Multipoint-to-multipoint communication can be implemented by combining the point-to-multipoint and multipoint-to-point connection algorithms. In an ATM multipoint-to-point connection, multiple sources send data to the...
详细信息
Multipoint-to-multipoint communication can be implemented by combining the point-to-multipoint and multipoint-to-point connection algorithms. In an ATM multipoint-to-point connection, multiple sources send data to the same destination on a shared tree. Traffic from multiple branches is merged into a single stream after every merge point. It is sometimes impossible for the network to determine any source-specific characteristics since all sources in the multipoint connection may use the same connection identifiers. The challenge is to develop a fair rate allocation algorithm without per-source accounting as this is inequivalent to per-connection or per-flow accounting in this case. We define fairness objectives for multipoint connections, and we design and simulate an O(1) fair ATM-ABR rate allocation scheme for point-to-point and multipoint connections sharing the same links. Simulation results show that the algorithm performs well and exhibits many desirable properties. We list key modifications necessary for any ATM-ABR rate allocation scheme to fairly accommodate multiple sources. (C) 2002 Elsevier Science B.V. All rights reserved.
Major focus in this software prototype development is the application of automated algorithmic techniques to the planning of communications services on military and commercial satellite systems. The emphasis to date i...
详细信息
ISBN:
(纸本)9781728142807
Major focus in this software prototype development is the application of automated algorithmic techniques to the planning of communications services on military and commercial satellite systems. The emphasis to date is on priority-based military systems, but the flexibility to accommodate commercial systems with varied service access mechanisms is built in. The overall goal is to decrease scenario planning times from days-to-weeks to minutes on standard, but, capable laptop computers. This level of performance has been met in developmental testing, and will enable resilient operations of satellite networks by supporting fast reallocation of terminal resources. The types of algorithms employed depend on the application. For service planning when terminal endpoints are within antenna beam coverage areas, a deterministic linear programming algorithm is employed. When beam coverage is not provided for new services, i.e., when many uncovered terminal endpoints exist, a probabilistic approach using a genetic algorithm is employed to heuristically search for antenna beam configurations meeting prioritized access requests. Alternatively, when lower volumes of new, uncovered endpoints exist, specialized beam adjustment algorithms are applied. Additionally, a framework has been developed to enable operations in an enterprise environment with automated service access request (SAR) processes, which further advances resilient satellite network operations.
暂无评论