HyRA: A Hybrid Resource Allocation Framework for RAN Slicing

TL;DR

This work introduces HyRA, a hybrid resource allocation framework for RAN slicing that combines dedicated per-slice allocations with shared resource pooling across slices and achieves up to 50-75% spectrum savings compared to dedicated-only and shared-only baselines.

Abstract

The advent of 5G and the emergence of 6G networks demand unprecedented flexibility and efficiency in Radio Access Network (RAN) resource management to satisfy diverse service-level agreements (SLAs). Existing RAN slicing frameworks predominantly rely on per-slice resource reservation, which ensures performance isolation but leads to inefficient utilization, particularly under bursty traffic. We introduce HyRA, a hybrid resource allocation framework for RAN slicing that combines dedicated per-slice allocations with shared resource pooling across slices. HyRA preserves performance isolation while improving resource efficiency by leveraging multiplexing gains in bursty traffic conditions. We formulate this design as a bi-level stochastic optimization problem, where the outer loop determines the dedicated and shared resource budgets and the inner loop performs per-UE scheduling under a novel water-filling approach. By using the sample-average approximation, the Karush-Kuhn-Tucker (KKT) conditions of the inner loop, and Big-M encoding, we transform the problem into a tractable mixed-integer program that standard optimization solvers can solve. Extensive simulations under diverse demand patterns, SLA configurations, and traffic burstiness show that HyRA achieves up to 50-75% spectrum savings compared to dedicated-only and shared-only baselines. These results highlight HyRA as a viable approach for resource-efficient, SLA-compliant RAN slicing in future mobile networks.

Figures (6)

• Figure 1: Water-filling interpretation of the inner loop problem. Subplots (a)–(d) show allocations under different dedicated and shared budgets for the same channel conditions across $4$ UEs (first two belonging to slice $1$, last two to slice $2$). Each stacked bar depicts the baseline inverse channel gain, followed by the allocated dedicated resources, and finally the shared resources per UE.
• Figure 2: Resource allocation for two network slices to deliver the desired SLA per UE. Each group of bars corresponds to a combination of UE counts and delay budgets of the two slices.
• Figure 3: Resource allocation for two network slices to deliver the desired SLA aggregated over slice UEs. Each group of bars corresponds to a combination of UE counts and delay budgets of the two slices.
• Figure 4: Impact of slice scaling on resource allocation under different delay budgets. Each group of bars corresponds to a combination of slice count and the corresponding delay budgets.
• Figure 5: Impact of traffic burstiness on resource allocation under different delay budgets. Lower $\alpha$ indicates higher burstiness.