Generalized nash bargaining solution for bandwidth allocation
Computer Networks: The International Journal of Computer and Telecommunications Networking
REPLEX: dynamic traffic engineering based on wardrop routing policies
CoNEXT '06 Proceedings of the 2006 ACM CoNEXT conference
Selection of efficient pure strategies in allocation games
GameNets'09 Proceedings of the First ICST international conference on Game Theory for Networks
| Hi-index | 0.00 |
Most of recent mobile equipment now supports different network technologies (WiFi, WiMax, LTE, Bluetooth and such like). Meanwhile, network operators offer services through these different technologies. The superposition of the different technologies (using different frequency band) increases the potential throughput of the system and hence global performance. Furthermore, new norms enable mobiles to dynamically switch between these different technologies while maintaining communications. This is known as vertical handover and the consequent network is called heterogeneous. This dynamic switching spares the need of predicting the users' behaviors to obtain efficient association schemes, opening the way for real-time dynamic association algorithms. In game theory, the recent evolutionary framework explicitly takes into account the dynamic nature by which individuals learn equilibria. The basic idea in the evolutionary framework is that individuals of populations naturally mimics the behavior of well-fitted elements. Based on the underlying mimics mechanism and of the initial conditions, the overall behavior of a population follows a trajectory of a differential equation. In this work, we compare two classes of algorithms that approximate several dynamics defined in evolutionary game theory when applied to the user-network association problem in heterogeneous networks. In particular, we study their performance in terms of quality of the obtained equilibria and convergence speed. We further study their complexity and robustness properties with respect to erroneous measurements.