A capacity broker architecture and framework for multi-tenant support in LTE-A networks

Abstract

Resource provisioning in multi-operator scenarios requires an estimate of the tenants' traffic needs. This is necessary in the scenario where a Mobile Network Operator (MNO) owns the Radio Access Network (RAN) and many Mobile Virtual Network Operators (MVNOs) act as resellers of their host network's capacity under their own brands, to their own customers. In such scenarios, the forecasted MVNO traffic is the basis for providing resources suitable with the corresponding MVNOs demand. To that end, the dynamic provision of resources among MVNOs should be performed in flexible, short-term time scales. In this paper, we effectively address this issue by integrating the capacity broker into the 3rd Generation Partnership Project (3GPP) network management network architecture using the minimum set of enhancements. In addition, to fully exploit its capabilities, we propose the Multi-tenant Slicing (MuSli) of capacity algorithm, to allocate resources towards MVNOs in coarse time scales. MuSli considers the estimated capacity and the impact of the traffic type (i.e., guaranteed QoS and Best-Effort) in each MVNO, to provide better utilization of the host network's capacity. Our results highlight the gains in the number of served requests without compromising their service quality.

RL-NSB: Reinforcement Learning-Based 5G Network Slice Broker

“…Unlike our proposal, this study introduces new 3GPP interfaces to accommodate the broker functionality. A scheme that integrates the capacity broker with enhancements on the 3GPP architecture is documented in [14]. Such capacity broker forecasts the network capacity when allocating guaranteed and best-effort slices, considering their respective SLAs.…”

“…If the total profit v(•) assigned to the 2-slice set requests C l * is greater than the half of the best profit retrieved after running all knapsack problems, we keep C l * as the best feasible set (line 10-11). Otherwise, we split the best set into two subsets F 1 and F 2 (line [13][14], where C l * is assigned to F 1 while the remaining items to F 2 . In this case, F 1 has a total profit less than the half of profit of the best solution whereas F 2 shows a total profit greater than the half of the best solution (recall that the sum of the profits of both subsets gives exactly the profit of the best solution).…”

Software Defined Network Service Chaining for OTT Service Providers in 5G Networks

“…), owned and managed by different NSPs, and are served by data centers with different capabilities and architectural design. Hence, the 5G network design should facilitate NSC for OSPs', offering two fundamental functionalities [7]:…”

Implementation of an SDN-Enabled 5G Experimental Platform for Core and Radio Access Network Support

“…Small-scale platforms have also been proposed [4] using single-board computers as SDN switches to lower the deployment cost. Other works address the Radio Access Network (RAN) part of 5G networks, focusing on RAN Virtualization [5] [6], or flexible resource allocation [7]. However, these are largely conceptual and do not consider the associated implementation challenges of real-world RANs.…”