本文選題：無線傳感器網(wǎng)絡 + 數(shù)據(jù)融合��； 參考：《重慶大學》2014年碩士論文

【摘要】：作為物聯(lián)網(wǎng)的底層技術支撐，無線傳感器網(wǎng)絡采集和傳輸監(jiān)測區(qū)域的各種信息，，為軍事國防、遠程醫(yī)療和環(huán)境監(jiān)測等應用提供實時可靠的數(shù)據(jù)，是物聯(lián)網(wǎng)的關鍵信息傳輸技術。但是，無線傳感器網(wǎng)絡節(jié)點分布密集并且采樣頻繁。感知數(shù)據(jù)之間的時空相關性造成冗余數(shù)據(jù)。大量冗余數(shù)據(jù)的傳輸對無線傳感器網(wǎng)絡有限的能量、存儲能力和網(wǎng)絡帶寬帶來了巨大的挑戰(zhàn)。數(shù)據(jù)融合能有效地減少冗余數(shù)據(jù)，提高數(shù)據(jù)收集效率和準確性。因此，無線傳感器網(wǎng)絡數(shù)據(jù)融合研究具有重要的學術意義和工程價值。 本文深入地研究了典型的數(shù)據(jù)融合算法的原理、特點和性能指標，結合單個無線傳感器網(wǎng)絡感知數(shù)據(jù)的時間相關性和多個無線傳感器網(wǎng)絡感知數(shù)據(jù)的空間相關性，提出時空相關性數(shù)據(jù)融合算法的架構，設計了兩種高效的數(shù)據(jù)融合算法。 本文主要的研究內容有如下幾點： ①針對單個無線傳感器網(wǎng)絡節(jié)點感知數(shù)據(jù)冗余度高的問題，本文提出了一種基于時間相關性的數(shù)據(jù)融合算法。在分段一元線性回歸模型的基礎上，通過對無線傳感器網(wǎng)絡感知數(shù)據(jù)的時間序列進行分析，建立預測模型，并根據(jù)該模型的各個參數(shù)和給定的誤差，自適應地調整下一個采集時間，并動態(tài)地優(yōu)化回歸模型。仿真和實驗結果證明，針對不同的數(shù)據(jù)變化率，該算法均能減少數(shù)據(jù)采集量和傳輸量，滿足數(shù)據(jù)精度。 ②針對多個無線傳感器網(wǎng)絡節(jié)點感知數(shù)據(jù)冗余度高的問題，本文提出一種基于空間相關性的數(shù)據(jù)融合算法。該算法將監(jiān)測區(qū)域根據(jù)無線傳感器網(wǎng)絡感知數(shù)據(jù)的空間相關性的強弱程度劃分為若干個相關區(qū)域。每個相關區(qū)域根據(jù)無線傳感器網(wǎng)絡節(jié)點的剩余能量，選取一個代表節(jié)點。多個相關區(qū)域根據(jù)代表節(jié)點的剩余能量和到匯聚節(jié)點的距離，選取一個簇頭節(jié)點。代表節(jié)點將相關區(qū)域的感知數(shù)據(jù)融合后傳遞給簇頭節(jié)點。簇頭節(jié)點將數(shù)據(jù)通過無線方式轉發(fā)給匯聚節(jié)點。仿真和實驗結果證明，該算法使得代表節(jié)點和簇頭節(jié)點分布均勻，數(shù)據(jù)傳輸量小，數(shù)據(jù)精度高。該算法具有很好的能效性。
[Abstract]:As the underlying technology support of the Internet of things, wireless sensor network (WSN) collects and transmits various kinds of information in the monitoring area, which provides real-time and reliable data for military defense, telemedicine and environmental monitoring applications. It is the key information transmission technology in the Internet of things. However, wireless sensor network nodes are densely distributed and frequently sampled. The temporal and spatial correlation between perceptual data results in redundant data. The transmission of large amounts of redundant data brings great challenges to the limited energy, storage capacity and bandwidth of wireless sensor networks. Data fusion can effectively reduce redundant data and improve the efficiency and accuracy of data collection. Therefore, the research of wireless sensor network data fusion has important academic significance and engineering value. In this paper, the principle, characteristics and performance index of typical data fusion algorithm are deeply studied, and the temporal correlation of sensing data of single wireless sensor network and the spatial correlation of sensing data of multiple wireless sensor networks are combined. The framework of spatio-temporal correlation data fusion algorithm is proposed, and two efficient data fusion algorithms are designed. The main contents of this paper are as follows: In order to solve the problem of high data redundancy in a single wireless sensor network node, this paper proposes a data fusion algorithm based on time correlation. Based on the piecewise univariate linear regression model, the prediction model is established by analyzing the time series of sensor network perceptual data, and according to each parameter and given error of the model, The next acquisition time is adjusted adaptively and the regression model is dynamically optimized. The simulation and experimental results show that the algorithm can reduce the amount of data acquisition and transmission and meet the accuracy of the data. In order to solve the problem of high data redundancy of multiple nodes in wireless sensor networks, this paper proposes a data fusion algorithm based on spatial correlation. The algorithm divides the monitoring region into several related regions according to the spatial correlation degree of the sensor network perceptual data. According to the residual energy of wireless sensor network nodes, each related region selects a representative node. A cluster head node is selected according to the residual energy of the node and the distance from the node to the convergence node. The representative node fuses the perceptual data of the related region and passes it to the cluster head node. The cluster head node forwards the data to the convergent node by wireless way. The simulation and experimental results show that the algorithm makes the representative node and cluster head node distribute uniformly, the data transmission is small, and the data precision is high. The algorithm has good energy efficiency.
【學位授予單位】：重慶大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TN929.5;TP212.9

【參考文獻】

相關期刊論文 前10條

1 李珂;吳曉培;吳躍;賈惠波;;用于土壤濕度測量的傳感器網(wǎng)絡動態(tài)采樣策略[J];電子測量與儀器學報;2011年04期

2 王雷春;馬傳香;;傳感器網(wǎng)絡中一種基于一元線性回歸模型的空時數(shù)據(jù)壓縮算法[J];電子與信息學報;2010年03期

3 樂俊;張維明;肖衛(wèi)東;唐九陽;;一種能量高效和均衡的無線傳感器網(wǎng)絡分簇數(shù)據(jù)融合算法[J];國防科技大學學報;2012年06期

4 林蔚;祝啟龍;;無線傳感器網(wǎng)絡節(jié)能型數(shù)據(jù)融合算法[J];哈爾濱工程大學學報;2011年10期

5 康健;左憲章;唐力偉;張西紅;李浩;;無線傳感器網(wǎng)絡數(shù)據(jù)融合技術[J];計算機科學;2010年04期

6 王繼良;周四望;唐暉;;基于回歸的無線傳感器網(wǎng)絡數(shù)據(jù)壓縮方法[J];計算機工程;2011年23期

7 胡海峰;楊震;;無線傳感器網(wǎng)絡中基于空間相關性的分布式壓縮感知[J];南京郵電大學學報(自然科學版);2009年06期

8 趙繼軍;魏忠誠;李志華;劉昊;連彬;;無線傳感器網(wǎng)絡中多類型數(shù)據(jù)融合研究綜述[J];計算機應用研究;2012年08期

9 王玲;石為人;石欣;宋寧博;冉啟可;;基于時間相關性的無線傳感器網(wǎng)絡數(shù)據(jù)壓縮與優(yōu)化算法[J];計算機應用;2013年12期

10 武春濤;胡艷軍;;無線傳感器網(wǎng)絡LEACH算法的改進[J];計算機技術與發(fā)展;2009年03期

本文編號：1870096