LLP-V2X: Low Latency-Power Vehicular Networking Towards 6G V2X
Zhaoyu Liu, Liu Cao, Lyutianyang Zhang, Dongyu Wei, Ye Hu, Weizheng Wang
TL;DR
To address energy-latency trade-offs in 6G V2X, this work proposes a multi-hop, multi-path architecture that jointly optimizes traffic splitting and per-path power, enabling end-to-end low latency under power constraints. It formalizes two convex optimization problems: E2E Low-Latency Multi-hop Multi-path Optimization (minimize $\sum_q \max_b u_{q,b}(n)$ subject to $C1$-$C5$) and Low-Power Multi-hop Multi-path Optimization (minimize $P(n)=\sum P_{i,j,k}(n)$ with latency targets $L_q^{max}$). It also introduces an Adaptive LLP MHMP Scheduler that switches between LL and LP modes based on observed tail-latency and channel state, solving for $\alpha_{i,j,k,q}(n)$ and $P_{i,j,k}(n)$ block-by-block. Simulation results show that MHMP outperforms single-path and path-selection baselines in both latency and power, with larger gains as the hop count increases. The framework offers a scalable route to energy-efficient, low-latency V2X in future 6G deployments and can be extended to IRS-assisted scenarios.
Abstract
The trade-off between energy and latency budgets is becoming significant due to the more stringent QoS requirements in 6G vehicular networks. However, comprehensively studying the trade-off between energy and latency budgets for 6G vehicular network with new Vehicle-to-Everything (V2X) features is still under-explored. This paper proposes a novel multi-hop, multi-path vehicular networking that jointly optimizes vehicular traffic splitting across candidate routes and per-link transmit power to achieve low-latency and low-power communications. Afterwards, we formalize two complementary problem formulations (minimum latency and minimum power) based on the proposed 6G V2X architecture and provide sufficient conditions. The performance of the proposed scheme is evaluated via well-designed simulations. Based on these theories, we design algorithm (LLP MHMP Scheduler) that switches on demand between a fixed-power minimum-latency mode and a fixed-latency minimum-power mode.
