Albatross

Project on opportunistic communication systems (part of Microsoft Research Cambridge Humanet project)

Project goal is to investigate feasibility, performance, and design of opportunistic communication systems. With opportunistic communication systems information is disseminated at contact instances of mobile devices such as those carried by humans or mounted on vehicles such as car or buses and infrastructure devices if there are any.

Why albatross? -- Empirical evidence suggests that animal movements such as that of albatross follow power law displacements such as that of Levy random walk see here and here.   

Publications

1. On Mobile User Behaviour Patterns, M. Vojnovic, Int'l Zurich Seminar on Communications (IZS 2008), March 12-14, 2008 (invited paper).
2. Power Law and Exponential Decay of Inter Contact Times between Mobile Devices, T. Karagiannis, J.-Y. Le Boudec, M. Vojnovic, ACM Mobicom 2007 (version with proofs MSR-TR-2007, March 2007).
   
   In contrast to earlier work, using a diverse set of measured mobility traces of human-carried devices, we find as an invariant property that there is a characteristic time, order of half a day, up to which the distribution of inter-contact time is well approximated by a power law and beyond it decays exponentially. We also find that contrary to earlier claims, already simple mobility models support the dichotomy in the distribution of inter-contact time.

Talks

1. ACM Mobicom 2007 talk slides.
2. Power Law and Exponential Decay---Mobile Devices, invited lecture, HyNet Colloquium, March 15, 2007 talk slides.
3. An early talk delivered to post-Sigcomm 2006 workshop, report here.
4. Related: ACM Mobicom 2006 tutorial on random trip models here.

Collaborators

• EPFL, Switzerland
• Thanks to Eric Horvitz and John Krumm (MSR Redmond) for providing us with vehicular data

Related projects

• Delay Tolerant Networking
• Haggle
• MSMLS
• UMass DieselNet

Last update, March 20, 2007