[ Jiang, Yunxia ; Song, Yankai ; Chen, Anhua ; Fu, Qi ] School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan, China;[ Chen, Zhigang ] College of Information Science and Engineering, Central South University, Changsha, China
Yu, Fei;Chang, Chin-Chen;Chen, Ben M.;Chen, Yongjun;Chen, Zhigang
Journal of Software,2008年3(9):1-2 ISSN：1796-217X
[ Chen, Ben M. ] National University of Singapore, Singapore;[ Chen, Yongjun ] Guangdong University of Business Studies, Guangdong, China;[Yu, Fei ] Peoples' Friendship University of Russia, Russia;[ Chang, Chin-Chen ] National Chung Hsing University, Taiwan;[ Chen, Zhigang ] Central South University, China
[ Chen, Zhigang ; Zhang, Deyu ; Zhang, Lin ] School of Software, Central South University, Changsha, China;[ Chen, Zhigang ; Zhang, Deyu ; Zhang, Lin ] Mobile Health Ministry of Education China Mobile Joint Laboratory, Changsha, China
Data fragments - Different treatments - Patient information - Performance analysis - Privacy protection - Reliable transmission - Security - Wireless network environment
Channel assignment and link scheduling are the key issues in wireless mesh networks with multiple interfaces and multiple channels. In this paper, we proposed a load balance link layer protocol, LBLP, which focuses on these issues aiming to improve the network capacity. The protocol, LBLP, classifies interfaces into three categories: static ones, dynamic ones and adaptive ones, with the following three main contributions. First, to minimize the interference among interfaces, a channel assignment for static interfaces is proposed by considering not only the number and positions of the channels but also the active status of the interfaces; Second, the link scheduler seeks a mechanism to map channel queues with the dynamic interfaces aiming to reduce the negative impacts of the interface switching delay; Third, according to the load, we adjust the status of the adaptive interfaces to avoid bottleneck so as to balance the load among interfaces. Simulation results show that LBLP improve the network throughput significantly despite the large switching delay and limited interfaces.
[Zeng, Zhiwen ] School of Information Science and Engineering, Central South University, Changsha 410083, China;[ Gao, Ya ; Chen, Zhigang ; Deng, Xiaoheng ] School of Software, Central South University, Changsha 410083, China
Global searches - Malicious nodes - Optimum solutions - P2p networks - Particle swarms - Searching algorithms - Short times - Simulation results - Trust models - Trust paths
Routing is a very challenging task in opportunistic networks due to the lack of continuous end-to-end path between nodes. To improve the routing performance, a type of community-based routing protocol is presented. This routing generally turns to the social attributes of nodes to construct the community, and utilizes community characteristics (nodes within the same community come in contact with each other more frequently) to select the relay nodes and make the forwarding decisions. Unfortunately, this routing has a problem of transmission bottleneck between communities. To address this issue, we propose an enhanced community-based routing assisted by ferry in opportunistic networks. This routing resorts to some super-nodes as ferry nodes to ferry messages between communities in order to overcome the network division. The simulation results demonstrate the efficiency and effectiveness of the proposed scheme by comparing it with existing routing schemes.