PaperlayerView exact revision on arXiv

Source-linked AI summary

A Tutorial on Movable Antenna-Enabled ISAC Systems: Fundamentals, Parameter Estimation, and Security Issues

Zhendong Li, Zhou Su, Jianle Ba, Linchu Chen, Yan Yang, Weichun Zhao, Jinyuan Huang, Tom H. Luan, Wen Chen, Qingqing Wu, Lipeng Zhu, Zhenyu Xiao, Weidong Mei, Nan Cheng, Ruijin Sun, Lin Chen, Ying Wang

arXiv:2610.00218v1eess.SP

TL;DR

MA-enabled ISAC brings antenna-position flexibility to shared communication and sensing systems, but its integration raises estimation and security challenges. This tutorial reviews system fundamentals, unified estimation models and methods, and physical-layer security approaches. It concludes that MA adds spatial freedom for balancing communication, sensing, and security, while identifying practical robustness and mobility concerns.

Abstract

from arXiv · show

Movable antenna (MA) technology has recently emerged as a promising paradigm for enhancing the flexibility and performance of wireless systems by allowing antennas to dynamically adjust their spatial positions within a confined region. Meanwhile, integrated sensing and communication (ISAC) has attracted significant attention as a key technology for next-generation wireless networks, aiming to integrate communication and sensing within a shared hardware and spectral framework. By leveraging the additional spatial degrees of freedom offered by MA, MA-enabled ISAC systems provide new opportunities to enhance both communication performance and sensing accuracy. However, the integration of MA into ISAC also poses new challenges in parameter estimation and security. In this tutorial, we provide a comprehensive overview of MA-enabled ISAC systems, with a particular focus on their fundamentals, parameter estimation, and security issues. First, we introduce the basic principles of MA and ISAC, and review representative MA-enabled ISAC systems and their potential applications. Then, we establish unified mathematical models for communication and sensing, and systematically review representative channel and sensing parameter estimation methods tailored to MA-enabled ISAC systems. Numerical comparisons are provided to illustrate the characteristics and performance of different estimation methods. Furthermore, we investigate the security issues of MA-enabled ISAC systems, introduce the fundamentals of physical-layer security (PLS), and summarize how the spatial reconfigurability of MAs can be exploited to enhance the security of ISAC systems. Representative case studies are presented to demonstrate the performance gains of MA-enabled secure ISAC systems. Finally, we identify several open issues and future research directions toward practical MA-enabled ISAC systems.

I. INTRODUCTION

This introduction situates MA-enabled ISAC within 6G’s integrated service vision and motivates antenna-position flexibility as a way to jointly adapt communication and sensing. It presents the tutorial as a systematic review of fundamentals, estimation methods, security, and open directions.

A. Background

  • 6G scenarios include immersive and massive communication, HRLLC, ubiquitous connectivity, integrated AI and communication, and ISAC, extending 5G service paradigms.ISAC shares hardware and radio resources for data transmission and environmental sensing, with communication and sensing also supporting one another.Source: A. Background · pp. 1–2

B. Motivations: From ISAC to MA-Enabled ISAC

  • MA adds antenna position as a design degree of freedom, allowing spatial responses to adapt to propagation conditions and offering opportunities to jointly improve communication and sensing.Reported benefits include mitigating deep fading, improving sensing accuracy, and reducing parameter-estimation CRBs; the added position freedom also couples deployment with estimation, beamforming, and security challenges.Source: B. Motivations: From ISAC to… · pp. 2–3

C. Related Works and Comparisons

  • Prior work covers ISAC, MA, fluid antennas, and related topics, but a comprehensive tutorial systematically treating MA-enabled ISAC architectures, methods, estimation, applications, and security is lacking.The article addresses this gap by reviewing MA and ISAC techniques, examining physical-layer security, and identifying open problems.Source: C. Related Works and Comparisons · p. 3

D. Contributions

  • The tutorial reviews MA architectures, spatial degrees of freedom, modeling, beamforming, positioning, trajectory optimization, and representative application scenarios.Applications include agriculture, healthcare, homes, factories, vehicles, oceans, low-altitude networks, and integrated air-space-ground networks.Source: D. Contributions · p. 3
  • A unified estimation framework treats communication and sensing channels together, reviews representative methods, and compares their accuracy and computational complexity.Tensor decomposition exploits high-dimensional signal structure to decouple parameter estimation, with stated benefits in accuracy and computational complexity.Source: D. Contributions · pp. 3–4
  • The security review covers physical-layer-security metrics and techniques, and examines joint antenna-position and beamforming optimization for strengthening legitimate links and suppressing eavesdroppers.Case studies consider mobility scenarios, channel uncertainty, and dynamic adversarial environments.Source: D. Contributions · p. 4

E. Organization

The tutorial proceeds from MA and ISAC fundamentals to modeling and later topics, while classifying movable antennas by spatial degrees of freedom. The types range from elements moving along one axis to panels with three-dimensional movement and rotation.

E. Organization

  • The tutorial organization places MA-enabled ISAC fundamentals before parameter estimation, security, and open issues.The listed topics include unified modeling, estimation algorithms and comparisons, physical-layer security, and future directions.Source: E. Organization · pp. 4–5

II. OVERVIEW OF MA-ENABLED ISAC SYSTEMS

  • Discrete models select antenna positions from a finite candidate set, whereas continuous models allow movement within a prescribed region and offer higher spatial DoF.The tutorial uses a 2D MA-enabled ISAC system as an illustrative example.Source: §II · p. 5

A. Movable Antenna

  • The tutorial identifies hardware architectures and classifies MAs according to their spatial degrees of freedom.Fig. 3 is captioned as showing MA hardware architectures.Source: 2) Types of MA · p. 5, Figure 3 · p. 6

2) Types of MA:

  • 1DMA elements move along a line, 2DMA elements within a rectangular region, and 3DMA elements within a rectangular parallelepiped.The described steering-vector models use the antenna positions and the target or receiver’s line-of-sight angles.Source: 2) Types of MA · p. 6
  • 6DMA panels can move in three dimensions and rotate in three dimensions, providing panel-level spatial adjustment.The tutorial also notes that combining movement and rotation dimensions permits customized MA dimensionalities from 1D to 6D.Source: 2) Types of MA · pp. 6–7

3) Comparison Between Conventional FPAs and MAs:

The section contrasts fixed array geometry with movable antennas’ position and orientation adjustments, which support adaptive channel exploitation and flexible coverage. It also outlines the added channel-acquisition and actuation burdens, alongside potential MA-enabled ISAC application settings.

3) Comparison Between Conventional FPAs and MAs:

  • MAs reposition elements or panels to adapt array geometry and channel use, while fixed-position arrays retain static spatial degrees of freedom after deployment.The comparison also attributes more effective beamforming and improved signal strength, SNR, and channel capacity to MA position or orientation adjustment.Source: 3) Comparison Between Conventional… · p. 7
  • As the movable range grows, the channel state space expands, increasing estimation, modeling, and feedback overhead.Systems need to estimate or predict channel information over the feasible movement region to exploit spatial adaptability.Source: 3) Comparison Between Conventional… · p. 7
  • MA actuation and control mechanisms increase hardware cost, power consumption, and maintenance complexity compared with fixed deployments.The passage cites motors, rails, rotation devices, and positioning controllers as examples.Source: 3) Comparison Between Conventional… · pp. 7–8

B. Application Scenarios

  • The tutorial identifies applications from digital agriculture and healthcare to smart homes, factories, vehicles, oceans, low-altitude networks, and space-air-ground networks.It presents these as settings where MA-enabled ISAC may combine communication with sensing.Source: B. Application Scenarios · p. 8

1) Digital Agriculture:

The application examples describe MA-enabled ISAC across agricultural, healthcare, domestic, industrial, transport, marine, aerial, and integrated space-air-ground settings. The system-model discussion distinguishes discrete position selection from continuous positioning and relates each choice to different spatial flexibility and optimization demands.

1) Digital Agriculture:

  • In digital agriculture, MA-integrated UAVs can collect crop, soil-moisture, and pest information while maintaining data links.The passage frames flexible antenna positions and beam directions as a way to address agricultural IoT coverage, reliability, and sensing-accuracy challenges.Source: 1) Digital Agriculture · p. 8

2) Smart Healthcare:

  • In smart healthcare, antenna-position adjustment is described as supporting vital-sign monitoring and data transmission in complex medical environments.The application discussion also identifies reliable communication, accurate sensing, and secure data collection as healthcare needs.Source: 2) Smart Healthcare · p. 9

3) Smart Home:

  • For smart homes, movable antennas can adjust to occupants to reduce multipath and occlusion while supporting communication stability and sensing accuracy.The passage gives adapting energy use in unoccupied rooms while maintaining comfort in active areas as an example.Source: 3) Smart Home · p. 9

4) Smart Factory:

  • In factories, antenna-position adjustment can sense robot trajectories and workpiece locations while maintaining reliable, low-latency communication.The passage identifies metal blockage and multipath as potential issues for fixed arrays in automated workshops.Source: 4) Smart Factory · p. 9

5) IoV:

  • For the IoV, MA-enabled ISAC can support perception of dynamic vehicle behavior while maintaining reliable, low-latency communication.The passage’s roadside-unit example detects sudden obstacles and sends warnings to nearby vehicles in real time.Source: 5) IoV · p. 9

6) Smart Ocean:

  • In smart oceans, flexible antenna adjustment is described as supporting ship-location and trajectory sensing while reducing sea-surface multipath interference.The example concerns illegal-fishing monitoring and real-time data transmission to command centers.Source: 6) Smart Ocean · p. 9

7) Low-Altitude Intelligent Network:

  • For low-altitude networks, MA spatial configuration is described as mitigating channel fluctuations and supporting UAV obstacle perception and collaborative path planning.The passage situates the network generally below 1000 meters, extendable to 3000 meters when necessary.Source: 7) Low-Altitude Intelligent Network · pp. 9–10

1) Signal and Channel Models:

  • Discrete modeling selects among finite candidate positions, while continuous modeling permits arbitrary positions in a feasible region and higher spatial DoF.Discrete models better align with practical hardware but generally yield combinatorial optimization; continuous models support fine-grained position or trajectory control.Source: 1) Signal and Channel Models · pp. 10–11

2) Antenna Positioning:

MA positioning methods seek antenna configurations that improve MA-enabled ISAC performance, using optimization, heuristic search, or learned policies. Their trade-offs include computational complexity, adaptability, and dependence on training data.

2) Antenna Positioning:

  • MA positioning methods fall into mathematical optimization, heuristic, and machine-learning-based approaches.The approaches differ in how they search antenna configurations and in their computational and data requirements.Source: 2) Antenna Positioning · p. 11
  • Optimization-based methods use structured formulations and tools such as convex optimization, fractional programming, gradient projection, and alternating optimization.They provide theoretical formulations and reliable performance gains, but complexity can rise in large-scale or non-convex scenarios.Source: 2) Antenna Positioning · p. 11
  • Heuristic methods use rule-based or intelligent searches for high-dimensional, discrete, or non-differentiable positioning problems.Examples include stochastic positioning with alternating optimization and greedy matching, grid refinement, and graph-based shortest-path search.Source: 2) Antenna Positioning · p. 11
  • Learning-based methods model the relationship between MA configurations and performance, with potential adaptability to dynamic or imperfect-CSI scenarios.Their performance depends on data quality, model design, and training complexity.Source: 2) Antenna Positioning · p. 11

3) Beamforming Design:

The section surveys beamforming and MA-enabled ISAC system models, then relates shared communication and sensing channel structure to unified parameter estimation. It also summarizes MA’s spatial flexibility and representative applications.

3) Beamforming Design:

  • Beamforming controls signal spatial distribution and affects communication quality and sensing performance; designs use mathematical optimization or machine learning.Optimization methods can offer reliable gains but face complexity in large, highly coupled systems, while learning methods can improve real-time adaptability but depend on training data, architecture, and convergence stability.Source: 3) Beamforming Design · p. 12

D. Chapter Summary

  • MA dynamically adjusts antenna positions to add spatial degrees of freedom, supporting flexibility, coverage, and adaptability in dynamic wireless environments.The chapter reviews MA hardware architectures and classifications, application scenarios, and studies of signal and channel modeling, positioning, and beamforming.Source: D. Chapter Summary · p. 12

A. Unified Mathematical Model Construction

  • MA-enabled ISAC models the received signal using communication and sensing channels, with shared training pilots and data symbols for communication and a sensing signal identical to the communication pilot.The channels share a similar mathematical structure, supporting unified estimation of channel and target parameters from the same pilots.Source: A. Unified Mathematical Model… · p. 13

B. Parameter Estimation Algorithms

The section compares least-squares and subspace-based approaches for estimating channel and target parameters. Least squares can recover a channel matrix but is costly for parameter estimation, while MUSIC and ESPRIT use signal-subspace structure in different ways.

1) LS-based Algorithm:

  • The LS approach estimates the channel matrix from orthogonal pilots and can reduce channel-estimation count by reconstructing the channel from estimates at several ports.Extracting delay and Doppler through Fourier-transform peaks, and other angles through correlation, adds substantial computational overhead; the passage concludes LS is impractical for parameter estimation in ISAC.Source: 1) LS-based Algorithm · pp. 13–14

2) Subspace-Based Algorithm:

  • MUSIC decomposes the received-signal covariance matrix into signal and noise subspaces and estimates parameters from peaks in a spatial spectrum.Joint steering vectors can resolve multiple parameters simultaneously, but expand the search codebook and increase computational and storage requirements.Source: 2) Subspace-Based Algorithm · p. 14
  • ESPRIT uses rotational invariance between two identical subarrays to extract parameters through a generalized eigenvalue problem, without spectral peak search or grid mismatch.MA positions can be adjusted to satisfy shift invariance, but antenna movement adds time overhead that may make ESPRIT unsuitable in time-varying channels.Source: 2) Subspace-Based Algorithm · p. 14
  • MUSIC permits arbitrary array geometries, whereas ESPRIT requires translational invariance; both use covariance-matrix eigenvalue decomposition and require the number of signal sources.Source: 2) Subspace-Based Algorithm · p. 15

3) CS-Based Algorithm:

The section presents compressed sensing and tensor decomposition as approaches to estimating MA-enabled ISAC parameters from multidimensional observations. Tensor methods preserve signal structure and decouple parameters, while their accuracy and computational costs depend on the chosen method and conditions.

3) CS-Based Algorithm:

  • OMP recovers sparse channel parameters by selecting dictionary atoms against the residual, updating coefficients by least squares, and stopping after P iterations or when residual energy is below a threshold.The selected support indices map to estimated parameter locations.Source: 3) CS-Based Algorithm · p. 15
  • OMP is sensitive to measurement-matrix correlation and requires the restricted isometry property; its complexity also grows rapidly with signal dimensionality and dictionary size.ROMP and CoSaMP are cited as approaches developed to improve robustness and efficiency.Source: 3) CS-Based Algorithm · p. 15

4) Tensor Decomposition-Based Algorithm:

  • Tensor decomposition processes multidimensional received data directly, extracting factor matrices whose factors carry parameter information and can decouple parameter estimation.The approach avoids the information loss and computational burden associated with vectorizing or matricizing multidimensional data.Source: 4) Tensor Decomposition-Based Algorithm · p. 15
  • ALS iteratively updates factor matrices to reduce fitting error, but can be computationally intensive, converge to local optima, and limit resolvable targets without structural constraints.It is general for CP decomposition and can accommodate constraints through constrained least-squares steps.Source: 4) Tensor Decomposition-Based Algorithm · p. 16
  • SCPD addresses rank deficiency with spatial smoothing, extracts Vandermonde factors using ESPRIT-inspired rotational invariance, and recovers remaining factors through Kronecker-product relationships.It avoids iterative optimization, while exploiting factor-matrix structure.Source: 4) Tensor Decomposition-Based Algorithm · p. 16
  • Table IV summarizes representative studies and their parameter-estimation algorithms, but the supplied passage does not provide comparative table entries.The listed work covers parameter-estimation approaches for MA-enabled ISAC.Source: 4) Tensor Decomposition-Based Algorithm · p. 16, Table IV · p. 17

C. Performance Analysis

The analysis compares estimation accuracy and computational complexity for representative algorithms, using reconstructed-channel NMSE as the accuracy metric. It reports SCPD's low-SNR accuracy advantage and complexity trade-offs, and introduces a CRB comparison framework.

1) Estimation Accuracy Analysis:

  • The analysis evaluates reconstructed-channel estimation using NMSE, with a setup of 12 transmit and 12 receive MAs and 10 RF chains at each end.The system uses a 28 GHz carrier and 100 MHz bandwidth.Source: 1) Estimation Accuracy Analysis · p. 17
  • SCPD achieves the best estimation accuracy at low SNR, and tensor decomposition methods show a significant NMSE advantage over baseline algorithms.MUSIC and OMP are noise-sensitive, whereas tensor decomposition is described as more noise-robust.Source: 1) Estimation Accuracy Analysis · p. 17, Figure 7 · p. 17
  • At five paths, the two tensor-based estimation algorithms do not differ significantly; too few or too many paths can impair accuracy through rank reduction or weakened factor-column orthogonality.With fixed RF-chain, subcarrier, and time-slot dimensions, tensor-decomposition uniqueness may become difficult to guarantee as path count rises.Source: 1) Estimation Accuracy Analysis · pp. 17–18, Figure 8 · p. 17

2) Computational Complexity Analysis:

  • SCPD has lower complexity than ALS for small factor dimensions, but its spatial-smoothing cost grows rapidly with dimensions; MUSIC is driven by search-grid and subcarrier sizes, while OMP scales with target and grid counts.The stated OMP complexity is O{RNgridQMSNsKM}.Source: 2) Computational Complexity Analysis · p. 18

3) CRB Analysis:

  • The CRB is formed as the inverse of the Fisher information matrix, providing a lower bound on the variance of unbiased estimators for comparison with tensor-based estimates.The log-likelihood and complex Fisher information matrix provide the calculation route described in the section.Source: 3) CRB Analysis · p. 18

D. Chapter Summary

The chapter summarizes estimation approaches and introduces physical-layer security for MA-enabled ISAC, where shared signals can expose both communication data and sensing information. It reviews wiretap models and security metrics, while noting that MA position flexibility creates a design dimension for balancing secure communication and sensing.

D. Chapter Summary

  • The chapter reviews LS, subspace, compressed-sensing, and tensor-decomposition estimation methods, which trade off accuracy, computational complexity, and prior-knowledge requirements.Tensor methods exploit multidimensional signal structure, while complexity, initialization sensitivity, and practical-channel robustness remain challenges.Source: D. Chapter Summary · p. 18

1) Introduction to PLS:

  • Because ISAC waveforms serve communication and sensing, unintended reception can expose confidential data as well as sensing intentions and parameters.PLS uses channel characteristics such as fading, noise, and diversity for information-theoretic secure transmission.Source: 1) Introduction to PLS · pp. 18–19

2) Wiretap Channel Models and Secrecy Capacity:

  • Wyner's wiretap model frames confidential transmission from Alice to Bob in the presence of Eve, with secrecy capacity defining the maximum reliable rate that remains confidential.For ISAC, a sensing target adds complexity because the waveform must also extract environmental parameters.Source: 1) Introduction to PLS · p. 19, 2) Wiretap Channel Models and Secrecy… · p. 19
  • In the Gaussian wiretap model, positive secrecy capacity requires Bob's channel quality to exceed Eve's, and larger channel disparity yields greater secrecy capacity.ISAC must engineer this channel disparity while maintaining sensing requirements.Source: 2) Wiretap Channel Models and Secrecy… · p. 19

3) Performance Metrics of PLS:

  • PLS metrics include ergodic secrecy capacity for long-term secure rate, secrecy outage probability for fixed-rate reliability, and secrecy energy efficiency for rate relative to power use.MA actuation energy adds a trade-off with secrecy and sensing gains.Source: 3) Performance Metrics of PLS · pp. 19–20

4) Techniques for Achieving PLS:

The section outlines conventional PLS techniques and describes how movable antennas can physically reconfigure channels for secure ISAC. Position optimization can strengthen Bob's channel while reducing Eve's gain, with sensing illumination retained as a design requirement.

4) Techniques for Achieving PLS:

  • Beamforming focuses energy toward the legitimate user and suppresses Eve, while ISAC beam design must also illuminate the sensing target sufficiently.The design balances communication SINR, sensing-beampattern matching, and Eve nulling.Source: 4) Techniques for Achieving PLS · p. 20
  • Artificial noise degrades Eve's reception while being placed in the legitimate channel's null space; in ISAC it must also avoid corrupting sensing echoes.Its transmit power is allocated between the information signal and artificial noise.Source: 4) Techniques for Achieving PLS · p. 20

B. Mechanisms of MA for Enhancing PLS in ISAC

  • MAs enable physical channel reconfiguration through transmitter position control, adding a design dimension for enhancing PLS in ISAC.The section organizes the mechanisms around channel customization, joint position and beamforming optimization, and collaborative design with artificial noise.Source: B. Mechanisms of MA for Enhancing PLS… · p. 20

1) Spatial Channel Customization:

  • By optimizing MA positions, different paths can combine coherently toward Bob while sensing-direction power is maintained; the reported channel-gain maps show increased gain for Bob and suppressed gain for Eve.The passage attributes this to selecting positions that bypass deep-fading regions.Source: 1) Spatial Channel Customization · p. 21, Figure 10 · p. 21
  • Jointly optimizing MA coordinates can make antenna signals destructively interfere at Eve, widening the legitimate-to-wiretap channel gap without compromising sensing performance.MAs can also control main-to-wiretap channel correlation, particularly when Eve is angularly close to the legitimate user.Source: 1) Spatial Channel Customization · p. 21

2) Joint Antenna Position and Beamforming Optimization:

The section describes joint antenna-position and beamforming design for secure ISAC, balancing secrecy-rate maximization with sensing constraints. It also explains how movable antennas can shape artificial noise toward an eavesdropper while limiting interference to the sensing target.

2) Joint Antenna Position and Beamforming Optimization:

  • The design maximizes secrecy rate subject to minimum sensing accuracy, transmit-power, and antenna-separation constraints, and may require artificial noise to lie in the legitimate-channel null space.Alternating optimization is used to decouple the coupled position and beamforming variables over a continuous non-convex space.Source: 2) Joint Antenna Position and… · p. 22
  • Jointly optimizing movable-antenna positions and artificial-noise projection can suppress multiple eavesdroppers while maintaining a desired sensing beampattern; near-field range focusing can separate same-angle users at different distances.The near-field case also achieves high-resolution target localization.Source: 2) Joint Antenna Position and… · p. 22

3) Collaborative Design of AN and Antenna Positions:

  • Jointly optimizing antenna positions and artificial-noise covariance can direct interference toward Eve while limiting noise leakage toward the sensing target.The target-direction leakage is bounded to preserve sensing accuracy; position changes can also manipulate the orientation of the legitimate-channel null space.Source: 3) Collaborative Design of AN and… · p. 22

C. Existing Works of MA-Enabled Secure ISAC

Existing work uses movable-antenna spatial flexibility to address secure ISAC tradeoffs, while the reviewed case studies organize design choices around BS and eavesdropper mobility. Static deployments can optimize positions from steady-state channel components, whereas beampattern behavior is also examined across region sizes and training epochs.

C. Existing Works of MA-Enabled Secure ISAC

  • Existing MA-enabled secure ISAC studies jointly adapt positions and other transmission controls to improve secrecy under sensing constraints, including IRS-assisted, near-field, and UAV-assisted settings.The review identifies dynamic-eavesdropper collaboration with sensing and secure communication as an open challenge in much prior work.Source: C. Existing Works of MA-Enabled Secure… · pp. 22–23

1) Static BS and Static Eve:

  • When the BS and Eve are static, steady-state channel parameters permit one-time antenna-position optimization that reduces Eve-channel correlation without requiring instantaneous Eve CSI, while preserving sensing and communication beams.The scenario’s key difficulty is distinguishing a legitimate user, sensing target, and potentially nearby Eve with limited FPA spatial degrees of freedom.Source: 1) Static BS and Static Eve · p. 23
  • The static-case beampattern evaluation reports that increasing the MA moving-region size affects pattern precision and that patterns evolve across training epochs while maintaining gain toward legitimate-user and Eve directions.The illustrated user directions are ±30° and ±60°, and the Eve direction is −15°.Source: Figure 12 · p. 23, Figure 13 · p. 23, 1) Static BS and Static Eve · p. 24

2) Static BS and Dynamic Eve:

For a static BS facing a mobile Eve, sensing echoes support Eve tracking and predicted CSI for transmission adaptation. For a mobile BS facing a static Eve, the review describes a UAV protocol that combines large-scale placement, local antenna adjustment, and slot-level beamforming.

2) Static BS and Dynamic Eve:

  • A fixed BS can treat a mobile Eve as a sensing target, reusing the probing waveform for target tracking and secure communication.The Eve state is estimated from reflected echoes using a matched filter, then EKF-based prediction supports CSI-informed adaptation and real-time null synthesis.Source: 2) Static BS and Dynamic Eve · p. 24

3) Dynamic BS and Static Eve:

  • A UAV BS facing a static Eve uses dual-spatial-scale transmission: optimize its 3D trajectory and hovering locations, fine-tune onboard antenna positions, and beamform in real time.The UAV serves downlink users and detects a target; its maneuverability can provide favorable legitimate links and sensing geometries, but can also expose transmissions to Eve.Source: 3) Dynamic BS and Static Eve · pp. 24–25

4) Dynamic BS and Dynamic Eve:

The section relates increasing mobility on both sides to increasingly predictive and coordinated secure-ISAC designs, then summarizes broader technical challenges and research directions. Across these topics, movable antennas add spatial design flexibility, while uncertainty, complexity, and practical constraints remain unresolved.

4) Dynamic BS and Dynamic Eve:

  • With both BS and Eve moving, the problem can be modeled as a non-cooperative Stackelberg game, with the network jointly designing beamforming, trajectory, and antenna positions under sensing and security requirements.The more general multi-UAV formulation jointly optimizes fleet placement, local antenna positions, and receive beamforming, with multi-agent learning and successive convex approximation among the cited solution techniques.Source: 4) Dynamic BS and Dynamic Eve · p. 25

E. Chapter Summary

  • Across mobility cases, countermeasures progress from deterministic spatial optimization in static settings to predictive, multi-agent allocation in fully dynamic networks while retaining sensing and secure communication.The chapter summary identifies robustness to channel uncertainty, dynamic mobility, and practical implementation constraints as open research directions.Source: 4) Dynamic BS and Dynamic Eve · p. 25, E. Chapter Summary · p. 25

A. Adaptive Beamforming and Antenna Positioning

  • Joint position and beamforming optimization can control beam directions and sidelobes, but its coupled non-convex structure and iterative updates create computational overhead, especially in large or dynamic systems.The discussion calls for lower-complexity methods, candidate-position selection, learning-assisted adaptation, and hardware-aware designs.Source: A. Adaptive Beamforming and Antenna… · p. 26

B. Low-Overhead and Scalable Parameter Estimation

  • Parameter estimation must capture position-dependent channel and sensing behavior, yet movement errors, motion limits, energy use, and spatiotemporal coupling complicate scalable estimation.The discussion calls for physically consistent models and low-overhead, robust methods that jointly consider trajectory design, measurement selection, and communication-sensing requirements.Source: B. Low-Overhead and Scalable Parameter… · p. 26

C. Security and Privacy Protection

  • MA mobility adds security and privacy risks under channel uncertainty and adversarial mobility, alongside tradeoffs among security, sensing accuracy, energy, and movement cost.The section calls for dynamic security models and privacy-preserving mechanisms under practical mobility and channel constraints.Source: C. Security and Privacy Protection · p. 26

D. Synergistic Integration of MA and RIS

  • MA and RIS offer complementary reconfiguration of the transmitter and propagation environment, but their joint optimization is coupled and high-dimensional, particularly in large or dynamic systems.Their cooperation remains largely unexplored, motivating low-complexity real-time control that accounts for hardware, mobility, and signaling overhead.Source: D. Synergistic Integration of MA and RIS · p. 26

VI. CONCLUSION

  • The tutorial synthesizes MA architectures and modeling, communication and sensing parameter-estimation methods, and PLS-based secure-ISAC approaches, concluding that MA adds a spatial design dimension for joint communication, sensing, and security.Its estimation review includes LS, subspace-based, CS-based, and tensor-decomposition methods, compared for accuracy and computational complexity.Source: §VI · p. 26

Related papers

Loading 2610.00218v1…