Performance evaluation of Networked Music Performance (NMP) over 5G scenarios through a Python-based simulator

Networked Music Performances (NMPs) enable remote musicians to perform together by means of low-latency audio/video streaming over a telecommunication network. Latency and reliability are two key parameters for the end-to-end transmission of audio information through networks in this context. The stringent requirements of NMPs with respect to those parameters pose a major challenge for 5G networks, which are expected to deliver significant Key Performance Indicator (KPI) improvements with respect to 4G. The objective of this thesis is to develop a Python-based simulator to define the placement of virtual machines operating as audio mixing servers within a set of candidate network nodes, while complying to the latency requirements of NMPs. The simulator emulates an optical network functioning as the backend for a 5G access network. Several users are connected to the 5G network, participating in remote musical interactions to simulate a Networked Music Performance scenario. This study aims to assess the performance and efficiency of the heuristic method, benchmarking it against optimal solutions derived from two Mixed Integer Linear Programming (MILP) models in terms of end-to-end latency, server load and computational timings.