Intelligent Reflecting Surface Assisted D2D Hybrid Communication for Internet of Vehicles: A Comprehensive Resource Optimization Strategy

Introduction

The rapid development of information and communication technologies has positioned the Internet of Vehicles (IoV) as a critical component of intelligent transportation systems. IoV enables vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication, facilitating applications such as autonomous driving, vehicle sensing, and precise positioning. However, urban IoV environments face significant challenges, including line-of-sight (LOS) link blockage due to buildings and vegetation, as well as limited spectrum resources. These issues degrade communication reliability and data transmission efficiency, necessitating innovative solutions to enhance network performance.

To address these challenges, this work proposes a novel hybrid communication framework that integrates Intelligent Reflecting Surfaces (IRS) and Device-to-Device (D2D) technologies in IoV scenarios. IRS, a passive metasurface with reconfigurable phase shifts, enhances signal coverage by intelligently reflecting incident signals to overcome obstructions. Meanwhile, D2D communication allows direct data exchange between nearby vehicles, reducing latency and improving spectral efficiency. By combining IRS with D2D and leveraging cognitive radio (CR) techniques, the proposed system optimizes resource allocation to maximize the sum rate while ensuring quality-of-service (QoS) requirements.

System Model and Problem Formulation

The proposed IRS-D2D hybrid communication system is designed for urban intersections where LOS links between base stations (BS) and vehicles are frequently obstructed. The network consists of a BS, an IRS with N reflecting elements, cellular vehicles (CVs), and D2D vehicles (DVs). The BS, typically installed at traffic lights, transmits road condition information, while the IRS, mounted on roadside structures, establishes alternative reflection paths to CVs when direct links are blocked. DVs communicate directly using D2D technology, particularly when they are outside the BS’s coverage range.

The system employs CR to improve spectral efficiency, allowing D2D users to share the same frequency bands as CVs under an interference threshold constraint. The total available bandwidth is divided into K orthogonal subchannels, each allocated to one CV and one DV pair. The channel gains follow Rayleigh fading, incorporating path loss and shadowing effects. The IRS phase shift matrix is optimized to maximize signal reflection toward intended receivers while minimizing interference.

The primary objective is to maximize the sum rate of both CV and DV links by jointly optimizing power allocation, IRS phase shifts, and spectrum resource allocation. The optimization problem is formulated with constraints on transmit power, interference thresholds, QoS requirements, and IRS reflection coefficients. However, this problem is non-convex and computationally complex, requiring an efficient solution approach.

Two-Stage Alternating Optimization Algorithm

To solve the non-convex optimization problem, a two-stage alternating optimization (AO) algorithm is proposed. The first stage focuses on optimizing power allocation and IRS phase shifts, while the second stage addresses spectrum resource allocation.

Stage 1: Joint Power Allocation and IRS Phase Shift Optimization

In this stage, the sum rate of CV links is maximized by optimizing transmit power and IRS reflection coefficients while keeping spectrum allocation fixed. The problem is decomposed into two subproblems:

Power Allocation Optimization: Given the IRS phase shifts, the power allocation problem is transformed into a convex form using slack variables and first-order Taylor approximations. The CVX solver is employed to obtain locally optimal power values that satisfy QoS and interference constraints.
IRS Phase Shift Optimization: With power allocation fixed, the IRS reflection coefficients are optimized to enhance signal reflection. The problem is reformulated using relaxation techniques to handle non-convex constraints, and CVX is again used to find the optimal phase shifts.

Stage 2: Spectrum Resource Allocation

Using the optimized power and phase shifts from Stage 1, the second stage optimizes subchannel allocation to maximize the overall system sum rate. The binary channel allocation variables are relaxed into continuous variables, and the problem is solved iteratively using alternating optimization. The algorithm ensures that each subchannel is assigned to at most one CV and one DV pair while meeting interference and QoS constraints.

Complexity Analysis

The AO algorithm’s complexity is dominated by the convex optimization steps in both stages. The overall complexity is polynomial, scaling with the number of vehicles, IRS elements, and subchannels. Despite its iterative nature, the algorithm converges efficiently, making it suitable for real-time IoV applications.

Performance Evaluation

Simulations are conducted to evaluate the proposed IRS-D2D hybrid communication system under various conditions. Key performance metrics include sum rate, spectral efficiency, and the impact of IRS elements, vehicle speed, and network density.

Impact of IRS Elements

The number of IRS reflecting elements significantly influences system performance. Results show that increasing the number of IRS elements enhances the sum rate due to improved signal reflection and coverage. However, beyond a certain point, additional elements provide diminishing returns due to increased interference paths. The proposed AO algorithm outperforms benchmark schemes such as random phase shift (Random), greedy optimization (Greedy), and fixed power allocation (RP), demonstrating superior spectral efficiency.

Effect of Vehicle Speed

Vehicle mobility affects channel conditions and, consequently, communication performance. As vehicle speed increases, the sum rate of BS-IRS-CV links decreases due to larger inter-vehicle distances and higher Doppler effects. However, the AO algorithm maintains better performance compared to other methods, particularly at higher transmit power levels, by dynamically adjusting power and phase shifts to mitigate mobility-induced degradation.

Network Density and Spectral Efficiency

The system’s spectral efficiency improves with the ratio of DVs to CVs, as D2D communication enables efficient spectrum reuse. However, excessive D2D links can introduce interference, necessitating careful resource allocation. The IRS-assisted system achieves higher spectral efficiency than non-IRS scenarios, validating the benefits of IRS in dense urban environments.

Conclusion

This work presents a comprehensive resource optimization strategy for IRS-assisted D2D hybrid communication in IoV. By integrating IRS with D2D and CR technologies, the proposed system addresses LOS blockage and spectrum scarcity challenges in urban vehicular networks. The two-stage AO algorithm efficiently optimizes power allocation, IRS phase shifts, and spectrum allocation to maximize the sum rate while ensuring QoS requirements. Simulation results demonstrate significant performance gains over existing schemes, highlighting the potential of IRS-D2D hybrid communication for future IoV deployments. Future research directions include extending the framework to multi-IRS scenarios and incorporating machine learning for dynamic resource management.

doi.org/10.19734/j.issn.1001-3695.2024.06.0223

Was this helpful?

0 / 0