延長有能量採集感測器之動態異質感測網路存活時間之研究

Abstract

發展無線感測網路之最大限制是如何去延長此網路之存活時間。過去研究大多分別考慮其動態性與異質性，或僅考慮靜態網路情境，較不符實際情況。此外，近來相關研究的趨勢是著重於開發能量採集感測器的新穎技術，它能使感測器透過吸收如太陽能或風能等自然能量以延長感測器本身之存活時間，進而延長整體網路存活時間。但目前受限於成本與技術問題，將所有無線感測器均改用能量採集感測器仍不可行，故當前最合適的作法是混合採用能量採集感測器與一般感測器，並同時兼顧架設成本與感測傳輸限制。因此，本計畫探討混合有能量採集感測器與一般感測器之動態異質感測網路之存活時間延長問題，當中參考不同時間點之感測器狀態來決定各感測器之休眠排程，以滿足覆蓋到所有目標物的需求，進而延長此網路存活時間。由於靜態無線感測網路之存活時間延長問題已被證明是NP-Complete問題，故此問題亦是，因此本研究首先建立此問題之數學模型，接著提出一改良式多族群和聲搜尋演算法以求解此問題，此方法能幫助解決動態問題並提高能量採集感測器之應用效益。透過實驗模擬以分析演算法所求解之網路存活時間、方法穩定性與運算時間。實驗結果顯示此演算法能得到較佳的結果，且可有效延長網路存活時間。

Lifetime extension has been the major constraint of developing wireless sensor networks (WSNs). Most of previous related works separately considered dynamics and heterogeneity of WSNs, or only considered static network scenarios, so that practical conditions are not achieved. In addition, recent works have tended to developing the technologies of energy-harvesting (EH) sensors, which can absorb natural power (e.g., solar and wind power) to extend lifetime of sensor devices. However, restricted to cost and technology problems, it is infeasible to replace all wireless sensors by EH sensors, so that an appropriate approach is to apply a hybrid WSN consisting of EH sensors and common sensors under constraints of installation cost and transmission. Therefore, this project investigates the problem of extending the lifetime of a dynamic heterogeneous WSN with EH and common sensors, in which we consider dynamic states of each sensor at different time points to determine the sleep schedule of each sensor, so that all targets to be monitored are covered, and the total network lifetime is extended. Since the case for state WSNs has been shown to be NP-complete, the concerned problem is also NP-complete. Hence, this work first models this problem mathematically, and then proposes an improved multi-population harmony search algorithm for this problem. By simulation, the network lifetime, stability, and executing time of the proposed algorithm are analyzed. Experimental results show that the proposed algorithm performs better, and indeed helps extend the total WSN lifetime.