Incentive evolutionary game model for opportunistic social networks

doi:10.1016/j.future.2019.07.034

Future Generation Computer Systems

Volume 102, January 2020, Pages 14-29

https://doi.org/10.1016/j.future.2019.07.034 Get rights and content

Highlights

•
An incentive evolutionary game model for Opportunistic social networks is proposed.
•
Effectively combines the credit based incentive approach with evolutionary game model.
•
Can eliminate the abnormal nodes with the evolutionary method of survival of the fittest.
•
Encourage network nodes to maintain stable and reliable cooperation behavior.
•
The results show that the proposed model is effective to improve the network performance.

Abstract

Opportunistic social networks transmit messages through the way of nodes cooperation. Because it is a kind of fully open networks, message forwarding is vulnerable to interference from abnormal nodes such as selfish and malicious nodes, who may result in disorder of network transmission. In order to suppress the bad impact on the network from those abnormal nodes, an incentive evolutionary game model for stimulating cooperation among nodes is proposed in this paper. By constructing our game model in routers, normal nodes and abnormal nodes are both encouraged to participate in network collaboration by self-evolution of gaming. Then after evolution, the entire network can reach a general cooperative state of nodes. In order to guide the evolution of gaming, a credit based incentive approach is proposed. The reward and punishment factor is designed in the revenue payment function of the game. The effectiveness is that the cooperative behavior is rewarded, and the non cooperative behavior is punished. Besides, nodes can update their policies of interaction with others by imitating their neighbors. The proposed model is implemented in several popular routers of opportunistic networks, and various scenarios are simulated. The experiments evaluate the performance of proposed model in terms of effectiveness, performance and scalability. The results show that our model is effective to improve the network performance when there are selfish nodes or malicious nodes, and It can be extended to a variety of network sizes. Besides, our incentive evolutionary game model can be easily applied to the current popular routers of opportunistic networks.

Introduction

In the transmission process of the opportunity social network communication, the message is passed through the cooperative mode of storage–carry–forward after the meeting of the nodes. However, there is also interference and destruction from internal selfishness or malicious nodes in the fully open opportunity network. For example, selfish nodes do not participate in cooperative forwarding, malicious nodes replay messages or block message transmission resulting in network transmission disorder. Therefore, the essence of a reliable message delivery is that the nodes have the willingness to cooperate. That means the node’s voluntary participation in routing cooperation, assisting other nodes to forward messages, and does not execute deception, camouflage or attack. At the same time, it does not assist in forwarding malicious information or interacting with malicious nodes. In fact, there is a social phenomenon among nodes in the opportunity network. For example, most nodes are willing to assist other nodes who meet frequently, but cooperation behavior is complex which is decided by multiple factors such as time, space, neighbors history and behavior. Therefore, understanding collaboration motivation in the opportunity network is a new problem to be solved urgently.

Actually, how to promote the willingness of collaboration among nodes and maintain a stable cooperative state in an opportunity network with selfish and malicious nodes inside is still a big problem to be solved. It can be divided into two sub problems: one is the mechanism problem of how the willingness of collaboration generates, and the other is the evolution problem. To solve the problem of collaboration willingness, we need to explore the reasons for the generation of node collaborations, and explore how to suppress the destructive impact on the entire network from abnormal behavior nodes. To solve the problem of evolution, we need to start from the modeling of the complex evolution process of opportunistic network, and we need to accurately depict the actual process of the evolution of node cooperative behavior.

Based on deep analysis of node behavior characteristics in opportunistic network, we introduces evolutionary game theory to explore the node cooperation mechanism in opportunistic network. The whole idea is motivated by a theoretical consideration because the selfish problems can be transfer to be cooperative problems between nodes. Selfish nodes show non-cooperative to other nodes. Usually, the assumption of network is that the node will dynamically adjust its behavior according to different network conditions to maximize their rewards. We assume that each node has an medium default trust value when it join the network. Other nodes can use default trust value to make the initial choice when they meet new node who are not seen before. When the node behaves good, the credits would be increased, when it behaves bad, the credits would be decreased. In opportunistic networks, due to the dynamic characteristics of nodes, it is almost impossible to obtain complete information of the whole network, and it is difficult for nodes to achieve perfect rationality.

Our contribution is that we introduce an evolutionary game theory into OSN as an analytical tool to study the evolution of cooperative behavior of nodes in opportunistic networks. Unlike traditional game theory, evolutionary game theory does not require participants to be completely rational, nor does it require perfect information. In opportunistic networks, different nodes are not uniformly mixed distribution, and the probability of encounter between nodes is not uniformly distributed. Usually, a node will only play games with the nodes in its communication range, not with all the nodes in the competition and cooperation relationship. Therefore, it is necessary to further apply evolutionary graph theory to study the cooperative mechanism of opportunistic network nodes in the context of considering network space.

The purpose of this paper is to understand the evolution procedure of cooperative and non cooperative behavior of nodes, analyze the formation mechanism of network nodes from individual selfish behavior to group cooperative behavior, and analyze the necessary conditions for generating collaboration in the process of node interaction. Besides, based on our evolution method, it is able to suppress malicious or selfish behaviors, and to promote the stable and reliable cooperative state of the nodes in the network.

Section snippets

Related work

Our model is related to both areas, evolutionary gaming and incentive mechanism. In the following related work, we mainly introduce them from two aspects.

Incentive evolutionary game theory model

The evolutionary game theory abandons the hypothesis of the perfect rationality of the classic game theory, and emphasizes the limited rationality of the participants and the dynamic evolution of the game process. The assumption of bounded rationality means that nodes in the opportunistic network can only know part of the knowledge of the whole network game state, and it is impossible to know the overall information of the game, which coincided with the characteristics of local. The

Simulation

We conducted a simulation of our framework in the Opportunistic Network Environment simulator (ONE) [21]. In order to close to physical environment, we use our campus (Shenzhen University Town) map as an experimental basic environment to simulate the opportunistic network of pedestrians carrying Bluetooth communication devices. The size of the experimental area is 2100 m $\times$ 1250 m, as shown in Fig. 2. We set up a total simulation time of 12 h (43200 s), and the scene update period is 0.1 s. The

Conclusions

An incentive evolutionary game model is proposed in this paper, which effectively combines the credit based incentive approach with evolutionary game model. It can eliminate the abnormal nodes with the evolutionary method of survival of the fittest, and encourage network nodes to maintain stable and reliable cooperation behavior, thereby inhibiting the impact of abnormal nodes on network performance. From the simulation results on effectiveness, performance and scalability, it shows that our

Declaration of Competing Interest

There is no conflicts of interest among the authors.

Acknowledgments

This paper was fully financially supported by King Saud University through the Vice Deanship of Scientific Research Chairs: Chair of Pervasive and Mobile Computing. Mohammad Mehedi Hassan is the corresponding author of this paper.

Dr. Eric Ke Wang is an associate professor of Harbin Institute of Technology (HIT), China. Currently, he work as a senior researcher at Key Laboratory of Shenzhen Internet Information Collaborative Technology and Application of HIT. He received a Ph.D. from department of computer science, the University of Hong Kong in 2009. His main research interests include information security. He has obtained two granted projects from National Science Funding (NSFC) of China. Besides, he has developed two

References (31)

SongQ.Q. et al.
Optional contributions have positive effects for volunteering public goods games
Physica A
(2011)
PercM. et al.
Coevolutionary gamesa mini review
BioSystems
(2010)
DengX. et al.
A belief-based evolutionarily stable policies
J. Theoret. Biol.
(2014)
GoswamiA. et al.
Evolutionary stability of reputation based incentive mechanisms in P2P systems
IEEE Commun. Lett.
(2017)
LiMingchu et al.
RIMNet: REcommendation incentive mechanism based on evolutionary game dynamics in peer-to-peer service networks
Knowl. Based Syst.
(2019)
Burton-ChellewM.N. et al.
Conditional cooperation and confusion in public-goods experiments
Proc. Natl. Acad. Sci. USA
(2016)
PfattheicherS. et al.
The intuitive system and antisocial punishment in the public goods game
Personality Soc. Psychol. Bull.
(2017)
PercM.
Sustainable institutionalized punishment requires elimination of wecond-order free-riders
Sci. Rep.
(2012)
SunR. et al.
An incentive mechanism for P2P network using accumulated-payoff based snowdrift game model
ChengD. et al.
Modeling, analysis and control of networked evolutionary games
IEEE Trans. Autom. Control
(2015)

LiJ. et al.

Pricing based power control game for cognitive radio networks

BanD. et al.

Distributed scheduling algorithm for barrier coverage in wireless sensor networks

LiJ. et al.

Computing nash equilibria and evolutionarily stable states of evolutionary games

IEEE Trans. Evol. Comput.

(2016)

ChaiB. et al.

Demand response management with multiple utility companies: A two-level game approach

IEEE Trans. Smart Grid

(2014)

ChengD. et al.

Evolutionarily stable policies of networked evolutionary games

IEEE Trans. Nneural Netw. Imitating Syst.

(2014)

Cited by (31)

Stable reputation management system in P2P networks: An evolutionary perspective
2024, Journal of Computational Science
In P2P networks, peers tend to “free-ride” on the contributions of others. Many reputation-based resource sharing mechanisms have been proposed to mitigate the problem of free riding. However, the evolutionary stability of these mechanisms that require peers to keep track of the reputations of other peers has not been investigated. We show that, without any extra benefit, systems with reputation-based incentive mechanisms are not evolutionary stable. We also show that if the fraction of peers that record reputation and reciprocate accordingly is above a certain threshold, then free riding can be controlled, and a reputation management system is beneficial only when the fraction of free riders is above a certain threshold. Furthermore, we show that imposing a round-based fee and dividing it among the peers as an incentive to keep track of reputation is an evolutionary stable strategy. In this process, we analyse all the possible equilibria in different settings of the system.
A comprehensive survey on Machine Learning techniques in opportunistic networks: Advances, challenges and future directions
2024, Pervasive and Mobile Computing
Machine Learning (ML) is growing in popularity and is applied in numerous fields to solve complex problems. Opportunistic Networks are a type of Ad-hoc Network where a contemporaneous path does not always exist. So, forwarding methods that assume the availability of contemporaneous paths does not work. ML techniques are applied to resolve the fundamental problems in Opportunistic Networks, including contact probability, link prediction, making a forwarding decision, friendship strength, and dynamic topology. The paper summarises different ML techniques applied in Opportunistic Networks with their benefits, research challenges, and future opportunities. The study provides insight into the Opportunistic Networks with ML and motivates the researcher to apply ML techniques to overcome various challenges in Opportunistic Networks.
Decentralized content sharing in mobile ad-hoc networks: A survey
2023, Digital Communications and Networks
The evolution of smart mobile devices has significantly impacted the way we generate and share contents and introduced a huge volume of Internet traffic. To address this issue and take advantage of the short-range communication capabilities of smart mobile devices, the decentralized content sharing approach has emerged as a suitable and promising alternative. Decentralized content sharing uses a peer-to-peer network among co-located smart mobile device users to fulfil content requests. Several articles have been published to date to address its different aspects including group management, interest extraction, message forwarding, participation incentive, and content replication. This survey paper summarizes and critically analyzes recent advancements in decentralized content sharing and highlights potential research issues that need further consideration.
A social rumor and anti-rumor game diffusion model based on sparse representation and tensor completion
2022, Journal of Network and Computer Applications
Citation Excerpt :
Starting from data on user behavior from the rumor and anti-rumor space, a sparse representation of the characteristics of the diffusion space (Zhang et al., 2019; Rubinstein et al., 2010) is made along with a homomorphic compensation of the topic data by tensor completion (Liu et al., 2012; Xie et al., 2019). Finally, a predictive study of user behavior in the topic space is made by combining evolutionary game theory (Bu et al., 2019; Jiang et al., 2014; Wang et al., 2020) with the antagonistic and symbiotic relationship between the rumor and anti-rumor in the diffusion process, and quantifying the influence of rumors and anti-rumors on user behavior. The contributions of this paper can be summarized as follows:
In view of the high dimensionality and complexity of the rumor diffusion space in social networks, and considering the game confrontation in the process of rumor diffusion, this paper proposes a rumor and anti-rumor game diffusion model based on sparse representation and tensor completion. First, taking advantage of the sparse representation’s ability to represent all of the original samples with as few atoms as possible to enable low-rank vectorization of the feature space. Second, in view of the fact that a tensor complement can recover lost data with high precision, a tensor complement with a time-decay function is used to complement the dynamic behavior data of users during the rumor process. Finally, considering the game relationship between rumors and anti-rumors during diffusion, evolutionary game theory is introduced. In addition, this paper proposes a cooperation-and-competition graph convolutional network (CC-GCN)-based model for predicting user behavior. Experiments demonstrate that the CC-GCN model not only improves the prediction accuracy of user behavior but also clearly reflects the game relationship between rumors and anti-rumors in the diffusion space.
Opportunistic routing in wireless networks using bitmap-based weighted tree
2021, Computer Networks
Citation Excerpt :
One of the great advantages of mobile wireless multi-hop network is that they can form spontaneously and quickly, that's why it's called "ad hoc networks" [1]. In fact, messages are sent in opportunistic networks, especially opportunistic social networks, through the collaboration of nodes [2]. In wireless networks, because of the mobility of nodes and that there is no specific duration for the connection between nodes, a method of routing and sending packets should be chosen that will find and send the packet in the shortest possible time.
Multi-hop wireless networks consist of a group of nodes that do not rely on pre-built infrastructures such as routers in wired networks or access points on managed wireless networks. Instead, each node contributes to the routing by forwarding data to other nodes. Each node can act as a router to transmit traffic to its intended destination. Opportunistic routing is used as an approach to increase the efficiency of multi-hop wireless networks. By using opportunistic routing, each packet is allowed to dynamically create a path from source to destination. This is done according to the conditions of the wireless links as the packet is being transmitted. In wireless networks, due to the mobility of nodes and no specific time for the connection between the nodes, it is necessary to select a method for routing and sending packets in the shortest possible time possible. The existence of interim and definitive communications, limited energy of the nodes, and additional network traffic are the challenges of multi-hop wireless networks. In the opportunistic routing, the next step of the packet transfer is determined in the same step. Choosing the next step is another key challenge in opportunistic routing. Most routing methods have problems such as low delivery rates, high delivery numbers, and high delivery times. This paper presents a bitmap-based approach. Using a bitmap, a tree is designed to send packets directly, without any additional and pointless routes, upon receiving a request to send a message to a specified node. Because in this way, each packet will have its unique path that finding it won't be time-consuming. The simulation results show that the proposed method has improved in terms of packet delivery ratio, delivery average, and the number of shipments compared to similar methods.
New trends and applications in social media analytics
2021, Future Generation Computer Systems
Citation Excerpt :
Next section provides a brief description of the main contents of each article. The papers selected for this issue are about new trends and applications in domains like social networks, big data, and Web of things (WoT) [23–28], sentiment analysis [29–31], community & component analysis, question & answering [25,32], network metrics [25,26,33], machine learning applications [25,27,30,31], metaheuristics [28,33,34] and data visualization [24]. The main contributions of each work are briefly summarized below.
The fast growth of social media platforms and their related applications have dramatically changed the way billions of people relate to each other on the Web. This evolution of social media has blossomed in a plethora of end-user, or user-centered, applications that required innovative and efficient techniques for data processing. This was made possible recently thanks to advances in data science and artificial intelligence in fields like pattern recognition, information fusion, knowledge discovery and data visualization. This special issue provides a set of 12 selected papers (out of 65 submissions) that represent latest advances and developments in these areas.

View all citing articles on Scopus

Dr. Chien-Ming Chen received the Ph.D. degree from the National Tsing Hua University, Taiwan. He is currently an Associate Professor with the College of Computer Science and Technology, Shandong University of Science and Technology, Shandong, China. He has published over 100 refereed papers in international journals or conferences.

Dr. Siu-Ming Yiu is an associate professor in the University of Hong Kong, he received a B.Sc. in Computer Science from the Chinese University of Hong Kong, a MS in Computer and Information Science from Temple University, and a Ph.D. in Computer Science from The University of Hong Kong. He has published over 100 refereed papers in international journals or conferences. And he received the honor of 2016 and 2017 ESI high cited scientist in Hong Kong.

Mohammad Mehedi Hassan (M’12 SM’18) is currently an Associate Professor of Information Systems Department in the College of Computer and Information Sciences (CCIS), King Saud University (KSU), Riyadh, Kingdom of Saudi Arabia. He received his Ph.D. degree in Computer Engineering from Kyung Hee University, South Korea in February 2011. He has authored and coauthored around 180+ publications including refereed IEEE/ACM/Springer/Elsevier journals, conference papers, books, and book chapters. Recently, his 4 publications have been recognized as the ESI Highly Cited Papers. He has served as chair, and Technical Program Committee member in numerous reputed international conferences/workshops such as IEEE CCNC, ACM BodyNets, IEEE HPCC etc. He is a recipient of a number of awards including Best Journal Paper Award from IEEE Systems Journal in 2018, Best Paper Award from CloudComp in 2014 conference, and the Excellence in Research Award from King Saud University (2 times in row, 2015 & 2016). He is on the editorial board of IEEE Access, and Elsevier Computer and Electrical Engineering Journal. He has also played role of the guest editor of several international ISI-indexed journals such as IEEE Internet of Things, Information Sciences (Elsevier), and Future Generation Computer Systems (Elsevier). He has secured several national and international research grants in the domain of cloud computing and sensor network. His research interests include Cloud computing, Edge computing, Internet of things, Body sensor network, Big data, Deep learning, Mobile cloud, Smart computing, Wireless sensor network, 5G network, and social network. He is a Senior Member of the IEEE.

Dr. Majed Alrubaian has completed his PhD at the Department of Information Systems in the College of Computer and Information Sciences (CCIS), King Saud University (KSU), Riyadh, Kingdom of Saudi Arabia in 2015. He has authored several papers in the refereed IEEE/ ACM/ Springer journals and conferences. He is a Student Member of the ACM. His research interests include social media analysis, data analytics and mining, social computing, information credibility, and cyber security.

Giancarlo Fortino is Full Professor of Computer Engineering at the Dept. of Informatics, Modeling, Electronics, and Systems of the University of Calabria (Unical), Italy. He received a Ph.D. in Computer Engineering from Unical in 2000. He is also guest professor at Wuhan University of Technology (Wuhan, China), high-end expert at HUST (China), and senior research fellow at the Italian National Research Council ICAR Institute. He is the director of the SPEME lab at Unical as well as co-chair of Joint labs on IoT established between Unical and WUT and SMU Chinese universities, respectively. His research interests include agent-based computing, wireless (body) sensor networks, and Internet of Things. He is author of over 400 papers in int’l journals, conferences and books. He is (founding) series editor of IEEE Press Book Series on Human-Machine Systems and EiC of Springer Internet of Things series and AE of many int’l journals such as IEEE TAC, IEEE THMS, IEEE IoTJ, IEEE SJ, IEEE SMCM, Information Fusion, JNCA, EAAI, etc. He is cofounder and CEO of SenSysCal S.r.l., a Unical spinoff focused on innovative IoT systems. Fortino is currently member of the IEEE SMCS BoG and of the IEEE Press BoG, and chair of the IEEE SMCS Italian Chapter.

View full text

Incentive evolutionary game model for opportunistic social networks

Highlights

Abstract

Introduction

Section snippets

Related work

Incentive evolutionary game theory model

Simulation

Conclusions

Declaration of Competing Interest

Acknowledgments

Physica A

BioSystems

J. Theoret. Biol.

IEEE Commun. Lett.

Knowl. Based Syst.

Conditional cooperation and confusion in public-goods experiments

Proc. Natl. Acad. Sci. USA

The intuitive system and antisocial punishment in the public goods game

Personality Soc. Psychol. Bull.

Sustainable institutionalized punishment requires elimination of wecond-order free-riders

Sci. Rep.

An incentive mechanism for P2P network using accumulated-payoff based snowdrift game model

Modeling, analysis and control of networked evolutionary games

IEEE Trans. Autom. Control