INVITED: Context-Aware Energy-Efficient Communication for IoT Sensor Nodes

The widespread proliferation of sensor nodes in the era of Internet of Things (IoT) coupled with increasing sensor fidelity and data-acquisition modality is expected to generate 30+ Exabytes of data per month by 2020. In this data driven IoT world, wireless communication is a significant consumer of energy, and paying careful attention to the balance between local and remote computation is critical to overall energy usage. The communication fabrics that will handle this enormous amount of IoT workload will need to be energy-efficient under changing contexts such as channel conditions, applications, QoS, data-rate requirements etc. Moreover, the IoT devices will often include multiple parallel communication fabrics; e.g. wired, proximity, mm-wave, 5G etc. We will discuss how self-learning can enable context-aware operation in such communication systems to allow minimum energy/bit and energy/information for any given communication scenario. The need for context-aware operation within and among multiple physical layers (PHYs) in future IoT workloads will be highlighted. Such energy-efficient communication (Shannon's Law) along with low-power computing (Moore's Law), is expected to harness the true potential of the IoT revolution and produce dramatic societal impact.