Biography
I am Somnath Mitra a PhD student at the WISe MANet Labs Department of Computer Science, University of Memphis under Professor Santosh Kumar
My primary research interest is in the field of Wireless Sensor Networks. Wireless Sensor networks is the realization of a vision called Smart Dust, where very small computers will be sensing the environemnt, processing information by collaborating with each other and making sense out of it all. This vision will help improve the quality of human life in several ways. Over the years (starting 1999) several researchers have successfully demonstrated the feasibility and usefullness of such systems.
At WISe MANet, our goal is to realize this vision with a humanist perspective. Everyday at our lab, we strive to answer one fundamental question: "How will our research make this world a better place?". Keeping this in mind, I am involved mainly in two projects. The first being AutoWitness, a burglar tracking project where our goal is to eliminate burglary from society. We propose to achieve this mamoth task, by embedding people's assets(such as tv, furniture, etc) with very low cost sensors tags. These tags will be ultra quiet and ultra power efficient to last on a single coin cell battery for ten years. These tags will detect theft and once theft is detected will raise alarms by sending out wireless messages to a an anchor node infrastrcutre. These tags will not give any clues to the burglar that their crime is being witnessed. The anchor node infrastructure is a low cost, sparse citywide deployment of wireless stations that will help the police to track and localize the burglar, convict the burglar and retrieve the stolen property.
The other project that I am involved in is called AutoSense. The goal of this project is to enable social, phsycological scientists to understand the relationship of psychosocial stress and the environment. We propose to develop a system called AutoSense which has a suite of sensors, and a gateway (a cellphone) that collects data from these sensors. The innovation in this project is to figure out a measure of stress which can be very subjective (varies from person to person and from event to event), based on all multiple modalities of sensing that AutoSense provides. Some challenges in this project are feature extraction, stress index computation, energy efficiency as we cannot stream all the data from all the sensors. This suite will also have a new Alcohol sensor specially designed for AutoSense that uses ISF to measure alocohol concentration in the body in a NON INVASIVE manner. This will enable social scientists to study group behaviour such as " Do young people drink more on Friday Nights ?",etc.(Is that a YES always !!!).