Title:  Towards Energy-efficient Wi-Fi Connectivity for the Internet of Things

Committee: 

Dr. Sivakumar, Advisor 

Dr. Sundaresan, Chair

Dr. Blough

Abstract: The objective of the proposed research is to study and enhance the energy-efficiency of Wi-Fi connectivity in lower-power battery-operated Internet of Things (IoT) devices. Wireless communication technologies have been making steady progress to support ultra-dense deployments as the number of connected devices globally is expected to surpass 28 billion units in 2023. It is also projected that over 14.7 billion connections will be Internet of Things (IoT) devices such as smart thermometers, soil moisture sensors, surveillance cameras, and inventory tracking devices. Among the competing wireless technologies, Wi-Fi is fast emerging as one of the prime choices for wireless connectivity for the Internet of things (IoT), particularly for indoor and medium-range applications (<100 meters). The power consumed by the Wi-Fi interface has always been a topic of interest and concern. This concern is significantly more pronounced in low-power devices such as IoT sensors where theWi-Fi interface can consume up to 78% of the total power while just maintaining connectivity with the access point. As a result, a significant proportion of IoT use cases require power-saving strategies. The Power Save Mechanism (PSM), introduced by the IEEE 802.11-2012 standard, is a widely used low-power operation technique for Wi-Fi client devices. We find that without any traffic, a continuously active Wi-Fi connected IoT module lasts only for 2 days while an STA using PSM lasts for over 6 months. We experimentally analyze a battery-operated Wi-Fi connected IoT module to accurately measure the impact of different network operations. We show that the battery life of the module is limited, while running thin uplink traffic, to ∼ 30% of its battery life on an idle connection, even when utilizing IEEE 802.11 PSM. We propose standard-compliant client-side strategies to increase power savings at client IoT module. The drastic increase in the average number of devices supported by wireless base stations is an issue for severely power-limited devices as high network congestion leads to longer active periods owing to increased contention. We quantify the effect of congestion on Wi-Fi power consumption by evaluating the performance of PSM with sparse uplink traffic under high channel congestion conditions. We outline the power-saving potential of Target Wake Time (TWT), a scheduling mechanism included in Wi-Fi 6, with preliminary results in a congested network. We discuss the potential to utilize TWT as a mechanism to increase power consumption and channel efficiency and to reduce co-channel interference.