基于兴趣和行为预测的移动社交网络动态资源发现机制

doi:10.11959/j.issn.1000-436x.2016070

通信学报 ›› 2016, Vol. 37 ›› Issue (4): 34-43.doi: 10.11959/j.issn.1000-436x.2016070

基于兴趣和行为预测的移动社交网络动态资源发现机制

李致远^1,²,陈汝龙¹,王汝传²

¹ 江苏大学计算机科学与通信工程学院，江苏镇江212013
² 南京邮电大学江苏省无线传感网高技术研究重点实验室，江苏南京210003

出版日期:2016-04-25 发布日期:2016-04-26
基金资助:
国家自然科学基金资助项目;国家自然科学基金资助项目;国家自然科学基金资助项目;国家自然科学基金资助项目;中国博士后科学基金资助项目;江苏省自然科学基金资助项目;镇江市重点研发计划（社会发展）基金资助项目;江苏大学高级专业人才科研启动基金资助项目

Exploiting interests and behavior prediction for dynamic resource discovery in mobile social networking

Zhi-yuan LI^1,²,Ru-long CHEN¹,Ru-chuan WANG²

¹ School of Computer Science and Telecommunications Engi ring, Jiangsu University, Zhenjiang 212013, China
² Jiangsu High Technology Research Key Lab for Wireless Sensor Networks, Nanjing University of Posts and Telecommunications, Nanjing 210003, China

Online:2016-04-25 Published:2016-04-26
Supported by:
The National Natural Science Foundation of China;The National Natural Science Foundation of China;The National Natural Science Foundation of China;The National Natural Science Foundation of China;The Project Funded by China Postdoctoral Science Foundation;The Natural Science Foundation of Jiangsu Province;The Key Research and Development Program Foundation (Social Development) of Zhejiang;The Senior Professional Scientific Research Foundation of Jiangsu Un versity

摘要/Abstract

摘要：

针对时延容忍的移动社交网络中的资源发现问题，在三维环境下提出一种基于兴趣和行为预测的动态资源发现机制(IBRD)。IBRD 首先从用户的文件资源和信息表中提取兴趣向量，然后通过节点间的余弦相似度计算构造初始的虚拟兴趣社区。之后，通过对移动社交数据的分析，建立半马尔可夫链模型以预测节点的行为和运动轨迹。依据模型的预测结果，实现虚拟兴趣社区的动态维护。最后，基于动态的虚拟兴趣社区构建高效的资源发现策略。IBRD机制在随机网络仿真器(ONE)上得以实现，实验结果表明，IBRD与同类模型相比，具有较高的资源发现成功率、较低的平均时延和通信开销。

关键词: 移动社交网络, 动态资源发现, 余弦相似性, 行为预测, 马尔可夫链

Abstract:

Resource discovery in delay-tolerant mobile social networks (MSN) continues to be challenging issue. An interests and behavior prediction-based dynamic resource discovery mechanism (IBRD) in 3-dimensional cartesian coor-dinate system was proposed. Firstly, IBRD extracted the interest vectors from the user's file resources and the profile table, and then the initial virtual interest communities through the cosine similarity computation between the nodes were con-structed. After mobile social networking data was anal the semi-Markov chain model was used to predict the beha-vior and movement trajectory of users. According to the prediction results of the Markov model, the dynamic ma en-ances of the virtual interest communities were realized. Next, an efficient resource discovery strategy based on the dynamic virtual interest communities was designed. Finally, proposed method was simulated on the platform of the opportunistic network environment simulator. Simulations results show that the proposed scheme consistently outperforms the state-of-the-art resource discovery schemes in terms of the searching efficiency the average delay and the communication cost.

Key words: mobile social networking, dynamic resource discovery, cosine similarity, behavior prediction, Markov chain

李致远,陈汝龙,王汝传. 基于兴趣和行为预测的移动社交网络动态资源发现机制[J]. 通信学报, 2016, 37(4): 34-43.

Zhi-yuan LI,Ru-long CHEN,Ru-chuan WANG. Exploiting interests and behavior prediction for dynamic resource discovery in mobile social networking[J]. Journal on Communications, 2016, 37(4): 34-43.

图/表 10

图1

图2

图3

图4

图5

表1

仿真实验参数"

参数	描述	默认值
MovementModel.world	仿真实验场景面积	600 m × 800 m
Size
Scenario.endTime	仿真时间	12 h
Scenario.nrofHosts	用户节点个数	100
Scenario.nrofHostGroups	分组数	5
BtInterface.transmit	传输带宽	250 kbit/s
Speed
BtInterface.transmit	传输距离	10~20 m
Range
Group.movementModel	组内节点移动模型	ShortestPathMap
		BasedMovement
Mobility Speed	节点的行走速度	0.5~1.5m/s
Message. Interval	产生消息的间隔	1~2s
Message.TTL	消息的生存周期	300 min
r	提取节点兴趣时的实验参数	0.5
$T_{s i m}^{1}$	构建局部兴趣社区的阈值	0.3
$T_{s i m}^{2}$	局部兴趣社区内资源发现的阈值	0.5
a	兴趣相似度的权重	0.6
b	运动模式相似度的权重	0.4

表1

图6

图7

图8

图9

参考文献 15

[1]	WU J , XIAO M , HUANG L . Homingspread: community home-based multi-copy routing in mobile social networks[C]// The 32nd IEEE Conference on Computer Communications(INFOCOM). Turin, Italy, c2013:2319-2327.
[2]	ZHU Y , XU B , SHI X , et al. A survey of social-based routing in delay tolerant networks: positive and negative social effects[J]. IEEE Communication Surveys ＆ Tutorials, 2013,15(1): 387-401.
[3]	张国印，李军．移动对等网络覆盖网[J]．软件学报， 2013,24(1): 139-152. ZHANG G Y , LI J . Overlays in mobile P2P networks[J]. Journal of Software, 2013,24(1): 139-152.
[4]	EAGLE N , PENTLAND A , LAZER D . Inferring social network structure using mobile phone data[J]. Proc Nat'l Academy of Sciences of USA, 2009,106(36): 15274-15278.
[5]	A European Union funded project in situated and autonomic comm[EB/OL]. , 2012.
[6]	SRINIVASAN S , MOGHADAM A , HONG S G . 7DS -node cooper-ation and information exchange in mostly disconnected net-works[C]// IEEE International Conference on Communications. Glasgow, United Kingdom, c2007:1-4.
[7]	SUN H X , WU C . Epidemic forwarding in mobile social networks[C]// The 11th IEEE International Conference on Communi-cations(ICC 2012). Ottawa, Canada, c2012:1421-1425.
[8]	TCHAKAROV J B , VAIDYA N H . Efficient content location n wire-less ad hoc networks[C]// The IEEE International Conference on Mo-bile Data Management(MDM'04). Berkeley, CA, United States, c2004:74-85.
[9]	HOH C , HWANG R . P2P file sharing system over MANET based on swarm intelligence: a cross-layer design[C]// The IEEE Wireless Com-munications and Networking Conference(WCNC '07). Hong Kong, China, c2004:2674-2679.
[10]	LI Y , XU D , XU J , et al. Spatial-aware interest group queries in loca-tion-based social networks[J]. Data ＆ Knowledge Engineering, 2014,(92):20-38.
[11]	YOU L , LI J , WEI C , et al. MPAR: a movement pattern-aware optimal routing for social delay tolerant networks[J].Ad Hoc 2015,(24):228-249.
[12]	EYUPHAN B , BOLESLAW K S . Exploiting friendship relations for efficient routing in mobile social networks[J]. IEEE Transactions on Parallel and Distributed Systems, 2012,23(12): 2254-2265.
[13]	CHEN K , SHEN H Y , ZHANG H . Leveraging social networks for P2P content-based file sharing in disconnected MANET[J]. IEEE Transac-tions on Mobile Computing, 2014,13(2): 235-249.
[14]	LIU G X , SHEN H Y , WARD L . An efficient and trustworthy P2P and social network integrated file sharing system[J]. IEEE Transactions on Computers, 2015,64(1): 54-70.
[15]	ARI K , JRG O , TEEMU K . The one simulator for DTN protocol evaluation[C]// The 2nd Int Conf on Simulation Tools and Techniques(ICST'09). c2009:10-35.

基于兴趣和行为预测的移动社交网络动态资源发现机制

Exploiting interests and behavior prediction for dynamic resource discovery in mobile social networking

在线阅读

PDF下载

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 15

相关文章 15

Metrics

推荐阅读 0

[1]	康海燕, 龙墨澜. 基于吸收马尔可夫链攻击图的网络攻击分析方法研究[J]. 通信学报, 2023, 44(2): 122-135.
[2]	彭滔, 钟文韬, 王国军, 罗恩韬, 熊金波, 刘忆宁, Hao Wang. 移动社交网络中面向隐私保护的精确好友匹配[J]. 通信学报, 2022, 43(11): 90-103.
[3]	佘维, 荣欣鹏, 刘炜, 田钊. 基于马尔可夫链的生成式区块链隐蔽通信模型[J]. 通信学报, 2022, 43(10): 121-132.
[4]	简鑫,韦一笑,刘钰芩,宋健,曾孝平,谭晓衡. 窄带物联网非连续接收机制功耗模型与优化[J]. 通信学报, 2019, 40(4): 107-116.
[5]	田鹤,赵海,王进法,林川. 互联网传播行为的时序演化与预测[J]. 通信学报, 2018, 39(6): 116-126.
[6]	简鑫,刘钰芩,韦一笑,宋健,王芳,付澍,谭晓衡. 窄带物联网覆盖类别更新机制性能分析与优化[J]. 通信学报, 2018, 39(11): 70-79.
[7]	田贤忠,郭敏,何佳存,刘高,祝驿楠. 能量捕获协作中继网络多中继节点选择策略[J]. 通信学报, 2017, 38(Z2): 1-7.
[8]	李峰,司亚利,陈真,申利民. 基于马尔可夫链的轻量级机会路由转发策略[J]. 通信学报, 2017, 38(5): 108-120.
[9]	赵泉华,石雪,王玉,李玉. 可变类空间约束高斯混合模型遥感图像分割[J]. 通信学报, 2017, 38(2): 34-43.
[10]	刘浩,陈志刚,张连明. 移动社交网络中基于拍卖模型的数据转发激励机制[J]. 通信学报, 2017, 38(11): 111-120.
[11]	周翰逊,郭薇,刘建,贾大宇. 基于马尔可夫链的网络蠕虫传播模型[J]. 通信学报, 2015, 36(5): 66-71.
[12]	刘燕,胡斌,徐立平. T函数Walsh谱值与差分转移概率快速算法[J]. 通信学报, 2015, 36(5): 167-173.
[13]	陈曙晖,徐成成. 基于两级存储的正则表达式匹配技术[J]. 通信学报, 2014, 35(6): 47-55.
[14]	陈曙晖，徐成成. 基于两级存储的正则表达式匹配技术[J]. 通信学报, 2014, 35(6): 7-55.
[15]	屈毓锛,陈晨,董超,王海. 基于Markov状态转移方法的网络编码时延分析[J]. 通信学报, 2013, 34(9): 77-83.