Computing Average Value in Ad Hoc Networks

  • Mirosław Kutyłowski
  • Daniel Letkiewicz
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2747)


We consider a single-hop sensor network with n=Θ(N) stations using R independent communication channels. Communication between the stations can fail at random or be scrambled by an adversary so that it cannot be distinguished from a random noise.

Assume that each station S i holds an integer value T i . The problem that we consider is to replace the values T i by their average (rounded to integer values). A typical situation is that we have a local sensor network that needs to make a decision based on the values read by sensors by computing the average value or some kind of voting.

We design a protocol that solves this problem in O(N/R ·logN) steps. The protocol is robust: a constant random fraction of messages can be lost (by communication channel failure, by action of an adversary or by synchronization problems). Also a constant fraction of stations may go down (or be destroyed by an adversary) without serious consequences for the rest.

The algorithm is well suited for dynamic systems, for which the values T i may change and the protocol once started works forever.


mobile computing radio network sensor network 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Chlebus, B.S.: Randomized communication in radio networks. In: Pardalos, P.M., Rajasekaran, S., Reif, J.H., Rolim, J.D.P. (eds.) Handbook on Randomized Computing, vol. I, pp. 401–456. Kluwer Academic Publishers, Dordrecht (2001)Google Scholar
  2. 2.
    Czumaj, A., Kanarek, P., Kutyłowski, M., Loryś, K.: Distributed stochastic processes for generating random permutations. In: ACM-SIAM SODA 1999, pp. 271–280.Google Scholar
  3. 3.
    Czumaj, A., Kutyłowski, M.: Generating random permutations and delayed path coupling method for mixing time of Markov chains. Random Structures and Algorithms 17, 238–259 (2000)zbMATHCrossRefMathSciNetGoogle Scholar
  4. 4.
    Gosh, B., Muthukrishnan, S.: Dynamic Load Balancing in Parallel and Distributed Networks by Random Matchings. JCSS 53(3), 357–370 (1996)Google Scholar
  5. 5.
    Jurdziński, T., Kutyłowski, M., Zatopiański, J.: Energy-Efficient Size Approximation for Radio Networks with no Collision Detection. In: Ibarra, O.H., Zhang, L. (eds.) COCOON 2002. LNCS, vol. 2387, pp. 279–289. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  6. 6.
    Lamport, L., Shostak, R., Pease, M.: The Byzantine Generals Problem. ACM TOPLAS 4, 382–401 (1982)zbMATHCrossRefGoogle Scholar
  7. 7.
    Stojmenovič, I.: Handbook of Wireless Networks and Mobile Computing. Wiley, Chichester (2002)CrossRefGoogle Scholar
  8. 8.
    Upfal, E.: Design and Analysis of Dynamic Processes: A Stochastic Approach. In: Bilardi, G., Pietracaprina, A., Italiano, G.F., Pucci, G. (eds.) ESA 1998. LNCS, vol. 1461, pp. 26–34. Springer, Heidelberg (1998)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Mirosław Kutyłowski
    • 1
  • Daniel Letkiewicz
    • 2
  1. 1.Inst. of MathematicsWrocław University of Technology 
  2. 2.Inst. of Engineering CyberneticsWrocław University of Technology 

Personalised recommendations