MWBS: An Efficient Many-to-Many Wireless Big Data Delivery Scheme

摘要

Wireless big data raises the demands on the networking schemes to support the efficient group data sharing over heterogeneous wireless technologies, which take many-to-many data delivery as the foundation. Information-centric networking (ICN) approach is a promising networking technology to support big data delivery, which has the potential to establish the harmony between networking and wireless big data sharing. However, the existing ICN schemes have not carefully addressed the many-to-many communications. To address this issue, we propose an efficient and secure many-to-many wireless big data delivery scheme (MWBS) to provide group-based data dissemination and retrieval with name-integrated forwarding. In MWBS, a bi-directional tree is securely constructed for each group through the procedures of group initiation, join, leave, publication, and multi-level inter-zone routing. Especially, Designated Forwarding and Cacheable Nodes (DFCNs) are introduced to act as the roots for the construction of such bi-directional trees. The implementation details of MWBS are provided for function verifications. To effectively deploy MWBS, we investigate the impacts to the MWBS performance from the number and locations of DFCNs, which show that the optimized number of DFCNs can reduce the total traffic cost and the DFCN close to users is preferred to be selected for a group. Finally, simulations are performed to evaluate the performance of MWBS, which show that MWBS can reduce the control packet overhead and the state storage overhead compared to the existing ICN schemes.