使用蜂群演算法之移動式無線感測網路定位研究

近幾年來智慧型移動裝置越來越普及並在各領域中應用。現今的許多裝置都需要定位功能，例如：醫療護理監測、手機定位或是行動導覽等，所以各種WSN定位方法逐漸受到重視。因此，本研究以移動無線感測器的定位問題做為探討。為了減少定位的誤差，本研究提出了蜂群演算法(artificial bee colony algorithm) 修正估測座標。蜂群演算法的運算原理是在一個空間內假設個食物源，食物源代表一個可行解，尋找可行解之外更好的可行解。可行解的優劣由適應值來判斷，判斷標準可能是食物源離蜂巢的遠近、花蜜的豐富度，若蜜蜂找到更好的可行解就會取代原本的；若是一直找不到更好的解，先記憶目前的解，再重新產生新的可行解，直到回合結束。如果適應值不佳，則放棄該食物源而尋找新目標，再過濾掉超過適應值的食物源進行修正，本研究利用此優點結合改良型蒙地卡羅演算法(IMCL)，IMCL負責前半段估測座標，蜂群演算法再修正估測座標，提高定位的精確度，縮小整體的定位誤差。除此之外，本論文考慮到不同的速度，節點量和不規則傳輸程度(DOI)，同時將考慮避免定位出來的估計座標超過估測範圍。本研究的模擬結果顯示本演算法，有效的減少定位誤差，與其他蒙地卡羅定位演算法MCL、MCB、IMCL相比最低誤差為0.44倍通訊範圍，優於其他三者。

關鍵字

蒙地卡羅定位演算法；無線感測器網路；移動式無線感測器網路；人工蜂群演算法

並列摘要

In recent years, smart mobile devices and the associated applications in various fields have been getting popular and many of them require the positioning function, for example: medical monitoring, mobile positioning and indoor navigation etc. Consequently, localization methods for wireless sensor networks play an important role in the aforementioned applications. In this thesis, a localization method for mobile sensor networks is presented. In order to reduce localization errors, the artificial bee colony (ABC) algorithm is used to modify the estimated coordinates. The fundamental concept of the ABC algorithm is to find better feasible solution (food source) from food sources in the specific space. According to the distance from food to bee colony and richness of pollen, the fitness function (profit function) is often used to calculate the fitness value of food (profitability) and these fitness values determine which food source to be collected. The food source with better fitness value replaces the original one. If the fitness value is not tolerable, then discard the food source and find a new target until the terminal condition is met. Using the idea, this study integrates the ABC algorithm and the improved Monte Carlo localization (IMCL) algorithm to improve the accuracy of positioning. The IMCL estimates the node coordinates and then the ABC refines the location. Also, different speeds, different number of nodes and degree of irregularity (DOI), and out of range prevention of coordinate estimation are considered in thesis. From the simulation results, it is seen that the proposed algorithm can effectively reduce the localization error. Comparing to the other Monte Carlo based localization algorithm MCL, MCB, and IMCL, the resulted estimation error is at most 0.44 times of communication range. It outperforms other three algorithms.

並列關鍵字

Monte Carlo localization algorithm ； wireless sensor network ； mobile wireless sensor network ； artificial bee colony algorithm

參考文獻

[2] I. F. Akyildiz, “A survey on sensor networks,” IEEE Communications Magazine, Aug. 2002, pp. 102-114.

[5] K. Langendoen and N. Reijers, “Distributed localization in wireless sensor networks: a quantitative comparison,” Computer Networks: The International Journal of Computer and Telecommunications Networking archive, vol. 43, issue 4, Nov. 2003.

[6] N. Bulusu, J. Heidemann and D. Estrin, “GPS-less low cost outdoor localization for very small devices,” IEEE Personal Communications Magazine, Oct. 2000.

[7] C. Alippi and G. Vanini, “A RSSI-based and calibrated centralized localization technique for wireless sensor networks,” Fourth Annual IEEE International Conference on Pervasive Computing and Communications Workshops, Mar. 2006, pp. 1-5.

[10] J. P. Sheu, W. K. Hu and J. C. Lin, “Distributed localization scheme for mobile sensor networks,” IEEE Transactions on Mobile Computing, vol. 9, issue 4, Apr. 2010, pp. 516-526.

國際替代計量

使用蜂群演算法之移動式無線感測網路定位研究

未授權

主題瀏覽