Abstract : The diversity of radio access technologies (e.g., GPRS, UMTS, HSDPA,Wi-Fi, WiMAX, LTE...), their complementary in terms of coverage area, technical characteristics (e.g., bandwidth, QoS) and commercial opportunities for the operators lead to the development of mobile terminals integrating multiple radio interfaces. The ability of mobile terminals to support various interfaces provides many interesting benefits, such as permanent and ubiquitous access, reliability, load sharing/load balancing, bandwidth aggregation, and muti-criteria interface selection. Mobile terminals with several radio interfaces have the possibility to choose the "best" interface according to several parameters such as application characteristics, user preferences, network characteristics, operator policies, tariff constraints, etc. It becomes also possible to associate the applications to the available network interfaces basing mainly on application requirements. In the thesis, we tackle the interface selection issue where a mobile terminal equipped with several interfaces has to select at any time the best interface or the best access technology according to multiple criteria. We particularly focus on the decision schemes and investigate the MADM methods The fundamental objective of the MADM methods is to determine among a finite set of alternatives the optimal one. MADM includes many methods such as Simple Additive Weighting (SAW), Weighting Product (WP), and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The first aim of the thesis is to study and analyze the Multiple Attribute Decision Making (MADM) methods for interface selection issue. A first contribution is to propose a simulation based study which highlights limitations of the methods in this context. TOPSIS suffers from the "ranking abnormality" problem. The ranking abnormality problem occurs when a low ranking alternative is removed from the candidate's list (e.g., one network is disconnected), the order of higher ranking alternatives will change abnormally. As a second contribution, we propose Distance to the ideal Alternative (DiA) algorithm which helps terminal to select dynamically the best interface by providing a ranking order between the interfaces. We show that DiA does not suffer from the ranking abnormality which is the shortcoming of the TOPSIS method. Simulation results validate the DiA algorithm. The third contribution tackles the flow/interface association issue (per-flow interface selection) where a mobile terminal equipped with several interfaces has to associate a specific application to the suitable interface. We first propose an interface utility function. This utility function allows identifying the interface which considers the satisfaction of the application requirements and economizes the energy consumption of the mobile terminal. We then propose an utility-based flow/interface association scheme that allows to associate the application to the appropriate interface. Network side attributes such as access delay and cost of using the network are also considered in the scheme. The Distance to the ideal Alternative (DiA) algorithm is used to rank the interfaces based on the interface utility values and the network side attributes. Simulation results are presented to validate the interface utility based scheme. Additionally, we propose a multiple flow/interface association scheme. In this case, a terminal running several applications tries to associate simultaneously each application flow to the suitable network interface while maximizing global utility. The multiple flow/interface association is an optimization problem. Particularly, it is related to stochastic heuristic optimization problems which are mainly based on search techniques of which solutions and search order depend on random variables. As the first step, we studies and realizes a simulation comparison of stochastic heuristic optimization methods such as local search, Tabu search, and simulation annealing algorithms. We then propose an oriented diversification technique of the Tabu search as an improvement. This allows the Tabu search to avoid being re-entrapped in the local optimization several times and to increase the performance of Tabu search in our context. Simulation results demonstrate that the modified Tabu search outperforms other stochastic heuristic algorithms in our context. We then head to a network centric approach while taking into account the flow/interface association. We consider a system of multi-interface mobile terminals with the ability of application/interface associations. As multiple terminals compete for common network resources, the system is modeled as a strategic game. Our objective is to find Nash equilibrium strategies of the game. We let the game evolve according to the so-called Replicator dynamic and then observe whether the system converges and whether the stationary points are Nash equilibria. We show that the Replicator dynamic is Positively Correlated and the system is a potential game. Our system converges to stationary points which include all Nash equilibria. Furthermore, the stationary points are proven to be efficient as they are solutions of the optimization problem of the total utility. An interesting edge is that our analytical results are valid for a general utility function which depends on the whole system state of connections. To validate our model and to demonstrate that the system converges to Nash equilibria, we implement two simulation scenarios using Nash learning algorithm with a specific bandwidth allocation scheme as well as a utility function that takes into account the application satisfaction level and the energy consumption.