Desktop version

Home arrow Engineering arrow Modeling and Optimization for Mobile Social Networks

Modeling of Selfishness-Aware Incentive for MSNs

In this chapter, we discuss the modeling of selfishness and propose an incentive mechanism in MSNs. We consider two types of selfish behaviors: weak selfishness and extreme selfishness. The information dissemination with weak and extreme selfishness are studied, respectively, through the ordinary differential equations (ODEs). Then, an incentive model is developed to stimulate users to participant cooperation.

Selfishness-Aware Incentive in MSNs

Due to the rapid growth of wireless communication [1,2] and sensor technologies

[3], various kinds of content [4-6] can be provided to both mobile networks and users. Currently, mobile crowd sensing (MCS) [7] has emerged as a new paradigm of MSNs. A large number of mobile individuals can use their mobile devices to deliver data, which can be provided to the MCS to collect the information of interest [8-13]. Based on the collected information [14, 15] from the participants, the MCS can get knowledge or make decision for the large scale systems. For example, the passengers on the bus can use their mobile phones to send the real time information, including the current location, the status of traffic, and the situation of overcrowding. Then, with the above information, the MCS can sense the environment around the bus and predict the arriving time, the number of spared seats on the bus, etc. However, it needs all participants to cooperatively collect the information from individuals. How to motivate participation of mobile individuals is an important issue.

However, in practice a large number of users exhibit various degrees of selfishness [16]. For example, some users mainly concern how to save their limited resources (e.g., battery, buffer, etc.), and thus they are not willing to provide information to others. Obviously, users’ selfishness will affect the performance of mobile crowds. In the previous studies, the selfishness is mainly divided into two parts: individual selfishness and social selfishness. The first is from the perspective of an individual, in which the user is not willing to relay and store information for others to save the limited buffer and power resources [17]. The second is from the perspective of a

© Springer International Publishing AG 2016 41

Z. Su et al., Modeling and Optimization for Mobile Social Networks,

DOI 10.1007/978-3-319-47922-4_3

community formed by some individuals with similar interests, where the individuals are more willing to help others in the same community but lack interest to contribute information to the ones out of the community [18].

In this chapter, different from the above conventional work, we characterize users’ selfishness based on different levels: weak selfishness and extreme selfishness. The weak selfishness means that an individual may be unwilling to forward information to others in order to conserve the limited resources but sometimes still has interest to contribute information to others. The extreme selfishness refers that an individual does not want to share information to help others. Here, the extreme selfishness is considered from two aspects. On one hand, some users who are not interested in the information will not accept or store it. On the other hand, some users who have possessed the information have no will to forward it to others. Then, an analytical model for information dissemination with weak and extreme selfishness is presented through the ordinary differential equations (ODEs). This model can be employed to evaluate the influences of both the extreme selfish and weak selfish behavior on the process of dissemination. In addition, real trace driven simulations are conducted to show the effect of both weak selfishness and extreme selfishness in information dissemination.

 
Source
< Prev   CONTENTS   Source   Next >

Related topics