Research Students

Graduate students engaged in research are at the core of a Research-One department. The people of an organization are its greatest assets; which, for a research group, are the graduate students.

I feel fortunate to work (and to have worked) with every one of my research students, listed below. Contact information is provided for current students - for past students, any information I have is listed, but they may be outdated and you can probably find more current information by searching the web.

(Note to all students, current and past: if I do not have up-to-date contact information for you, please send it to me. Got a better picture? Send it to me!)
  Prospective students, please feel free to get in touch with my current and former students; I certainly encourage you to do so if you are considering working with me.  

All completed thesis documents are on record and can be obtained from the Electronic Theses and Dissertations webpage of the D H Hill Library.

Last updated: August 6, 2014
Current Students
Aravindhan Dhanasekaran [Lab Notebook]
Aravind is working toward his MS thesis on the general area of SDN. Specifically, he is looking at the problem of integrating the type of payload matching that Rob and Kasyap worked on (see below) with packet scheduling within the datapath.
Portrait : Aravind Dhanasekaran

Portrait : Harshavardhan Joshi Harshavardhan Joshi [Lab Notebook]
Harsh took an M.S. degree from NCSU (ECE Dept.), and then spent some years working for Cisco. He is back in academic research full time now, with interests in SDN and security.

Robinson Udechukwu [Web] [Lab Notebook]
Rob received a B.S. from our department in 2010, and is back working toward the Ph.D. degree after a stint in the industry, we are glad to have him back! He is interested in wireless mesh network design and management, and has been looking at the software defined networking paradigm as a vehicle to realize such strategies, as well as assessing the pros and cons of doing so.
Portrait : Rob Udechukwu

Portrait : Magreth Mushi Magreth Mushi [Web] [Lab Notebook]
Magreth arrived at NCSU from Tanzania, where she has been working as a faculty member, and in Fall, 2012 started graduate studies at NCSU leading to a Ph.D. degree. She has been surveying the newly emerging area of Science of Security, and plans to pursue research in that area related to networking problems.

Shireesh Bhat [Email]
Shireesh came to NCSU to study for a Master's degree, but decided to stay for the long haul - good for him! He is advised by my colleague George Rouskas, and co-advised by me. He is working on semantic issues in future Internet architecture, specifically on defining the service definition and composition semantics for our ChoiceNet project.
Portrait : Shireesh Bhat
Graduated Students
Portrait : Kasyap Marmavula Ramachandra "Kasyap" Marmavula [Lab Notebook]
Graduated (M.S.) in Summer, 2014. Kasyap designed and demonstrated an extension to OpenFlow to allow specification of arbitrary byte matches in application layer header and payload in a packet, and extended the Linc software switch to realize this extension. Apart from his thesis work, he was an active contributor to my CentMesh project, and was instrumental in hosting the CentMesh Drones Challenge 2014. Kasyap joined Citrix after graduating.

Trisha Biswas [Web] [Lab Notebook]
Graduated (Ph.D.) in Spring, 2014. Trisha worked mainly in the area of multihop wireless networks, in the direction of robustness in the face of natural and malicious jamming, or other attacks. Her primary work has been in Petal Routing, which uses several different types of geo-diverse or diffuse multipaths to provide network-layer redundancy in the face of jamming. She also workd on using linear systems theory to characterizing the performance of such approaches. She went on to join Aruba Networks after graduation.
Portrait : Trisha Biswas

Portrait : Ahmet Can Babaoglu Ahmet "Can" Babaoglu [Web] [Lab Notebook]
Graduated (Ph.D.) in Spring, 2014. Can (he goes by his middle name - which is pronounced "Jon") has worked primarily toward future Internet architecture issues - he contributed to parts of the SILO project, and studied and designed future Internet verification architectures, especially in context of ChoiceNet, our future Internet project. He was also deeply involved in GENI, carrying the development side of my GENI project singlehandedly. In his spare time he published some research on optical network design as well, on converter placement. He went on to join Riverbed Technologies upon graduation.

Hui Wang [Email]
Graduated (Ph.D.) in Spring, 2013. Hui was advised by my colleague George Rouskas, and co-advised by me. She worked on efficient formulations and relaxations of Integer Linear Programs for traffic grooming in optical networks.
Portrait : Hui Wang

Portrait : Ashutosh K Grewal Ashutosh K Grewal
Graduated (M.S.) in Summer, 2012. He designed a protocol to attach sensor-rich devices to CentMesh. Ashu went on to join Juniper Networks in Sunnyvale, California.

Parth H Pathak
Graduated (Ph.D.) in Summer, 2012. Parth worked on wireless mesh network design - early on, he studied power control on its own, and later focused on the survivability and network continuity aspects of mesh networks. He published several papers and collaborated on a book with me, which was published in late 2012. Parth moved on to a post-doctoral fellowship at UC Davis.
Portrait : Parth Pathak

Portrait : Sankalp Nimbhorkar Sankalp U Nimbhorkar
Graduated (M.S.) Spring, 2012. Sankalp worked on implementing and verifying the performance of a back-pressure based spectrum width allocation in a mesh network. He joined VMWare after graduation.

Gaurang S Sinkar
Graduated (M.S.) Fall, 2011. Gaurang proposed a cooperative Sybil detection algorithm based on a puzzle based challenge-response approach, suitable for sensor networks. Gaurang joined Cisco after graduating.
Portrait : Gaurang Sinkar

Portrait : Gaurish Deuskar Gaurish Deuskar
Graduated (M.S.) Summer, 2010. Gaurish's thesis explored the possiblity of gaining some throughput efficiency by small packet aggregation in a multi-hop wireless network with back-pressure rate control without sacrificing fairness or congestion control too much. It turned out that the introduction of aggregation actually allowed improvement of the back-pressure mechanism. Last I heard Gaurish works for Qualcomm.

Umang S Patel
Graduated (M.S.) Fall, 2010. Umang's thesis work proposed a way to quantify the vulnerability of a multi-hop path in a wireless network to jamming, and proposed static and adaptive approaches to finding multiple redundant paths with low vulnerability. Umang joined Qualcomm after graduating.
Portrait : Umang Patel

Portrait : Anjing Wang Anjing Wang
Graduated (Ph.D.) Summer, 2010. Anjing's thesis work focused on future Internet architecture - he was the main contributor to SILO. He studied various design decisions for the SILO architectural vision, and realized prototype implementations. Anjing also studied the various technologies and techniques that fall under the term network virtualization. He was advised by me, and co-advised by George Rouskas. Anjing joined Ericsson after graduating.

Mohan L Iyer
Graduated (Ph.D.) Summer, 2010. Mohan was advised by George Rouskas, and co-advised by me. His thesis work was centered around providing bandwidth services for cloud providers on demand. Mohan studied network resource allocation to optimize such bandwidth services, SILO extensions to support high-bandwidth flows, grooming low-bandwidth flows, and heirarchical waveband switching. Mohan graduated to join Oracle.
Portrait : Mohan Iyer

Portrait : Suresh Subramanian Suresh K Subramanian
Graduated (Ph.D.) Summer, 2010. Suresh was a mid-career Ph.D. student - he was working for Lucent when he started; he moved to Cisco during his Ph.D. Suresh investigated performance issues of SIP. While the performance of media or voice over a data network has been studied extensively, the performance of the control plane is also crucial but generally had not been studied. Suresh used queueing models, simulation, and actual experimentation with real SIP software, to study and characterize the performance of the SIP control plane.

Prashant Jaikumar
Graduated (M.S.) Fall, 2008. Prashant studied an extension of the p-Cycle fault tolerance mechanism based on the concept of differential capacity - the observation that the amount of spare capacity that needs to be reserved on a link of a p-Cycle must be sufficient to protect traffic on other links that are covered by the p-Cycle, but not on itself.
Portrait : Prashant Jaikumar

Portrait : Deepa Seshadri Deepa Seshadri
Graduated (M.S.) Summer, 2008. Deepa studied green grooming - traffic grooming with the objective of minimizing the energy expenditure of the complete set of lightpaths.

Manoj Vellala
Graduated (M.S.) Spring, 2008. Manoj worked on the SILO project, and contributed concepts and prototype effort.
Portrait : Manoj Vellala

Portrait : Dheeraj Kandula Dheeraj Kandula
Graduated (M.S.) Spring, 2008. Dheeraj studied vehicular networks and Delay/Disruption tolerant networks, and proposed a TCP variant that was suitable for use in message-ferry networks.

Vineet K Sahijwani
Graduated (M.S.) Fall, 2006. Vineet's work was aimed at increasing the efficiency of middleware for large transaction based systems which transmit a lot of XML back and forth. Multiple on-the-wire XML representations exist; Vineet's work showed how the system can utilize different representations at different times to increase efficiency. Vineet graduated to join Cisco.
Portrait : Vineet Sahijwani

Portrait : Bhushan Bhatt Bhushan Bhatt
Graduated (M.S.) Fall, 2006. Bhushan followed up on earlier work by Sharat Visweswara to design a medium access approach for wireless sensor networks based on the idea of period inflation - nodes closer to the gateway have controlled longer periods for periodic sleep/wakeup.

Shu Huang
Graduated (Ph.D.) Summer, 2006. Shu dealt with many optimization problems that come up in the design of large networks, including articulating various realistic models of dynamic traffic, designing a slack allocation algorithm to maximize mean time to traffic blocking, and designing a wireless multihop network under a realistic physical (additive) model of interference. After graduation, Shu joined the Rennaissance Computing Institute (RENCI) of North Carolina.
Portrait : Shu Huang

Portrait : Pritesh Patwa Pritesh K Patwa
Graduated (M.S.) Summer, 2006. Pritesh studied vehicular networks, in particular, he extended an existing dense-fluid model for roadway traffic flow for the case where a vehicular network allowed drivers to be informed in advance of congestion ahead, and take alternate routes - thus changing the traffic flow itself. Pritesh joined Microsoft after graduating.

Divya Gupta
Graduated (M.S.) Summer, 2006. Divya worked on developing simple models for the throughput and delay of multihop wireless networks - her main contribution was the concept of loner links in such networks: links which interfered with so many other links that they effectively demanded a schedule slot each its own. The set of loner links dominates the schedule length in such cases.
Portrait : Divya Gupta

Portrait : Bensong Chen Bensong Chen
Graduated (Ph.D.) Summer, 2005. Bensong was advised by me and co-advised by George Rouskas. He worked on various aspects of heirarchical traffic grooming. obtained NP-completeness results for various grooming problems, constructed heuristic algorithsm, and also designed clustering algorithms for the heirarchical algorithms to operate on. Bensong joined Google Labs after graduation.

Sharat Visweswara
Graduated (M.S.) Summer, 2004. Sharat studied the general idea of sensors spending most of their time in sleeping mode to conserve energy and waking up periodically to perform and forward measurements, but under the assumption that the period of the various sensors are different according to a statistical distribution due to clock drift and other reasons. He showed that over time this resulted in the periods interacting in an unexpected way, and went on to quantify the benefit of lower sleep traded off with missed readings. Sharat joined Amazon after graduating.
Portrait : Sharat Visweswara

Portrait : Ruarob Tiyachate Ruarob Tiyachate
Graduated (M.S.) Spring, 2004. Ruarob studied the problem of joint grooming and survivability in optical networks, articulated the various levels at which the problems could be decoupled and their efficiency implications, and designed a heuristic algorithm. He returned to Thailand after graduation to join Celestia.

Apurva Goel
Graduated (M.S.) Spring, 2004. Apurva studied controlled delaying of sensor reading packets to ensure a local periodicity for sensor networks utilizing a sleep/wakeup cycle.
Portrait : Apurva Goel

Portrait : Ruhiyyih Mahalati Ruhiyyih Mahalati
Graduated (M.S.) Fall, 2003. Ruhiyyih attacked the problem of reconfiguration of the virtual topology of an optical network that also utilizes grooming. The objective functions for these two goals are difficult to compare, since one is a rate and the other is incurred only at distinct epochs of time. Ruhiyyih was able to formulate a joint problem, which she then went on to linearize, and also designed heuristic approaches. Ruhiyyih joined Microsoft after graduation.

Koundinya Srinivasarao
Graduated (M.S.) Summer, 2003. Koundinya worked on traffic grooming in ring networks, and designed two heuristic approaches based on decomposition; one based on grouping the nodes, and the other on partitioning the traffic matrix. He showed that the second approach is more general but more difficult, and developed a test for traffic matrices based on clustering to determine when the first approach would work equally well. Last I heard, Koundinya was working at OPNET.
Portrait : Koundinya Srinivasarao

Portrait : Prashant Iyer Prashant Iyer
Graduated (M.S.) Spring, 2003. Prashant was advised by Carla Savage (with whom he went on to work for his Ph.D.), and co-advised by me. Prashant studied the problem of traffic grooming in paths, and showed that several flavors of the problem, distinguished by their cost functions, were not equivalent, but were all NP-complete. He also articulated the NP-completeness cliff in several directions, showing that the problem was tractable if the traffic distribution was uniform, or there were only two wavelengths, or all traffic was bound for a single egress node.

Shu Huang
Graduated (M.S.) Spring, 2003. Shu worked on traffic grooming in the path topology, and showed that even in this very elemental topology, some versions of the problem were NP-complete, including even cases where the virtual topology had already been specified. He also designed heuristic algorithms for the problem. After graduation, Shu returned to China to work for a period, and later joined NCSU again as a graduate student to work for his Ph.D. degree with me.
Portrait : Shu Huang