Abstract: In recent times, Vehicle-to-Everything (V2X) networks have gained significant attention due to their role in enabling safety-critical and delay-sensitive applications such as cooperative driving, platooning, extended sensing, and autonomous mobility. V2X networks allow vehicles to communicate with surrounding vehicles, roadside infrastructure, pedestrians, and network entities to improve road safety and traffic efficiency. Integrated sensing and communication (ISAC) further enables shared radio resources to support both information exchange and environment perception in vehicular networks. However, practical V2X systems also involve computation-intensive sensing and processing workloads, particularly for radar-based perception, cooperative awareness, and edge-assisted decision making. This motivates integrated sensing, communication, and computation (ISCC)-aware resource allocation for NR-V2X networks.

In this talk, we present MAPPO-SPS, an ISCC-aware cross-layer resource allocation and scheduling framework for 5G NR-V2X networks. The proposed framework extends conventional sensing-based semi-persistent scheduling (SB-SPS), the standard distributed resource allocation scheme in NR-V2X Mode 2, by jointly considering sensing, communication, and computation resources. The scheduling problem is modeled as a cooperative partially observable Markov game and solved using multi-agent proximal policy optimization with centralized training and decentralized execution. To evaluate the effectiveness of the proposed method, MAPPO-SPS is compared with benchmark scheduling schemes in terms of communication reliability, sensing accuracy, latency, throughput, and energy consumption.

Event Details
Title: Resource Allocation and Scheduling for Integrated Sensing, Communication, and Computation in NR-V2X Networks
Date: June 09, 2026 at 11:00 AM
Venue: ESB 244/Google Meet (https://meet.google.com/tqn-fmux-qkd)
Speaker: Ms. Indulekha K P (EE22D006)
Guide: Dr. Venkatesh T G
Type: PHD seminar

Updated: