Thèse soutenue

Etude de problèmes de routage dans les réseaux de capteurs sans fil et la logistique

FR  |  
EN
Auteur / Autrice : Wenhua Zhao
Direction : Haoxun Chen
Type : Thèse de doctorat
Discipline(s) : Optimisation et Sûreté des Systèmes
Date : Soutenance en 2012
Etablissement(s) : Troyes
Ecole(s) doctorale(s) : Ecole doctorale Sciences pour l'Ingénieur (Troyes, Aube)

Résumé

FR  |  
EN

In this thesis, several routing problems in wireless sensor networks or logistics are studied. Firstly, the routing of mobile sensors for event capture with different event occurrence frequencies at different points of interest is studied. This problem is formulated as a cyclic multiple traveling salesman problems (mTSP) with inter-visit time constraints of nodes and solved by a column generation approach. Secondly, the routing of mobile sensors or actors for data collection or event handling in a stochastic environment where data are generated or events happen randomly is studied. The problem is formulated as a probabilistic multiple TSP problem and solved by a variable neighborhood search (VNS) method. We use two methods to evaluate the change of the expected cost in the local search: analytic method and simulation based estimation method and derive some cost change formulas for the two methods. Numerical experiments demonstrate that the VNS with estimation-based cost evaluation is more efficient than the VNS with analytic cost evaluation. Finally, the routing of mobile actors for event handling in a stochastic and safety-critical environment is studied, where each event must be handled before its deadline or the handling of the event later than its deadline incurs a penalty cost. This problem is formulated as a probabilistic multiple TSP with hard deadlines or soft deadlines and solved by a VNS with estimation-based cost change evaluation. All methods proposed in this thesis are evaluated by numerical experiments