Guest Editorial: Advances in Communication Networking

This special issue of the Computer Communications journal is dedicated to present state of the art research results in the advances in Communication Networking. We accepted 19 papers out of many submissions from all over the world. The majority of these papers are extended versions of selected papers accepted in the 2006 International Symposium on Performance Evaluation of Computer and Telecommunication Systems, SPECTS’2006. This means that they have undergone a thorough and a critical review process by experts in the field. The first version of each of these papers has been refereed by at least three technical program committee members of SPECTS’2006 and the extended version of each selected paper has undergone a through second review cycle by at least three referees. This process speaks about the quality of these papers. Yuen and Schulzrinne propose a Bloom filter based (BF) scheme to improve search efficiency. A data holder summarizes its content by a Bloom filter and broadcasts it to the data querier. The data querier then transmits a query only when the requested data is available. For the single neighbor scenario, they derive the long run bandwidth utilization of a connection, as a function of random connection time T. For the multiple neighbors scenario, they derive the search efficiency, which accounts for the possibility of multiple Bloom filter transmissions over a busy period, and a higher query success probability during querying. The performance evaluation results show that the Bloom filter overhead is negligible even at low bandwidth scenarios, when the number of data objects is modestly large. In the dissemination of many types of content, it is expected that the query success probability is small. The BF scheme is therefore a desirable alternative to the QB scheme by offering a superior search efficiency, and robustness against performance falloff associated with the QB scheme when the query success probability is small. On the other hand, they use a node-centric analysis and focus on the coverage area of an arbitrary observer node. This avoids the use of discrete locations in a location-centric model, which requires the use of unrealistic interference and mobility models. The performance evaluation study uses the M/M/1 queue to model the effect of node mobility