利用移動式感測網路在時間限制下之目標物偵測

Abstract

無線感測網路已經被用在使用移動式感測器來偵測移動式目標。考 慮在一個封閉的區域空間內，一些移動式感測器會持續不斷地進行巡邏，並且嘗試去發現在此空間內的移動式目標。由於目標會在一定時間內離開這塊空間， 感測器必須要在給定的時間之內發現目標。本論文探討如何在給定時間限制之下去找出需要的感測器數量，以符合應用的要求。我們使用馬可夫模型來計算目標出現的機率，並且根據此機率推算出期望的偵測時間，再利用這些關係來推論出所需要的感測器數量。另外本論文也提供感測器在合作模式之下對時間的影響，並且提出一個讓感測器在分散式的環境交換資訊的演算法。 最後我們使用模擬來驗證模型的準確性、在不同的環境底下對於偵測時間的影響，並且成功推算出所需要的感測器數量。

Wireless sensor networks have been widely studied in arious applications such as environmental monitoring , battlefield surveillance, and intrusion detection recently. One of the important applications is to detect mobile targets (e.g., invaders) by using mobile sensor nodes. Consider a closed region where there are several mobile sensors patrolling the area and some mobile targets in the region. The mobile sensors need to find the targets before a time period because the mobile targets may escape from the monitored region after a certain period of time. In order to meet the deadline, many mobile sensors should be deployed to reduce the detection time. This work aims at finding the number of sensors required in the region through realizing relations among expected detection time, detection strategies, and the number of sensors. Thus, given the number of sensors and detecting robability, we propose a Markov model to derive the expected detection time. Our proposed model can then help determine how many sensors are sufficient such that sensors are able to detect the targets before a given deadline. We also provided a cooperative scheme for sensors that can reduce expected detection time and a distributed protocol for sensors with limited transmission range. The simulation results show that the model can predict the expected detection time precisely, and the number of sensors required derived from the model is adequate.