李澤軍等
摘要:針對具有障礙物的復(fù)雜3D平面定位誤差大以及能耗高的問題,如何進行未知節(jié)點的定位,提出了一種具有障礙物的不規(guī)則復(fù)雜3D平面定位策略,該策略將復(fù)雜三維平面劃分為水平邏輯層并標記節(jié)點的所屬層,然后在具有障礙物的水平層進行子區(qū)域的劃分及合并,最后利用大氣壓傳感器測距進行計算未知節(jié)點的坐標從而得到3D復(fù)雜平面的定位.通過與最新SV定位機制及COLA定位機制在定位誤差及能耗進行綜合仿真實驗對比,本文的定位策略3DCCD平均定位誤差比SV少了5%,比COLA的平均定位誤差少8%以上,定位精度上比SV和COLA提高了6.5%以上,及在能耗上比SV和COLA減少了2.7%左右.因此該定位策略(3DCCD)在解決復(fù)雜不規(guī)則圖形的不確定節(jié)點具有較大的優(yōu)越性.
關(guān)鍵詞:定位誤差;邏輯層;子區(qū)域;定位機制
中圖分類號:TP39 文獻標識碼:A
The Study of Positioning Strategy of Irregular 3D Plane
in Sensor Networks
LI Zejun1,2, CHEN Min1,2
(1.College of Information Science and Engineering, Hunan Univ,Changsha, Hunan 410082, China;
2.School of Computer and Information Science,Hunan Institute of Technology,Hengyang , Hunan412002,China)
Abstract:Considering the large positioning error and high energy consumption concerning irregular and complex 3D plane with obstacles, this paper put forward a new strategy to locate the unknown nodes. This policy divides the complex threedimensional plane into different logic layers and labels each layer node it belongs to. Then, subregions are divided and merged in horizontal layers. Finally, the positioning of 3D complex plane is possible by calculating the location coordinate of unknown nodes after atmospheric pressure sensor ranging. Compared with SV and COLA, the 3DCCD positioning strategy in the average positioning error is 5% less than SV and 8% less than COLA. Its positioning precision increases more than 6.5%, and the energy consumption reduces 2.7% than SV and COLA. The 3DCCD has great advantages in terms of positioning error, accuracy and energy.
Key words:positioning error;logic layer;subdomain ;positioning mechanism
復(fù)雜環(huán)境下的定位機制大多數(shù)都是基于2D網(wǎng)絡(luò)拓撲環(huán)境,導(dǎo)致節(jié)點路徑估算產(chǎn)生較大的誤差以及誤差難以控制、傳感器節(jié)點能耗大等諸多因素.如:ACDL經(jīng)典方法采用2D定位機制解決復(fù)雜網(wǎng)絡(luò)定位問題,該方法設(shè)定一臨界點,傳感器節(jié)點利用鄰居節(jié)點的通信來判定凹凸表面,并將復(fù)雜表面劃分為幾個子區(qū)域,節(jié)點通過MDS建立節(jié)點的相對位置.最后利用相對位置建立全局2D平面圖.這樣方法過度臨界值的選擇,當(dāng)復(fù)雜不平表面無臨界值時,該策略將導(dǎo)致失敗,另外該定位方法對邊界噪音比較敏感.目前,學(xué)者們嘗試從2D平面擴展到3D平面,其主要采用的定位方法有兩種,一是通過將3D復(fù)雜平面映射到2D平面來定位其坐標[1,2].二是采用特殊裝置來測量凹凸平面的高度值和拋錨節(jié)點的距離[3].上述方法對于簡單的3D表面定位具有較高的精確度,但對于復(fù)雜表面以及具有較大凸高時,其定位誤差明顯增大和定位不明確.而SV(SingleValue)根據(jù)不同節(jié)點映射到不同2D平面上且平面的點具有唯一性.然后根據(jù)二維平面來定位三維平面.一定程度上解決了較簡單的3D平面節(jié)點重疊問題[4].但SV的代價較高,對于不規(guī)則三維平面也將導(dǎo)致節(jié)點重疊現(xiàn)象.COLA定位機制[5]采用移動錨節(jié)點來定位三維平面,該方法的精確度較高,但對于復(fù)雜3D的平面,由于錨節(jié)點的不斷移動產(chǎn)生大量的誤差值,難以適用大規(guī)模3D平面定位.
3.3合并子區(qū)域及構(gòu)建層區(qū)域
在進行子區(qū)域的建立時采用廣播的方式進行信息傳遞,導(dǎo)致在層區(qū)域內(nèi)形成若干個子區(qū)域并且子區(qū)域可能產(chǎn)生重疊現(xiàn)象.若節(jié)點p接收多個構(gòu)建子區(qū)域信息時,以第一時間收到的信息作為標記加入到該子區(qū)域中,其余消息將被丟棄.否則若子區(qū)域標記不相同則合并兩個子區(qū)域直到合并消息不存在.
復(fù)雜凹凸平面的分解與子區(qū)域的合并算法描述如下:
4復(fù)雜平面定位與計算
在復(fù)雜3D平面定位計算中主要考慮模型構(gòu)建差生的誤差以及定位方法和過程.傳感器節(jié)點在進行通信時其物理層的傳播模式可以任意,其公式為[7,8]:
圖11表示3DCCD算法的開銷在不同測量誤差時都要比SV及COLA算法的開銷要小,這是由于3DCCD采用的迭代次數(shù)少的緣故.
6結(jié)束語
本文研究復(fù)雜3D平面的定位機制,該策略首先將復(fù)雜三維平面進行邏輯水平分層,然后在水平層進行子區(qū)域的劃分及合并處理,最后用大氣壓傳感器測距進行計算未知節(jié)點的坐標從而得到3D復(fù)雜平面的定位.該思路為不規(guī)則復(fù)雜三維平面提供了新的思路.該策略尤其適用于不確定節(jié)點的定位.
參考文獻
ZHAO Y,WU H,JIN M,et al.Localization in 3D surface sensor networks: Challenges and solutions[C]//The 31st Annual IEEE International Conference on Computer Communications (IEEE INFOCOM 2012).Orlando, Florida,USA,2012:55-63.
〖LL〗[2]LIU Q,REN P,ZHOU Z. Threedimensional accurate positioning algorithm based on wireless sensor networks[J].Journal of Computers, 2011, 6(12): 2582-2589.
[3]LIU C,JIANG H,ZENG L D.Unitary matrix pencil algorithm for rangebased 3d localization of wireless sensor network nodes[J]. Journal of Networks, 2012, 7(9):1384-1390.
[4]趙德鑫,李婷,黃芝平,等.無人潛航器舷側(cè)陣聲吶匹配場被動定位方法研究[J].湖南大學(xué)學(xué)報:自然科學(xué)版,2013,40(8):76-82.
ZHAO Dexin,LI Ting,HUANG Zhiping,et al.Matchedfield source localization using the flank array of unmanned under water vehicle[J].Journal of Hunan University:Natural Sciences, 2013,40(8):76-82. (In Chinese)
[5]STOLERU R,HE T,MATHIHARAN S S,et al.Asymmetric eventdriven node localization in wireless sensor networks[J]. Parallel and Distributed Systems, IEEE Transactions on, 2012,23(4): 634-642.
[6]楊高波,吳瀟,張兆揚,等.基于過渡像素的視頻圖像文本檢測與定位[J]. 湖南大學(xué)學(xué)報:自然科學(xué)版,2011,38(6):69-74.
YANG Gaobo,WU Xiao,ZHANG Zhaoyang,et al.A transition pixels based text detection and localization for video images[J]. Journal of Hunan University:Natural Sciences, 2011,38(6):69-74.(In Chinese)
[7]CUI H,WANG Y.Fourmobilebeacon assisted localization in threedimensional wireless sensor networks[J]. Computers & Electrical Engineering, 2012, 38(3): 652-661.
[8]WANG R J,QIN Z G.A weighted 3D localization algorithm based on partial hopsize in wireless sensor network[J]. International Journal of Advancementsin Computing Technology,2012, 4(9): 504-513.
[9]WANG Y,LIU Y,GUO Z.Threedimensional ocean sensor networks: A survey[J]. Journal of Ocean University of China, 2012, 11(4): 436-450.
[10]王小平,羅軍,沈昌祥.無線傳感器網(wǎng)絡(luò)定位理論和算法[J].計算機研究與發(fā)展, 2011,48(3):353-363.
WANG Xiaoping,LUO Jun,SHEN Changxiang. Theory and algorith mson localization in wireless sensor networks[J]. Journal of Computer Research and Development, 2011,48(3):353-363. (In Chinese)