Annu Annu

Annu Annu

Auto-generated profile, based on publicly available information

Annu Annu

Indian Institute of Technology, Hyderabad, India

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 February 2024 in IEEE Open Journal of Vehicular Technology
Verified
Journal Article
Open access
ANAnnu
Annu
RPRajalakshmi Pachamuthu
Rajalakshmi Pachamuthu
The advent of 6G marks a transformative phase in wireless communication, ushering in a hyperconnected experience. This paper explores optimizing sidelink technologies in Vehicle-to-Everything (V2X) communication through integrating 6G capabilities. Emphasizing challenges in sidelink resource allocation, the study introduces mathematical solutions. The survey further investigates the evolution of Cellular-V2X (C-V2X) and sidelink standardization within the 6G V2X Sidelink context, highlighting key features and applications. Additionally, it examines inter-UE coordination, resource re-evaluation, and pre-emption operations in 5G-V2X Sidelink, addressing associated challenges and mathematical formulations. The paper focuses on power-based resource allocation in 5G-V2X, addressing challenges and proposing solutions for the 6G V2X Sidelink landscape. Encompassing direct communication, collision issues, spectrum compliance, resource fairness, uncertainty, and interference management, the survey comprehensively explores challenges and solutions in current sidelink resource allocation. It evaluates traditional and emerging techniques, such as cognitive radio, cooperative communication, power control, dynamic spectrum access, ML-aided allocation, blockchain-enabled allocation, and edge computing-driven allocation. The resource allocation requirements for diverse V2X services in 6G V2X Sidelink are outlined, explicitly focusing on Vehicular to Vehicular (V2V), Vehicular to Infrastructure (V2I), Vehicular to Pedestrian (V2P), and other V2X services, addressing their specific needs.

7 April 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
AAAnnu Annu
Annu Annu
RPRajalakshmi Pachamuthu
Rajalakshmi Pachamuthu
JSJim Sayer
Jim Sayer
We propose a distributed framework for traffic congestion estimation in 5G NR-V2X sidelink (Mode 2) networks that integrates rate–distortion–based compression with consensus filtering. The resulting Compressed Consensus Estimator (CCE) enables vehicles to exchange compact, quantized estimates within existing SCI/PSSCH fields, preserving full 3GPP compliance without requiring extra spectrum. Analytical results establish mean-square error bounds, convergence under quantization and packet loss, and largedeviation reliability. The framework is validated through both system-level simulations (WiLabV2X + SUMO) and a real vehicular dataset, showing that CCE achieves estimation accuracy close to the centralized Cramér–Rao bound, converges within one second, and remains robust under dense traffic and high packet loss. Compared with uncompressed consensus, adaptive gossip, and observer-based baselines, CCE consistently yields lower estimation error and faster convergence, while meeting NR-V2X latency and scalability requirements, enabling practical ITS deployment.


24 June 2026 in Vehicular Communications
Verified
Journal Article
Closed access
ANAnnu
Annu
RPRajalakshmi Pachamuthu
Rajalakshmi Pachamuthu
JSJames R. Sayer
James R. Sayer

9 March 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
AAAnnu Annu
Annu Annu
RPRajalakshmi Pachamuthu
Rajalakshmi Pachamuthu
Persistent V2X Mode 2 degrades communication reliability, limiting the effectiveness of standard Semi-Persistent Scheduling (SPS). We propose MarkovGNN, a hybrid approach that integrates Markov chain modeling with Graph Neural Networks (GNNs) to predict and mitigate collisions. By modeling resource selection dynamics and leveraging graph-based learning, Markov-GNN enables adaptive, decentralized scheduling. Trained on pre-processed sidelink NR V2X dataset and validated in WiLabV2X simulations, it significantly reduces collision durations and improves packet delivery, outperforming SPS, reinforcement learning (RL) baselines, and graph-based heuristics. Statistical validation confirms its ability to drive the network toward a collision-free equilibrium, making it a robust solution for ultra-reliable V2X communications. Mathematically, the Markov–GNN chain is shown to reach collisionfree absorption in finite time with probability 1. Our Markov-GNN solution is lightweight enough to operate in near real-time on typical vehicular hardware, thereby ensuring practical deployment feasibility.

9 June 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
AAAnnu Annu
Annu Annu
RPRajalakshmi Pachamuthu
Rajalakshmi Pachamuthu
Reliable resource sensing and selection are critical for collision avoidance in 5G NR V2X Mode 2 networks operating without centralized coordination. Conventional sensing-based semi-persistent scheduling (SPS) relies primarily on received-power measurements and fixed-threshold energy detection, which can suffer from noise uncertainty, weak-signal miss detections, and hidden-terminal collisions in dense vehicular environments. To address these limitations, this paper proposes an Eigenvalue-Based Spectrum Sensing and Scheduling (EBSS) framework that integrates covariance-spectrum analysis into the NR V2X sensing process. The proposed approach derives an eigenvalue-ratio sensing statistic from a binary hypothesis-testing formulation and employs covariance eigen-structure information to improve resource-occupancy detection. Theoretical analysis is developed using covariance perturbation theory, eigenvalue inequalities, and random matrix theory to characterize detector behavior, threshold selection, robustness, and computational complexity. The resulting sensing metric is incorporated into the standard SPS resource-selection procedure while preserving compatibility with existing NR V2X Mode 2 operation. Simulation results under dynamic vehicular scenarios demonstrate significant reductions in collision probability, improvements in packet delivery ratio and latency, and enhanced sensing reliability compared with conventional sensing-based SPS approaches. The proposed framework achieves these gains without requiring channel estimation, pilot signaling, centralized coordination, or machine-learning training, making it a practical and scalable solution for future NR V2X networks.