Publications and Projects

One of the biggest challenges to wider adoption of environmental sensing networks is the lack of ubiquitous power infrastructure. You can’t just plug electronics into the dirt… but what if you could? Our work, “Towards Deep Learning for Predicting Microbial Fuel Cell Energy Output”, which was accepted to ACM COMPASS 2024, progresses this vision by creating a deep learning model that takes in environmental data (like soil moisture, temperature, and electrical conductivity), and outputs a prediction of how much energy the soil battery will produce for a future time horizon.

Our work “Soil-Powered Computing: The Engineer’s Guide to Practical Soil Microbial Fuel Cell Design” was published in IMWUT (Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies).

Our work on carbon-aware load balancing was presented at the second annual ACM HotCarbon workshop, and will appear in the ACM SIGEnergy Energy Informatics Review journal.

Our work on sustainable Information and Communications Technology was presented at the first annual HotCarbon workshop co-located with OSDI, and also published in the ACM SIGEnergy Energy Informatics Review journal.

Our work on “Hardware to enable large-scale deployment of soil microbial fuel cells” was presented at the ENSsys 2022 workshop, which is part of the SenSys conference.

Our work on “Early Characterization of Soil Microbial Fuel Cells” was accepted at IEEE ISCAS 2022.

Our work on low-cost, high accuracy in-situ soil moisture sensing has been accepted for publication at ACM COMPASS 2021. The paper is titled Low-cost In-ground Soil Moisture Sensing with Radar Backscatter Tags. ACM COMPASS aims to be the forum for the presentation and publication of original research from a broad array of disciplines, including computer and information sciences, social sciences, environmental sciences and engineering, that support the growth of sustainable societies worldwide.

Our short paper titled Farming Electrons: Galvanic vs. Microbial Energy in Soil Batteries has been accepted to IEEE Sensors Letters.

Our short paper titled Time-of-Flight Soil Moisture Estimation Using RF Backscatter Tags has been accepted to IGARSS 2020, the IEEE International Geoscience and Remote Sensing Symposium. It was virtually presented in October 2020.

I’ll be presenting a live demo of our soil moisture sensing work at IPSN ‘20 in Sydney, Australia during April 21-24. Here’s a video for those who can’t make it to Australia:

Our paper, “RF Soil Moisture Sensing via Radar Backscatter Tags” is under review and a pre-preint is available on the arXiv.

Our poster, “Wireless Soil Moisture Sensing”, won an award for best poster at the SenSys ‘19 N2Women Workshop.

Our paper has been accepted for publication at IPSN ‘19 (Information Processing in Sensor Networks): BackCam, Wireless Computer Vision Using Commodity Devices by Colleen Josephson, Lei Yang, Pengyu Zhang and Sachin Katti.

In Summer 2018 I was a research intern at Microsoft Research in Redmond, WA. My mentor was Ranveer Chandra and I worked on next-generation soil sensing in the Farmbeats project. We all got to demo out work to Bill Gates himself, and in October it was featured on his blog. There was even a video made where I made a few brief appearances showcasing our backscatter tag and digging holes for experiments.

At the CoNEXT 2017 conference in Seoul, South Korea, we introduced the design and implementation of FreeRider, the first system to enable backscatter communication with multiple commodity radios, such as 802.11g/n WiFi, ZigBee, and Bluetooth, while these radios are simultaneously used for productive data communication. Furthermore, we are, to our knowledge, the first to implement and evaluate a multi-tag system.

In 2016 I circumnavigated the globe, visiting friends and family across five continents. I used a digital recorder to make field recordings of the various interesting things I heard during my travels. This is a playlist of the highlights.

MIT’s EECS department has a Masters of Engineering (MEng) program where qualifying undergraduates can apply to stay for an additional year and complete a master’s degree. The program requires both coursework and a research thesis. My research advisor was Professor Muriel Medard, and my topic was on using network coding for anonymous network communications.

This was my term project for 6.829, MIT’s graduate computer networking class. We compared the performance of two wireless bit-rate selection algorithms and made improvements to one of them. Bit-rate selection is the process of choosing which bit rate to send with over the wireless link. Unlike wired networks, it is very common for wireless networks to send at lower rate that the maximum the channel supports. This is because wireless channel conditions are rarely ideal due to factors such as fading and interference. The goal is to select a bit rate that achieves the highest throughput, and to keep that rate updated as channel conditions change.