Data Encoding Techniques for Reducing Energy Consumption in Network-On-Chip

Abstract

Energy consumption has become a major challenge in on-chip communication architectures such as Network-on-Chip as a result of the growing complexity of system-on-chips (SoCs) (NoC). Encoding techniques have recently gained attention as a potentially useful strategy for lowering the amount of power that a NoC design requires. These techniques intend to alter the manner in which data is transmitted through the communication stack in order to cut down on the number of transitions in the data and, as a consequence, lower the amount of power that is consumed. This paper presents an overview of data encoding techniques that can be used for NoC and discusses the effectiveness of these techniques in lowering the amount of energy that is consumed. Additionally, the importance of energy efficiency as a design constraint for modern SoCs is emphasised throughout the course of this paper. Based on the findings, it appears that data encoding techniques are a viable option for lowering the level of energy consumption in NoCs and have a significant bearing on the level of energy efficiency achieved by contemporary SoCs.