Compresso: Latency-Aware Transmission of Compressed IoT Measurement Data Over SDN

Abstract

Measurement data obtained from “things” in the Internet of Things (IoT) faces challenges in efficient transmission due to the low-bandwidth data transmission link. We observe that measurement data are fixed in size and format, and low-entropy in the time domain, indicating that compression can be benefited. Rather than employing a single compression algorithm as advocated in existing literature, we argue that optimal transmission can be achieved by jointly considering compression overheads and network status, where software-defined networking (SDN) is employed to enforce network statistics and packet forwarding. This article presents a new paradigm that achieves optimal transmission of SDN-empowered compressed measurement data. We formulate the problem as an optimization problem and prove its nonpolynomial hardness time complexity. Due to this complexity, we introduce Compresso, a heuristic algorithm that efficiently solves the problem. We conduct rigorous simulations, and the results demonstrate the efficiency of the new paradigm and Compresso, i.e., attaining comparable performance to the optimal solution with 50% time usage reduction.