Home
Research
Publications
Talks
 Research Interests and Musings
Ph.D. Dissertation
Other Misc. Research
Publications
Simulating Heterogeneous Multi-Channel Wireless Networks in ns-2

My research interests and rambling thoughts on research

My research interests lie in the areas of wireless networking and distributed computing. I am interested in both theoretical and protocol design aspects pertaining to enabling communication and/or distributed computation in multi-hop wireless networks.

I like to believe in the Utopian ideal of establishing rigorous theoretical foundations, and thereafter translating them to practically feasible designs. Alas, like many a (much-wiser) person before me, I have found that this is indeed quite non-trivial, and find it fun to distort T.S. Eliot's words (see below) in this regard. However, I am still very much engaged in the effort, and hold on to the hope that eventually I might succeed.
Between the idea, And the reality, Between the motion, And the act, Falls the Shadow
Between the conception, And the creation, Between the emotion, And the response, Falls the Shadow
-- T.S. Eliot (The Hollow Men)

The researcher's viewpoint: Between the theory, And the practice, Falls the Shadow

Finally, why do I like doing research? I guess there is something quite appealing about trying to find your way around, amidst things you don't quite understand yet, but hope to eventually figure out! As a child I read some lines by Max Planck on this issue that I found quite profound...I guess they make for an apt quote here:
Science...means unresting endeavour and continually progressing development towards an aim which the poetic intuition may apprehend, but which the intellect can never fully grasp.-- Max Planck

Ph.D. Dissertation (graduated October 2008)

Performance of wireless networks subject to constraints and failures

  • Multi-channel wireless networks with interface and channel heterogeneity
    This work is motivated by scenarios where multiple spectral bands/channels may be available for use, but there may be heterogeneity in interface capabilities as well as channel characteristics. Interface heterogeneity implies that all radio-interfaces may not be capable of tuning to all available channels, i.e., they may be subject to switching constraints. Channel heterogeneity implies that the achievable link-rates may be different for different channels. In order to develop an initial theoretical understanding of interface heterogeneity, we proposed some constraint models, and studied the asymptotic capacity scaling behavior of a randomly deployed network under these models. The capacity constructions yielded interesting insights into the coupled nature of channel/interface selection in networks where radio-interfaces do not all operate on the same set of channels. We have also considered the scheduling implications of channel heterogeneity in networks of realistic scale, and devised some scheduling algorithms that can provide acceptable performance using limited information-exchange. Building on theoretical insights, we designed a channel and interface management protocol for multi-channel multi-radio wireless networks with heterogeneous interfaces and channels. The objective was to handle devices with different type/number of interfaces, and adapt to traffic. The design drew inspiration from some of our theoretical work, as well as from other existing theoretical work in the literature, e.g., the Dynamic Backpressure Scheduler of Tassiulas and Ephremides. We have evaluated the designed protocol via ns-2 simulation. Details of the protocol implementation are available in Chapter 6 of my Ph.D. Thesis. I am currently working on some extensions to this work.
  • Reliable communication in wireless networks in the face of Byzantine/Crash-Stop Faults
    In this work we have studied issues pertaining to achievability of reliable communication in multi-hop wireless networks, where nodes may exhibit Byzantine failure. In the absence of end-to-end encryption, the presence of Byzantine nodes complicates the task of reliable multi-hop data communication, as the Byzantine nodes may relay/report wrong message values. We have established conditions for achievability of reliable broadcast under various network and fault placement models, subject to some idealizations. We have also examined the issue of achieving reliable local broadcast in a realistic wireless environment.
The average Ph.D. thesis is nothing but a transference of bones from one graveyard to another.
--J. Frank Dobie (A Texan in England)
(I would really like to believe that, in case of my thesis, the graveyards in question were sufficiently far apart to make the transference a slightly non-trivial matter, but...[:)])

Other (non-dissertation) Research

  • Real-time scheduling in multi-rate wireless networks
  • Reprogramming multi-functionality wireless sensor networks (I worked on this issue while an intern at Motorola Labs, Schaumburg in Summer 2004).
  • Rate and power control for energy-efficiency in many-to-one networks
Selected Publications
Conference Papers:

Short Conference Papers/Workshop Papers/Poster Abstracts:

Posters:

  • V. Bhandari and N. H. Vaidya, Heterogeneous Multi-Channel Wireless Networks: Routing and Link Layer Protocols, Poster at ACM Mobicom 2007
  • V. Bhandari and N.H. Vaidya, Implementing a Reliable Local Broadcast Primitive in Wireless Ad hoc Networks, Poster at ACM PODC 2006
  • V. Bhandari and N.H. Vaidya, A Rate Control Mechanism for Energy-Efficient Many-to-One Communication (Mobicom 2004 poster)

Recent Technical Reports: