transceiver matching and explore waveform, channel-modeling, and uplink-downlink designs suited to changing scenarios.The review highlights near-field modeling, newer waveforms, and terminal-assisted sensing under high mobility where channel orthogonality is challenging.Source: 1) Signal Transmission · p. 18

2) Sensing Algorithm:

  • Sensing algorithms need greater robustness to estimation errors, timing and frequency offsets, mobility, and imperfect self-interference cancellation, while reducing computation for constrained devices.Imperfect cancellation can raise radar-receiver noise floors and make distant targets harder to locate accurately.Source: 2) Sensing Algorithm · pp. 18–19

3) Data Fusion:

  • Future distributed sensing fusion must balance accuracy and computational burden across soft, neutral, hard, and symbol-level methods while accounting for node synchronization.Signal-level fusion requires high synchronization accuracy, whereas data-level fusion leaves room to improve sensing accuracy.Source: 3) Data Fusion · p. 19

1) Multi-Band Cooperation:

The section outlines future directions spanning multiband and frame design, network control and architecture, new deployment degrees of freedom, and security.

1) Multi-Band Cooperation:

  • Combining frequency bands could address single-band interference limitations by pairing low-frequency coverage and penetration with higher-frequency sensing precision and bandwidth.Source: 1) Multi-Band Cooperation · p. 19

2) Flexible Frame Structure :

  • Flexible frame structures should adapt sensing resources and signal correlation to service and network conditions, with dynamic power control to reduce near-far effects and neighboring-cell interference.Source: 2) Flexible Frame Structure · p. 19

3) Dynamic Power Control :

  • Non-cellular architectures and flexible base-station operating modes are proposed to address uneven resource use and reduce interference from shared resources and S&C operation.Source: 3) Dynamic Power Control · p. 19

4) New DoF :

  • RIS offers signal-path control, while UAVs can provide agile line-of-sight paths for sensing and tracking, though UAV processing, energy, and interaction remain constrained.The section identifies UAV-assisted target tracking and SLAM with lower power consumption as a prospective direction.Source: 4) New DoF · pp. 19–20

1) Service-Driven Network Architecture:

  • Service-driven architectures could adjust power, frequency, and signal paths to real-time demands, while SDN and AI may improve prediction and adaptation despite computational and security challenges.Source: 1) Service-Driven Network Architecture · p. 20

2) Enhanced Security and Privacy :

  • Wireless sensing’s capacity to monitor people and sensitive sites creates privacy and security risks, motivating stronger encryption, access control, and transmission monitoring.Source: 2) Enhanced Security and Privacy · p. 20

VIII. CONCLUSION

  • The survey synthesizes ISAC interference management for self-interference, mutual interference, clutter, and cross-link interference, and outlines challenges and future directions.Source: §VIII · p. 20
Loading 2403.16189v1…