Movable Antenna-Assisted Integrated Sensing and Communication Systems

Basic Information

• Paper ID: 2501.01217
• Title: Movable Antenna-Assisted Integrated Sensing and Communication Systems
• Authors: Chengjun Jiang, Chensi Zhang, Chongwen Huang, Jianhua Ge, Dusit Niyato, Chau Yuen
• Classification: eess.SP (Signal Processing)
• Publication Date: January 2, 2025 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2501.01217

Abstract

Movable antennas (MAs) enhance the flexibility of beamforming gain and interference suppression by adjusting their positions within designated transceiver regions. This paper proposes an MA-assisted integrated sensing and communication (ISAC) framework that deploys MAs at both the receiver and transmitter of a base station to reconfigure the channel array response. An optimization framework is developed to maximize the sensing signal-to-interference-plus-noise ratio (SINR) through joint optimization of the receive beamforming vector, transmit beamforming matrix, and MA positions, while satisfying minimum SINR requirements for each user. To address the non-convex problem involving complex coupled variables, an alternating optimization algorithm is designed, incorporating techniques such as Charnes-Cooper transformation, second-order Taylor expansion, and successive convex approximation (SCA).

Research Background and Motivation

1. Problem Definition

With the development of 6G wireless technology, future industrial scenarios (such as intelligent transportation, autonomous driving, and industrial IoT) require not only high-quality wireless connectivity but also high-precision and robust sensing capabilities. Traditional separate design of communication and sensing systems results in low resource utilization efficiency, whereas integrated sensing and communication (ISAC) systems can achieve complete integration of radar sensing and communication functions on a unified platform.

2. Limitations of Existing Methods

• Fixed Position Antenna (FPA) Systems: Cannot fully exploit spatial variations in wireless channels within designated transceiver regions
• Large-Scale MIMO Systems: While providing numerous degrees of freedom, they suffer from large antenna sizes, high power consumption, and complex RF resource management
• Existing ISAC Optimization: Primarily focused on beamforming design, insufficiently exploiting the additional spatial degrees of freedom provided by antenna position optimization

3. Research Motivation

Movable antenna technology dynamically adjusts antenna positions, orientations, and configurations to reconfigure transmission channels and optimize signal reception and transmission. Compared to traditional FPA, MAs provide additional spatial degrees of freedom that can enhance beamforming gain, suppress interference in complex environments, and achieve performance levels unattainable by FPA.

Core Contributions

1. Proposed a Novel MA-Assisted ISAC Optimization Framework: Leveraging the additional spatial degrees of freedom provided by 2D MAs, jointly optimizing the receive beamforming vector, transmit beamforming matrix, and antenna positions
2. Designed an SCA Algorithm Based on Alternating Optimization: For the highly non-convex optimization problem, decomposing it into two classes of subproblems for beamforming and antenna position optimization, with convergence and complexity analysis provided
3. Derived Closed-Form Solutions and Theoretical Proofs: Providing closed-form expressions for the receive beamforming vector of radar echoes, obtaining the transmit beamforming matrix through Charnes-Cooper transformation and semidefinite relaxation, and proving it yields a rank-one solution
4. Verified Significant Performance Improvements: Simulation results demonstrate that the MA-assisted ISAC system achieves over 50% sensing SINR improvement compared to baseline schemes while requiring fewer antennas, RF chains, and power consumption

Methodology Details

Task Definition

Consider an ISAC system comprising a pair of transceiver base stations, K users, and a point target. The transmitter is configured with N MAs and the receiver with M MAs, both in planar array configurations. The optimization objective is to maximize the target sensing SINR while guaranteeing communication SINR requirements for all users.

System Model

Channel Model

Adopting a field response-based channel model utilizing far-field assumptions and plane wave models:

Communication Channel: The channel vector between user k and the base station is $h_k(t) = F_k^H(t)g_k$

where $F_k(t)$ is the field response matrix and $g_k$ is the path response vector.

Sensing Channel: The bidirectional channel matrix for target q is $H_q(r,t) = h_{r,q}(r)h_{t,q}^H(t)$

Signal Model

The transmitted signal comprises communication and radar signals: $s = \sum_{k=1}^K w_k s_{c,k} + \sum_{n=K+1}^N w_n s_{r,n}$

The receive SINR for user k is: $\gamma_{c_k} = \frac{|h_k^H(t)w_k|^2}{\sum_{n≠k}|h_k^H(t)w_n|^2 + σ_k^2}$

The target sensing SINR is: $\gamma_r = \frac{\sum_{n=1}^N |α_d u_r^H H_d(r,t)w_n|^2}{\sum_{l=1}^L \sum_{n=1}^N |α_l u_r^H H_l(r,t)w_n|^2 + σ_r^2||u_r||^2}$

Optimization Problem Formulation

The original optimization problem P1 is: $\max_{\{w_n\}_{n=1}^N, u_r, t, r} \gamma_r$$\text{s.t. } \gamma_{c_k} ≥ γ_{th_k}, \forall k$$t_n ∈ C_t, ||t_n - t_a||_2 ≥ D$$r_m ∈ C_r, ||r_m - r_b||_2 ≥ D$$\sum_{n=1}^N ||w_n||^2 ≤ P_{th}$

Alternating Optimization Algorithm

1. Receive Beamforming Optimization

Fixing other variables and optimizing $u_r$. This is a classical SINR maximization problem solved using the MVDR method to obtain the closed-form solution: $u_r^* = \frac{(\sum_{l=1}^L R_l' + σ_r^2 I)^{-1} d_d'}{d_d'^H (\sum_{l=1}^L R_l' + σ_r^2 I)^{-1} d_d'}$

2. Transmit Beamforming Optimization

Introducing matrix variable $W_n = w_n w_n^H$, using Charnes-Cooper transformation to handle fractional objective functions, and employing semidefinite relaxation for rank-one constraints. The relaxed solution is proven to naturally satisfy the rank-one condition.

3. Antenna Position Optimization

For receive and transmit antenna positions, employing second-order Taylor expansion and SCA methods respectively:

• Using auxiliary variables to convert fractional constraints into linear constraints
• Performing second-order Taylor expansion approximation of non-convex functions
• Linearizing minimum distance constraints

Technical Innovations

1. Dual-End MA Deployment: First simultaneous optimization of 2D movable antenna positions at both transmitter and receiver in ISAC systems
2. Joint Optimization Framework: Unified consideration of beamforming design and antenna position optimization
3. Theoretical Guarantees: Proving the tightness of semidefinite relaxation to ensure algorithm convergence to local optimal solutions
4. Efficient Solution: Converting complex non-convex problems into efficiently solvable convex subproblems through problem decomposition and convex approximation techniques

Experimental Setup

Simulation Parameters

• Number of users: K = 2, Number of clutter: L = 2
• Base station positions: Transmitter (0,0,5), Receiver (0,20,5)
• Target and clutter positions: (0,10,5), (10,10,5), (-10,10,5)
• User positions: (15,10,0), (-15,10,0)
• Number of propagation paths: $L_k^t = L_q^t = L_q^r = 4$
• Path loss exponents: Users 2.5, Target 2.2, Clutter 2.3
• Communication SINR threshold: $γ_{th} = 0$dB

Comparison Schemes

1. Proposed: Dual-end MA optimization scheme
2. Receive-MA: MA only at receiver, FPA at transmitter
3. Transmit-MA: MA only at transmitter, FPA at receiver
4. FPA: Fixed position antennas at both transmitter and receiver

Evaluation Metrics

• Sensing SINR performance
• Convergence performance
• Performance comparison under different system parameters

Experimental Results

Main Results

1. Convergence Performance

The algorithm converges after approximately 20 iterations, with antenna quantity having minimal impact on convergence speed, verifying the algorithm's fast convergence and effectiveness.

2. Transmit Power Impact

At $P_{th} = 25$dBm, the proposed scheme achieves approximately 57.54% performance improvement over FPA:

• Proposed scheme: approximately 12.5dB
• Receive-MA scheme: approximately 10.8dB (38.43% improvement)
• Transmit-MA scheme: approximately 9.5dB (21.12% improvement)
• FPA scheme: approximately 8.0dB

3. Antenna Quantity Impact

Transmit Antenna Quantity ($N = 4$ to $N = 10$, $M = 6$):

• Proposed: Improvement from 9.45dB to 12.49dB
• Receive-MA: Improvement from 6.37dB to 10.82dB
• Transmit-MA: Improvement from 2.98dB to 9.22dB

Receive Antenna Quantity ($M = 2$ to $M = 7$, $N = 6$):

• Proposed: Improvement from 6.44dB to 10.97dB
• Receive-MA: Improvement from 1.85dB to 9.21dB

4. Communication Requirement Impact

As communication SINR threshold increases from -5dB to 20dB:

• Proposed scheme: Slight decrease from 12.48dB to 11.44dB
• Other schemes show more pronounced decreases, demonstrating MA technology's advantage in balancing sensing and communication requirements

5. Movement Range Impact

Maximum performance is achieved when $A/λ = 2.6$, indicating that antenna spatial requirements are bounded, supporting the trend of maintaining antenna miniaturization in future applications.

Ablation Studies

Through comparison of different MA configuration schemes, validating:

1. Significant advantages of dual-end MA deployment over single-end deployment
2. More pronounced interference suppression effects from receiver-end MA
3. Transmitter-end MA primarily improving signal coverage and directivity

Case Analysis

• Beam Pattern Analysis: Transmit-MA scheme achieves higher gain in target direction compared to FPA, with narrower main lobe and better sidelobe suppression
• Channel Gain Analysis: The optimization algorithm moves antennas to high-power gain regions while considering both main signal enhancement and interference suppression

Related Work

ISAC System Research

• MIMO Technology: Leveraging large-scale MIMO and millimeter-wave technology to enhance ISAC performance
• Antenna Selection: Optimizing sensing-communication tradeoffs through transmit/receive antenna selection
• RIS-Assisted: Enhancing ISAC system performance using reconfigurable intelligent surfaces

Movable Antenna Research

• Communication Applications: MA applications in MIMO communication, physical layer security, and multiple access
• Sensing Applications: MA applications in wireless sensing and multi-beam forming
• ISAC Applications: Existing work primarily considers one-dimensional MA or single-end deployment

Contribution of This Work

Compared to existing work, this paper is the first to propose dual-end 2D MA-assisted ISAC systems with a complete optimization framework and theoretical analysis.

Conclusions and Discussion

Main Conclusions

1. MA technology significantly enhances ISAC system performance, with dual-end deployment achieving optimal results
2. Receiver-end MA demonstrates more pronounced interference suppression effects
3. The proposed algorithm effectively solves complex joint optimization problems
4. MA technology reduces resource requirements while maintaining performance

Limitations

1. Channel State Information Assumptions: Assumes perfect CSI; practical applications require consideration of channel estimation errors
2. Single-Target Scenario: Primarily considers single-target sensing; multi-target scenarios require further extension
3. Computational Complexity: Computational complexity in large-scale systems may become a bottleneck
4. Hardware Implementation: Practical deployment and control mechanisms for MAs require further research

Future Directions

1. Distributed optimization and cooperative beamforming techniques
2. Hierarchical antenna control strategies
3. Synergistic integration of MAs with other advanced hardware technologies
4. Robust optimization design under imperfect CSI

In-Depth Evaluation

Strengths

1. Technical Innovation: First proposal of dual-end 2D MA-assisted ISAC systems with novel technical approach
2. Theoretical Completeness: Providing comprehensive problem modeling, algorithm design, and theoretical analysis
3. Algorithm Effectiveness: Cleverly solving complex non-convex optimization problems through multiple mathematical tools
4. Experimental Sufficiency: Comprehensive simulation verification of method effectiveness and superiority

Weaknesses

1. Practical Considerations: Perfect CSI assumption is difficult to satisfy in practice
2. Hardware Constraints: Insufficient consideration of MA movement speed and precision limitations
3. Scalability Analysis: Lacking complexity and scalability analysis for large-scale systems
4. Comparison Baselines: Missing comparisons with other ISAC optimization methods

Impact

1. Academic Value: Provides important theoretical foundation for MA applications in ISAC systems
2. Practical Value: Offers new technical pathways for 6G ISAC system design
3. Reproducibility: Detailed algorithm descriptions and complete simulation parameters ensure good reproducibility

Applicable Scenarios

• 6G wireless communication systems
• Intelligent transportation and autonomous driving
• Industrial IoT sensing applications
• Millimeter-wave/terahertz communication systems

References

The paper cites 51 relevant references covering important works in ISAC systems, movable antenna technology, optimization theory, and other domains, providing a solid theoretical foundation for the research.

This paper makes important contributions to MA-assisted ISAC system design. The proposed method possesses strong theoretical value and application prospects, providing new insights for future 6G systems' integrated sensing and communication design.