Game Theory and Wireless Networks
Link: https://www.intechopen.com/chapters/62516
Game theory has been largely applied in people’s daily life and some of its application is really important. This article introduces the application in the computer science field where engineers try to increase and maintain the performance of wireless networks. We cannot deny that in recent years, people have become more and more addicted to the Internet and network performance has become more and more important. Imagine a time, when you are rush to meet the deadline in your apartment, unfortunately, the wifi becomes slow since there are too many residents refreshing the internet at the same time. Does that sound like a recipe for disaster?
Let’s first introduce the existing problem of network performance and why game theory is applied. A wireless network is similar to a system consisting of a number of nodes that communicate with each other via the wireless data connection, i.e. a media. When a source node transmits the data to the destination node, it requires other nodes’ help, but not all nodes are willing to relay the data to others. In this case, the performance of the network largely declines.
The game theory is applied to increase performance if a data collision happens, in which two nodes of the network want to access the media at the same time. This collision is quite expensive since it will result in packet loss, the decline of the network capacity, and even the termination of the network life span.
An optimal pricing technique is then deployed among a number of nodes to achieve various resource allocation policies. With a compensation price, the unwilling relay nodes will then choose to use their resources by making some payoff, and thus be willing to forward the messages to the other nodes.
The model is aiming to maximize the utility of each node. Each one has its own payoff for sending or not sending messages to the allocation, then a Nash equilibrium can be found. A dynamic differential game and target cost function can also be successfully derived for the Nash equilibrium between all nodes. Like what we discussed in the class about game theory, two players are extended to multiple nodes and reaction functions are derived depending on the choice of the other players, nodes here. Super fascinating!