ZigBee網路之效能最佳化

Abstract

ZigBee是一個無線個人區域網路的通訊標準，具有極低成本、極低複雜度、極為省電的特性。在眾多ZigBee網路拓樸中，Cluster-tree是特別適用於低耗電和低複雜度的無線感測網路，因為這樣的網路支援ZigBee的省電運作模式和輕量路由。這篇論文的主要目的是針對ZigBee Cluster-tree網路達成效能的最佳化。首先，本篇論文考量ZigBee的小範圍Cluster-tree網路，提出了一個可支援general traffic distribution的分析系統，針對其獨有的省電運作模式作一個全盤的效能分析。本篇論文接著延伸至大範圍的ZigBee網路。針對吞吐量最佳化的議題，提出了一個以養父母概念為基底的架構，可增加頻寬的使用量且不會造成多餘的訊息交換。為了在這個架構上面達到吞吐量最佳化，本研究模型化了一個節點限制的最大流量問題，並針對這個問題提出了一個可完全套用於ZigBee的分散式演算法，並證明此演算法具有收歛性且可以達到最佳解。針對同樣的大範圍ZigBee網路，本篇論文最後提出一個提供服務品質保證的架構。延用此論文的第一個部份研究成果，此架構能夠針對general traffic distribution的即時網路流量來源，達到機率式的傳輸延遲保證。經過一連串的網路模擬以及完整數學分析模型，本篇論文所提出的改進方法已被驗證的確可達到顯著的效能提升。

ZigBee, a unique communication standard designed for low-rate wireless personal area networks, has extremely low complexity, cost, and power consumption for wireless connectivity of inexpensive, portable, and moving devices. Among the well-known ZigBee topologies, the cluster-tree is especially suitable for low-power and low-cost wireless sensor networks due to its support of power saving operation and light-weight routing. The objective of this dissertation is to optimize the performance for ZigBee cluster-tree networks. This dissertation starts with the consideration of star networks, i.e., small-scale cluster-tree networks, for ZigBee. The low-power operation of a star-based ZigBee network follows the IEEE 802.15.4 specification. A comprehensive analysis for low-power operation for IEEE 802.15.4 is conducted. Specifically, an analytical model which can accommodate a general traffic distribution is developed. This dissertation then extends the performance study for ZigBee to accommodate large-scale cluster-tree networks. For throughput optimization, an adoptive-parent-based framework is presented for a ZigBee cluster-tree network to increase the bandwidth utilization without any extra message exchange. To optimize the throughput in the framework, a vertex-constraint maximum flow problem is formulated, and a distributed algorithm with full compatibility of the ZigBee standard is developed. This dissertation is concluded by developing a QoS-oriented framework to provide stochastic QoS guarantee for urgent information delivery in ZigBee cluster-tree networks. This framework can accommodate a general traffic model extended by our first work for IEEE 802.15.4 to adapt diverse traffic characteristics in various applications. The capabilities of the proposed approaches are demonstrated by a series of experiments.