Mansi Sood

Carnegie Mellon University

Position: PhD Candidate
Rising Stars year of participation: 2021
Bio

Mansi Sood is a Ph.D. student in Electrical and Computer Engineering at Carnegie Mellon University. Before this she completed her B.Tech. in Electrical Engineering and M.Tech. in Communications and Signal Processing from the Indian Institute of Technology (IIT) Bombay India. Her research focuses on two main themes: analyzing spreading processes over real-world networks and designing securely connected ad-hoc networks. She is a recipient of the best paper award at the IEEE International Conference on Communications. Her research at Carnegie Mellon University has been supported by several prestigious fellowships including the Philip and Marsha Dowd Fellowship CyLab Presidential Fellowship Knight Fellowship and David H. Barakat and LaVerne Owen-Barakat CIT Dean’s Fellowship. In addition to her passion for research and teaching she is an avid visual artist and engages in community programs for making science and art more accessible to diverse audiences.

Heterogeneous models for designing resilient topologies and controlling spreading phenomena in real-world networks

Heterogeneous models for designing resilient topologies and controlling spreading phenomena in real-world networks
My research studies real-world networks with emphasis on two main themes : i) Designing resilient ad-hoc networks A key challenge in designing ad-hoc networks such as sensor networks and next-generation internet architectures is to establish a securely connected network of devices while minimizing operational costs. Random K-out graphs are receiving attention as a model to construct sparse yet well-connected topologies in a fully distributed fashion in sensor networks and cryptocurrency networks. In these applications it is important to model heterogeneity in resource access among the participating agents; and design topologies that are securely connected and resilient to the failure or capture of participating agents. Our recent work analyzes connectivity in random K- out graphs and their heterogeneous variants under random node failures and targeted attacks on nodes or links. We show that random K-out graphs yield a large giant component that persists even in the event of node failures. ii) Modeling spreading phenomena in real-world networks The second focus of my research is to develop epidemiological models that account for variability in transmission risks associated with the emergence of new variants and changes in the contact network structure under policy interventions. Our recent work proposes a modeling framework that accounts for the multi-layer structure typical of human contact networks. Specifically we assume that the risk of transmission depends not only on the type of strain carried by an infective individual but also on the nature of links used to infect their neighbors. For characterizing epidemic outbreaks caused by mutating pathogens over multi-layer contact networks we derive the probability of emergence of an epidemic the expected fraction of individuals infected with each strain and the phase transition point at which an epidemic emerges.