Optimal Relay Services for MSNs
In this chapter, the optimization of relay service for MSNs is studied. In MSNs, each user has his own virtual currency and can earn currency as a relay for other users. A bundle carrier selects relay users based on his current resource usage. A bargain game is employed to model the transaction pricing between the bundle carrier and the relay user, which leads to a subgame perfect Nash equilibrium as the agreement of two players to maximize their benefits. The simulation shows that the proposed scheme can efficiently improve both delivery ratio and delay performance in MSNs with an optimal relay service.
Relay Services in MSNs
In MSNs [1-4], users can create and share their content with each other by using mobile devices equipped with short-range wireless interfaces via peer-to-peer opportunistic links [5-8]. In MSNs, the transmission path between the source and the destination is unstable or even unavailable sometimes . Moreover, since the MSNs have unique features with different types of social ties among mobile nodes, e.g., friends, relatives, etc., the transmission path is even harder to set up than the conventional networks without social ties. Therefore, a store-carry-forward fashion is used to deliver bundles to the destination in MSNs, where this delivery fashion requires mobile nodes to provide relay services.
Although many studies have been carried out to design relay schemes for wireless networks [10-12], most of the relay services assume that the transmission path between the source and the destination always exists and is stable, where these studies cannot be directly applied to MSNs. In addition, the social ties among mobile nodes bring new challenges to the study of bundle delivery in MSNs. Therefore, it is still a new and open problem to design social-aware relay services for bundle delivery in MSNs.
In this chapter, as shown in Fig. 4.1, a novel relay scheme for delivering bundles in MSNs is proposed. First, each node has its own virtual currency and can earn
© Springer International Publishing AG 2016 55
Z. Su et al., Modeling and Optimization for Mobile Social Networks,
Fig. 4.1 Schematic of an incentive scheme in the MSN
currency as a relay for other nodes. When a node refuses to forward the bundle of others, it will not get paid. Therefore, this node will lose a chance to earn currency to afford the relay service from other nodes in the future, with the result that the proposed scheme can efficiently prevent nodes from being selfish. Next, a bundle carrier selects relay nodes based on its current status of limited resources. Specifically, if the status of the carrier is loose, it will select one of its friends to be a relay with a low agreement price. Otherwise, if the status of the carrier is tense, it will select any node it encounters, i.e., even a nonfriend node, with a high agreement price. Then, a bargain game is employed to model the transaction pricing between the bundle carrier and the relay node, which leads to a subgame perfect Nash equilibrium as the agreement of two players to maximize their benefits. In addition, with simulation experiments, it proves that the proposed scheme is efficient to improve both delivery ratio and delay.