A Bayesian Model for Mobility Prediction in Wireless Sensor Networks

Abstract : Wireless sensor networks (WSN) have specific features such as low transmission range, stringent energy consumption constraints, limited memory and processing power. For this reason, tailored protocols have been proposed, optimizing network protocols with respect to various assumptions: one commonly exploited property of WSN is the stability of the topology due to permanent installation of sensor nodes. However, in some applications, some of the wireless sensor nodes might be mobile, for instance when they are associated with users. In that case, specific extensions of WSN protocols need to be designed. Then a first step is the characterization of nodes' mobility. This is the focus of this article: we propose a general method of mobility estimation for wireless sensor networks. Namely, using a Bayesian framework, we derive a mobility prediction model to estimate the node velocity (starting from no knowledge) from observed events. In this article, we focus on events represented by the most minimal information: mere observation of link duration with neighbors. Simulations illustrate that, even with such limited information, the mobility can be well inferred, and results show the good performance of the method.
Document type :
Conference papers
Complete list of metadatas

Cited literature [27 references]  Display  Hide  Download

https://hal.inria.fr/hal-01405277
Contributor : Cédric Adjih <>
Submitted on : Tuesday, November 29, 2016 - 4:40:00 PM
Last modification on : Thursday, February 7, 2019 - 5:34:16 PM
Long-term archiving on : Monday, March 27, 2017 - 8:35:06 AM

File

PEMWN_IK.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01405277, version 1

Collections

Citation

Fatma Somaa, Cédric Adjih, Inès Korbi, Leila Saidane. A Bayesian Model for Mobility Prediction in Wireless Sensor Networks. 5th IFIP International Conference on Performance Evaluation and Modeling in Wired and Wireless Networks (PEMWN 2016), Nov 2016, Paris, France. ⟨hal-01405277⟩

Share

Metrics

Record views

235

Files downloads

468