Seemron Neupane

My research interest lies in building wireless systems that remain robust when real world conditions violate clean modeling assumptions. Specifically, I am interested in how handsets can use low fidelity, cheap to observe signals, such as coarse CQI reports, HARQ retransmission patterns, thermal readings, and modem state transitions, to adapt physical layer and scheduling parameters under non stationary channels, without relying on expensive measurements or ideal feedback loops.

Topics I want to explore include:

• Adaptive link and PHY configuration using indirect proxy signals for regime shift detection under mobility and interference.
• Cross layer failure modeling understanding how PHY and upper layer interactions produce long tail stalls and throughput collapse that average case metrics hide.
• Evaluation methodology that surfaces tail behavior and reproducible edge case failures, rather than only reporting means.

My applied background in RF drive testing, device validation, and embedded systems gives me a measurement first perspective. I have seen the gap between what models assume and what devices actually do in the field, and I want to work on methods that close it in a principled, reproducible way.