Research
My research deals with the large-scale design and automation of high-speed low-latency services / applications for clients by utilizing the 5G+ wireless network architectures. More specifically, I am interested in transactive models which allocate the network resources and services to the clients based on their individual preferences and data sharing policies. I focus on designing low complexity scheduling algorithms which ensure network wide coordination among the servicing nodes of the network operator and provide timely and feasibe recommendations to the clients. I have a keen interest in optimization techniques and deep reinforcement learning techniques. Some key issues addressed by my team include client behavior modeling, faster learning under uncertainty, scalability, privacy, data caching and maintenance, and network-wide service provisioning. Presently, my research scholars have taken up some important research problems in the vehicular edge networking.
Vehicular Edge Networks
Vehicular edge networks are a combination of electric vehicles (EVs) and 5G+ multi-access edge computing. The importance of VEC for transportation networks and power distribution networks is gaining importance in both research and commercial. This is evident in the recent trends in the publications under leading journals such as IEEE Trans. on Intelligent Transportation Systems, IEEE Trans. on Smart Grid, IEEE Trans. on Vehicular Technology and IEEE Trans. on Industrial Informatics. This is also evident in the investments done by major taxi service providers and The EVs are used for dispatching items or to provide ride hailing services. The 5G+ multi-access edge computing is expected to aid service discovery, assignment of EVs to clients/charging stations, tracking and repositioning of EVs besides providing seamless high-speed low-latency connectivity for the delivery of critical messages (e.g., road safety, congestion, navigation, object detection) and infotainment services. Owing to the uncertainties and the large-scale nature of the problem, a combination of distributed optimization, deep reinforcement learning, and network management are required. Issues such as sustainability, client/service guarantees, privacy, fairness, speed and convergence to near optimal solutions have to be addressed.
Research grants
"Solar Energy Investment and Energy Trading Problem ", PES University Internal funding, PESUIRF/ECE/2020/05 (Start date: 23-09-2020, End date: 24-09-2022)
Research scholars
Ms. Veena S works on "Proactive Caching and Computing Services in Vehicular Edge Networks"
Ms. Keerthi G Reddy works on "Charging, Assignment, Navigation and Repositioning of Electric Taxi Fleet"
Past doctoral students
Dr. Anuradha Kannan, RSNIT, worked on "Modeling and Analysis of Demand-Side Management"
Links
Google scholar: https://scholar.google.co.in/citations?user=_81IkLIAAAAJ&hl=en
PES University: https://staff.pes.edu/nm1243