eMBB-URLLC Resource Slicing: A Risk-Sensitive Approach (1902.01648v1)

Abstract: Ultra Reliable Low Latency Communication (URLLC) is a 5G New Radio (NR) application that requires strict reliability and latency. URLLC traffic is usually scheduled on top of the ongoing enhanced Mobile Broadband (eMBB) transmissions (i.e., puncturing the current eMBB transmission) and cannot be queued due to its hard latency requirements. In this letter, we propose a risk-sensitive based formulation to allocate resources to the incoming URLLC traffic while minimizing the risk of the eMBB transmission (i.e., protecting the eMBB users with low data rate) and ensuring URLLC reliability. Specifically, the Conditional Value at Risk (CVaR) is introduced as a risk measure for eMBB transmission. Moreover, the reliability constraint of URLLC is formulated as a chance constraint and relaxed based on Markov's inequality. We decompose the formulated problem into two subproblems in order to transform it into a convex form and then alternatively solve them until convergence. Simulation results show that the proposed approach allocates resources to the incoming URLLC traffic efficiently while satisfying the reliability of both eMBB and URLLC.

• The paper introduces a risk-sensitive CVaR framework to balance resource allocation between eMBB and URLLC in 5G networks.
• It decomposes the non-convex scheduling problem into eMBB user allocation and URLLC placement subproblems using convex optimization and Markov’s inequality.
• Simulations demonstrate that the approach maintains eMBB reliability above 90% under various URLLC loads, ensuring equitable service.

Overview of eMBB-URLLC Resource Slicing: A Risk-Sensitive Approach

This paper explores the intersection of enhanced Mobile Broadband (eMBB) and Ultra Reliable Low Latency Communications (URLLC) within the context of 5G New Radio (NR) technologies. eMBB aims to maximize peak data rates for mobile broadband applications, whereas URLLC targets services with stringent reliability and latency requirements, such as autonomous vehicles and industrial IoT devices.

Risk-Sensitive Resource Allocation Strategy

The authors propose a risk-sensitive framework using Conditional Value at Risk (CVaR) to optimize resource allocation between eMBB and incoming URLLC traffic. Traditional approaches that favor maximizing average data rates may disadvantage eMBB users with lower data rates when puncturing occurs. By leveraging CVaR, the risk, particularly for eMBB users experiencing low data rates, is mitigated.

Problem Formulation and Solution Approach

To address the non-convex nature of the problem, the authors decompose it into two subproblems targeting eMBB user scheduling and URLLC placement. They utilize Markov’s inequality to relax the URLLC reliability constraint, transforming it into a linear form that can be efficiently solved. The solution iterates between these subproblems, leveraging convex optimization techniques to achieve convergence.

Simulation Outcomes

Simulation results demonstrate the efficacy of the proposed approach in maintaining eMBB reliability above 90% across varying URLLC loads. Moreover, the results elucidate the trade-offs between eMBB data rates and URLLC reliability constraints. In scenarios of increased URLLC demand, the risk-sensitive method ensures equitable allocation amongst eMBB users without disproportionately affecting those with lower bandwidth allocations.

Implications and Future Prospects

The risk-sensitive resource slicing approach presents significant implications for optimizing 5G NR traffic management. By safeguarding eMBB users with lower data rates in high-demand contexts, this methodology enhances overall network performance. Future research could expand upon this foundation to include more complex network architectures and to explore dynamic market-based resource allocation mechanisms. Additionally, the integration of machine learning models for predictive resource management may further refine this approach.

In summary, the paper offers a theoretically robust and practically viable solution to the dynamic multiplexing of URLLC traffic in 5G networks, fostering improved quality of service across diversified use cases. Researchers and system architects engaged in optimizing multi-service 5G NR deployments may find the insights and methodologies presented beneficial for advancing both network reliability and user experience.