Joint Bandwidth Orchestration and User Association in 5G Network Slicing

Network slicing requires the on-demand resource allocation according to the requirements from different slices. In this paper, we aim to incorporate an enhanced Mobile Broadband (eMBB) slice and an ultra-Reliable Low-Latency Communications (uRLLC) slice in the same substrate network, where the eMBB slice aims to receive a high throughput and the uRLLC slice requires a low transmission delay. To achieve the different requirements, we jointly consider user association and bandwidth orchestration in the substrate network, and formulate the problem as a mixed-integer programming (MIP). We propose two different approaches to solve the MIP, i.e., a high-complexity exhaustive search (ES) algorithm and a low-complexity high probability guarantee heuristic (HPGH) algorithm. Simulation results show that our proposed approach can seek the tradeoff between throughput and transmission delay, and allocate on-demand bandwidth resources dynamically.