在靜態無線感測網路中，如何達成有效的週期性資料收集並延長網路生命週 期已經成為一個重要的研究議題。在這篇文章中，我們提出一個分散式k 流量平衡樹的方法來延長週期性資料收集的網路生命週期。一棵 k 流量平衡樹是指在一個網路中任兩個非葉節點的小孩個數差距不會超過k。我們讓每個節點在挑選父母的時候，會以目前擁有的小孩數以及其鄰居節點個數最少的節點為優先挑選者，來避免任兩個非葉節點的小孩個數差距超過k。此外，為了避免在無線資料傳輸下可能產生的封包碰撞，我們提出一個以分時傳送為基礎的分散式傳輸排程方法。模擬結果顯示我們的方法不但能延長網路生命週期且讓因為排程產生的整體時槽使用時間趨近於集中式方法所產生的整體時槽使用時間。 In static wireless sensor networks, it is an important research issue to achieve effectively periodic data collection and prolong the network lifetime. In this thesis, a distributed k-load-balanced tree scheme is proposed to prolong the network lifetime for periodic data collection. A k-load-balanced tree is a tree that the difference of the number of children of each two non-leaf nodes in a network is less than k. Each node chooses the node with the least current number of children and number of neighbor as parent for avoiding that the difference of the number of children of each two non-leaf nodes in a network is more than k. Furthermore, based on the k-load-balanced tree, a distributed TDMA-based scheduling scheme is proposed to avoid packet collisions in wireless data transmissions. The simulation results show that our scheme can prolong the network lifetime and the network delay time is close to the result of centralized method.
