Asian Science Citation Index is committed to provide an authoritative, trusted and significant information by the coverage of the most important and influential journals to meet the needs of the global scientific community.  
ASCI Database
308-Lasani Town,
Sargodha Road,
Faisalabad, Pakistan
Fax: +92-41-8815544
Contact Via Web
Suggest a Journal
Articles by xinyu Wang
Total Records ( 5 ) for xinyu Wang
  Jian Wang , Jianling Sun , xinyu Wang and Hang Chen
  Process allocation for load distribution can improve system performance by utilizing resources efficiently. For primary-backup based fault tolerant system, a classic load-balancing process allocation method (two-stage allocation algorithm) has been proposed that can balance the load before as well as after faults occurrence. But two-stage allocation algorithm has bad scalability since its load-balancing performance reduces dramatically when each primary process is duplicated more than once (i.e., has more than one backup process). In this study, we present an improved algorithm named RSA (R-Stage Allocation algorithm) that can have the load better balanced no matter how many backup processes each primary process owns; Simulations are also used to compare the proposed algorithm with the two-stage allocation algorithm and the experimental results show that when extending to replicated R-Resilient processes, RSA has significantly better load distribution performance than two-stage allocation algorithm.
  Xinyu Wang , Xiaohu Yang , Jianling Sun and Zhengong Cai
  Once software is in production, maintenance works are inevitable e.g., bug fixes, requirement changes, etc. During the long-term software maintenance, documents become gradually inconsistent with the system and the source code becomes the most reliable source for design recovery. Currently there exist many approaches to identify component from source code, which adopt different component metrics and clustering algorithms. However, these approaches are not satisfying in the precision and efficiency. This study proposes a new component extraction approach based on 0 metrics and hierarchical clustering algorithm, which greatly improves the precision and efficiency.
  Chao Huang , Jianling Sun , Xinyu Wang and Yuanjie Si
  In this study, we propose the redundancy and inconsistency checking algorithms to support the policy management of systems employing role based access control model. Present method is based on the formal definition of the policy redundancy and policy inconsistency. Via constructing the role graph, we analyze the redundancy and inconsistency one by one. According to the features of each type of redundancy and inconsistency, present algorithm checks all the possible violations and generates the related policy elements to help the security administrator to amend the policy afterwards. The performance test demonstrates that the approach is efficient enough for practical usage. Present approach could guarantee the conciseness as well as consistency of the access control policy, at same time reduce the burden of access control administration significantly.
  Yongsheng Fu , Xinyu Wang and Shanping Li
  Distributed Hash Table (DHT) has proven to be an efficient platform for building a variety of scalable and robust distributed applications like content sharing and location in the internet. However, the adaptation of DHT technology to Mobile Ad-hoc NETwork (MANET) is not straightforward. Network scalability and routing as well as information distribution are major problems for nodes in a MANET, who are only aware of their immediate neighborhood. Several algorithms implement DHT using geographic information in MANET, but they can not adapt well in large-scale network without an efficient localization mechanism. This study propose a new DHT implementation named Distributed Index based on Geographic Hash Table (DI-GHT) in MANET. In DI-GHT, using hashing function, the shared resource location information (index) is mapped to nodes in a geographic area rather than a geographic position. The network is partitioned into domains and DI-GHT distributes resource index in all domains. The requestor finds the index information in the nearest domains using the hash function and then retrieves the resource. The simulation results and analysis show that DI-GHT outperforms original Geographic Hash Table (GHT) in terms of query success rate and message cost.
  Cheng Chang , Ye Wang , Xinyu Wang and Yongsheng Fu
  The lifetime of ad hoc networks is limited by the battery life of individual sensors. As deployed sensors may be owned and supported by different stakeholder, they are selfish agents by instinct, only maximizing their own utility. However, selfish behaviors degrade the network performance significantly. To address this problem, we propose an Anti-Collusion and Fair Reputation-based Incentive mechanism (ACFRI). The ACFRI mechanism measures the reputation of each node based on not only the past routing behaviors but also the evaluated residual energy. Only the neighbor node is responsible for punishment on selfishness of the source node. Other nodes on the routing path can monitor routing behaviors of the source node and punish the neighbor node if it colludes with the source node. The ACFRI mechanism can achieve fairness by the objective reputation and make collusive behaviors unattractive. The experimental results show that the ACFRI mechanism can punish selfish nodes and achieve fairness and anti-collusion successfully, under 10% throughput overhead.
Copyright   |   Desclaimer   |    Privacy Policy   |   Browsers   |   Accessibility