Distributed Decision Fusion Rule using M-ary Source Coding Scheme on Unsymmetrical Hypothesis in Wireless Sensor Network

[[abstract]]無線感測器所構成的無線感測器網路，如今被廣泛應用在各種環境觀測上，用途也相當廣泛，例如軍事、生活、商業活動等等，而無線感測器在做無線傳輸時，能源消耗是以電池做為能量來源，因此電池限制了感測器的工作時間以及效率，所以如何有效控管消耗的電力，延長其有效時數，是一項值得探討的議題。本研究在此設計管理出提高能源效率的演算法來提升正確率。 在傳統非對稱環境中，僅以二元的作法來記錄所觀測之目標，本論文將使用多維訊源編碼，建立於非對稱分散式環境中，將感測器所觀測到之訊號編碼分區，從原本的二元拓展至多元，再根據融合中心的決策法則改善傳統方法的正確率。經由實驗和理論證實此機制相較於傳統，能改善正確率以及達到節省能源的目標。 關鍵詞 : 非對稱分散式感測器網路; 決策融合; 多維訊源編碼機制[[abstract]]Wireless Sensor Network, WSN, composed of wireless sensors is widely used in the application of environment detection such as military, life, or business activities. The wireless sensor is mainly battery-powered and hence power consumption is one of the major issues in WSN. This study focuses on the design of an energy-efficient algorithm whose performance is better than that of the conventional method. In the classical unsymmetrical environment, only binary algorithm is used to record the observation of a sensor. In this thesis, M-ary source coding scheme is adopted on unsymmetrical hypothesis, i.e., the decision on the observation by a sensor is extended from binary algorithm to M-ary algorithm. The results show that the goal of better rate and lower energy consumption compared with that of the conventional algorithm is achieved via the porposed M-ary source coding scheme. Key words: unsymmetrical distributed sensor network, decision fusion, M-ary source coding[[note]]碩