Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Document Type

Campus-Only Access for Five (5) Years

Degree Name

Doctor of Philosophy (PhD)

Degree Program

Computer Science

Year Degree Awarded


Month Degree Awarded


First Advisor

Deepak Ganesan

Second Advisor

Prashant Shenoy

Third Advisor

Arun Venkataramani

Fourth Advisor

Dennis L. Goeckel

Subject Categories

OS and Networks


The past few years have seen a dramatic growth in wireless sensing systems, with millions of wirelessly connected sensors becoming first-class citizens of the Internet. The number of wireless sensing devices is expected to surpass 6.75 billion by 2017, more than the world's population as well as the combined market of smartphones, tablets, and PCs. However, its growth faces two pressing challenges: battery energy density and wireless radio power consumption. Battery energy density looms as a fundamental limiting factor due to slow improvements over the past several decades (3x over 22 years). Wireless radio power consumption is another key challenge because high-speed wireless communication is often far more expensive energy-wise than computation, storage and sensing. To make matters worse, wireless sensing devices are generating an increasing amount of data. These challenges raise a fundamental question --- how should we power and communicate with wireless sensing devices. More specifically, instead of using batteries, can we leverage other energy sources to reduce, if not eliminate, the dependence on batteries? Similarly, instead of optimizing existing wireless radios, can we fundamentally change how radios transmit wireless signals to achieve lower power consumption? A promising technique to address these questions is backscatter --- a primitive that enables RF energy harvesting and ultra-low-power wireless communication. Backscatter has the potential to reduce dependence on batteries because it can obtain energy by rectifying the wireless signals transmitted by a backscatter reader. Backscatter can also work by reflecting existing wireless signals (WiFi, BLE) when these are available nearby. Because signal reflection only consumes uWs of power, backscatter can enable ultra-low-power wireless communication. However, the use of backscatter for communicating with wireless sensing devices presents several challenges. First, decreasing RF power across distance limits the operational range of micro-powered backscatter devices. This raises the question of how to maintain a communication link with a backscatter device despite tiny amount of harvested power. Second, even though the backscatter RF front-end is extremely power-efficient, the computational and sensing overhead on backscatter sensors limit its ability to operate with a few micro-Watts of power. Such overhead is a negligible factor of overall power consumption for platforms where radio power consumption is high (e.g. WiFi or Bluetooth based devices). However, it becomes the bottleneck for backscatter based platforms. Third, backscatter readers are not currently deployed in existing indoor environments to provide a continuous carrier for carrying backscattered information. As a result, backscatter deployment is not yet widespread. This thesis addresses these challenges by making the following contributions. First, we design a network stack that enables continuous operation despite decreasing harvested power across distance by employing an OS abstraction --- task fragmentation. We show that such a network stack enables packet transfer even when the whole system is powered by a 3cmx3cm solar panel under natural indoor light condition. Second, we design a hardware architecture that minimizes the computational overhead of backscatter to enable over 1Mbps backscatter transmission while consuming less than 100uWs of power, a two order of magnitude improvement over the state-of-the-art. Finally, we design a system that can leverage both ambient WiFi and BLE signals for backscatter. Our empirical evaluation shows that we can backscatter 500bps data on top of a WiFi stream and 50kbps data on top of a Bluetooth stream when the backscatter device is 3m away from the commercial WiFi and Bluetooth receivers.