An Online Combinatorial Auction Based Resource Allocation and Pricing Mechanism for Network Slicing in 5G

Abstract

The wide variety of applications that appear in fifth-generation mobile network (5G) require the mobile network to provide different services. In this background, the concept of network slicing based on virtualization is proposed. Network virtualization decouples the infrastructure of the network from the services provided and the resources of isolated network slices abstract from the infrastructure resources of infrastructure providers (InPs). The resources included in each slice are isolated from each other, and the resources of each slice are allocated to the subordinate service providers (SPs) by the mobile virtual network operator (MVNO) corresponding to the slice. Therefore, an effective resource allocation mechanism is important for network slicing. This paper designs an online combinatorial auction based resource allocation and pricing mechanism for MVNO, which applies the theory of exclusive or (XOR) auction, and this mechanism is invoked to calculate the allocation results and the payments of winning SPs when new requests arrive or resources are released. The proposed mechanism is proved to be individual rational and incentive compatible, and the numerical simulations testify that this mechanism can also achieve higher social welfare and service provider loading rate.