We introduce a challenging problem in establishing and initially configuring a Free Space Optical (FSO) network. In such networks, it is assumed that each communication node is a base station, including a router and o...
详细信息
ISBN:
(纸本)0769523129
We introduce a challenging problem in establishing and initially configuring a Free Space Optical (FSO) network. In such networks, it is assumed that each communication node is a base station, including a router and optical transceivers, and its number of transceivers is limited. In addition, the FSO networks are characterized by narrow beam, directional links (operating at 1550 nm, for example). The problem is to initially configure the transceivers to form a connected topology - an NP-complete problem because of the transceiver limitation. It also needs to configure the transceivers in a "distributed" fashion, because a node can have direct knowledge of only its neighbors. We have developed a fully distributed approximation algorithm, which constructs a spanning tree with maximal node degree at most one larger than that in the optimal solution. Due to its distributed nature, this algorithm outperforms known serial algorithms. For a graph with 200 nodes generated in some randomized model, speedups greater than 6 have been demonstrated.
This paper explores the problem of efficiently ordering interprocessor communication operations in statically-scheduled multiprocessors for iterative dataflow graphs. In most digital signal processing applications, th...
详细信息
This paper explores the problem of efficiently ordering interprocessor communication operations in statically-scheduled multiprocessors for iterative dataflow graphs. In most digital signal processing applications, the throughput of the system is significantly affected by communication costs. By explicitly modeling these costs within an effective graph-theoretic analysis framework, we show that ordered transaction schedules can significantly outperform self-timed schedules even when synchronization costs are low. However, we also show that when communication latencies are non-negligible, finding an optimal transaction order given a static schedule is an NP-complete problem, and that this intractability holds both under iterative and non-iterative execution. We develop new heuristics for finding efficient transaction orders, and perform an experimental comparison to gauge the performance of these heuristics.
暂无评论