Architecture for Internet of Things: Optimizing Network Performance Over Limited Data
In recent years there has been an explosive growth of the IoT network with an estimated 50 billion devices expected to join the network by 2020. A new architecture needs to be developed that can scale with the network. These scalability challenges open up numerous avenues for network research. Some examples include (1) developing new techniques for network management and control to handle the needs of large-scale dynamic networks (2) researching how to smartly deploy sensors to secure cyber-physical systems and (3) combining big data analytics with cognitive techniques to learn network behavior and operate networks. The expected focus of this project will be to see how sparse sampling of network parameters may be used to infer network states and optimize performance.

I am excited by this SuperUROP as I will have the opportunity to gain a more in-depth understanding of networking as well as learn about new advances in Internet of Things. Previously I worked on data transfer protocols for the CLARITY project an air quality network system developed within the MIT CEE department. I have also interned at Cisco Systems where I received exposure to concepts in software-defined networking.