AAKANKSHA CHOWDHERY

My name is pronounced as Aa-kan-sha. I was born in India and completed BTech in Electrical Engineering at IIT Delhi in 2007 finishing at the top of graduating class. I will be graduating from Stanford University with a PhD in Electrical Engineering supervised by Prof. John M. Cioffi.

I was recently awarded Paul Baran Marconi Young Scholar Award 2012. I was also recepient of Stanford's DARE Doctoral fellowship 2010-2012.

Email: achowdhery-at-stanford-DOT-edu
[LinkedIn]

Research Vision

My research vision is to design practical algorithms that enable quality user experience for internet-enabled devices ubiquitously. Toward this vision, I have applied "Dynamic Spectrum Management" techniques to wireline copper-access networks, wireless networks as well as home powerline communication networks. In simple terms, "Dynamic Spectrum Management" is a powerful framework that collects data of current communication network's state and configures them dynamically to achieve higher data-rates and quality of service. The design of such algorithms requires in-depth understanding of communications theory as well as simple machine learning algorithms and statistical modeling tools.

Research Contributions

One of the key contributions of my dissertation research is Dynamic Spectrum Management (DSM) techniques for next-generation copper-access networks that promise successful co-existence of multi-100 Mbps to Gbps copper-access networks with legacy networks in the next-generation. This work enables gradual incremental upgrade of digital subscriber lines (DSLs) to Gbps backhaul for wifi connections/small wireless cells everywhere. Thus, we can get order-of-magnitude improvement in data-rates and stability of DSLs allowing millions of internet-enabled devices to leverage the benefits on their DSL plus wifi connections. I have also led the standardization of this work in American and UK telecom networks thus taking it one step closer to real-world deployments.

Among other interesting research projects, I recently investigated the opportunity for dynamic spectrum access in spectrum bands between 30 MHz-6 GHz at Microsoft Research. Much of this spectrum is currently underutilized and can be opened up for "wifi" type secondary unlicensed use. This work developed a new goodness metric to compare the opportunity in different spectrum bands based on data-sets collected at Microsoft Spectrum Observatory.