彭鳳英 焦鍵
關(guān)鍵詞: 無(wú)線傳感網(wǎng)絡(luò); 定位; 移動(dòng)模型; 路徑規(guī)劃; 交替最小算法; 移動(dòng)錨節(jié)點(diǎn)
中圖分類號(hào): TN915.02?34; TP393 ? ? ? ? ? ? ? ? ? 文獻(xiàn)標(biāo)識(shí)碼: A ? ? ? ? ? ? ? ? ?文章編號(hào): 1004?373X(2019)03?0018?05
Abstract: The anchor node position plays an important role for accurate node localization in wireless sensor networks (WSNs). Therefore, a Gauss?Markov?based mobile anchor?localization (GM?MAL) algorithm is proposed in this paper. An adaptive mobile path planning of anchor node is proposed on the basis of Gauss?Markov mobility model. The strategies of velocity adjustment, perpendicular bisector, virtual repulsion and virtual attraction are used to plan the path in path planning stage. The non?convex optimization problem is converted into a bi?convex form, and solved with alternating minimization algorithm (AMA), which can acquire a shorter mobile path of anchor node. The experimental data shows that the virtual attraction strategy can improve the path planning accuracy, and cover more surveillance regions. In comparison with linear localization algorithm, the GM?MAL algorithm can improve the localization accuracy.
Keywords: wireless sensor network; localization; mobile model; path planning; alternating minimization algorithm; mobile anchor node
無(wú)線傳感網(wǎng)絡(luò)(Wireless Sensor Networks,WSNs)已廣泛應(yīng)用在各個(gè)領(lǐng)域中,如森林火災(zāi)檢測(cè)、戰(zhàn)場(chǎng)偵察、入侵檢測(cè)、目標(biāo)跟蹤以及健康康復(fù)等[1?2]。部署于WSNs內(nèi)的傳感節(jié)點(diǎn)實(shí)時(shí)感測(cè)環(huán)境數(shù)據(jù),然后再將數(shù)據(jù)傳輸至控制中心,進(jìn)而實(shí)現(xiàn)對(duì)環(huán)境區(qū)域的監(jiān)測(cè)。然而,感測(cè)數(shù)據(jù)必須附加較準(zhǔn)確的位置信息,一旦離開(kāi)了位置數(shù)據(jù),感測(cè)數(shù)據(jù)就失去意義。因此,節(jié)點(diǎn)定位成為WSNs的研究熱點(diǎn)之一[3]。
目前,現(xiàn)有的定位算法可分為測(cè)距和非測(cè)距兩類。其中非測(cè)距算法利用傳感節(jié)點(diǎn)與錨節(jié)點(diǎn)間的連通性,而測(cè)距算法是利用節(jié)點(diǎn)與節(jié)點(diǎn)間的距離或角度信息估計(jì)節(jié)點(diǎn)位置[4]。相比非測(cè)距定位算法,測(cè)距算法的定位精度較高。
在測(cè)距定位算法中,錨節(jié)點(diǎn)位置對(duì)定位精度有重要的影響。通常,錨節(jié)點(diǎn)數(shù)越多,定位精度越高。然而,增加錨節(jié)點(diǎn)數(shù)也會(huì)增加定位成本。此外,在靜態(tài)錨節(jié)點(diǎn)場(chǎng)景中,一旦對(duì)傳感節(jié)點(diǎn)定位后,這些錨節(jié)點(diǎn)就不再有價(jià)值,因?yàn)楣?jié)點(diǎn)是靜態(tài)的,網(wǎng)絡(luò)拓?fù)洳粫?huì)發(fā)生變化。因此,利用移動(dòng)的錨節(jié)點(diǎn)策略,并由移動(dòng)錨節(jié)點(diǎn)不斷廣播它的位置,在提高定位精度的同時(shí),降低定位成本[5]。
為此,本文提出基于高斯?Markov(Gauss?Markov,GM)移動(dòng)模型的移動(dòng)錨節(jié)點(diǎn)的節(jié)點(diǎn)定位算法(Gauss?Markov?based Mobile Anchor?localization,GM?MAL)。GM?MAL算法先規(guī)劃移動(dòng)路徑,錨節(jié)點(diǎn)依據(jù)此路徑移動(dòng),并實(shí)時(shí)調(diào)整移動(dòng)速度和方向,在移動(dòng)過(guò)程中,錨節(jié)點(diǎn)不斷廣播自己的位置信息,使得周圍的傳感節(jié)點(diǎn)能獲取與錨節(jié)點(diǎn)的距離信息,即測(cè)距。當(dāng)傳感節(jié)點(diǎn)獲取足夠多的測(cè)距值后,傳感節(jié)點(diǎn)便可估計(jì)自己的位置。
GM?MAL算法主要由錨節(jié)點(diǎn)移動(dòng)路徑規(guī)劃和定位兩部分組成。即先制訂錨節(jié)點(diǎn)的移動(dòng)路徑,然后,錨節(jié)點(diǎn)依據(jù)此路徑移動(dòng),使得傳感節(jié)點(diǎn)能夠獲取與錨節(jié)點(diǎn)的距離信息;隨后,傳感節(jié)點(diǎn)再依據(jù)距離信息估計(jì)自己的位置。
1.1 ?路徑規(guī)劃
從圖5可知,當(dāng)[σ>0.3]后,提出的GM?MAL定位算法的RMSE低于同類的線性定位算法,并且隨[σ]的增加,優(yōu)越性越發(fā)特出。例如,當(dāng)[σ]=2時(shí),GM?MAL算法的RMSE約為2.2,而線性定位算法的RMSE達(dá)到5.2。這些數(shù)據(jù)表明,提出的AMA定位算法能夠估計(jì)目標(biāo)位置。
從圖5可知,當(dāng)[σ>0.3]后,提出的GM?MAL定位算法的RMSE低于同類的線性定位算法,并且隨[σ]的增加,優(yōu)越性越發(fā)特出。例如,當(dāng)[σ]=2時(shí),GM?MAL算法的RMSE約為2.2,而線性定位算法的RMSE達(dá)到5.2。這些數(shù)據(jù)表明,提出的AMA定位算法能夠估計(jì)目標(biāo)位置。
針對(duì)無(wú)線傳感網(wǎng)絡(luò)的節(jié)點(diǎn)定位問(wèn)題,本文提出GM?MAL算法。GM?MAL算法針對(duì)移動(dòng)錨節(jié)點(diǎn),提出基于GM移動(dòng)模型的路徑規(guī)劃算法,并利用AMA算法估計(jì)節(jié)點(diǎn)位置。通過(guò)引入虛引力,對(duì)MAN的移動(dòng)方向進(jìn)行控制。此外,將定位問(wèn)題的非凸結(jié)構(gòu)轉(zhuǎn)化為雙凸形式,進(jìn)而利用AMA算法求解。實(shí)驗(yàn)數(shù)據(jù)表明,虛引力策略增加了定位節(jié)點(diǎn)數(shù),此外,應(yīng)用AMA算法提高了定位精度。
參考文獻(xiàn)
[1] YANG L, YANG L, HO K C. Moving target localization in multistatic sonar by differential delays and Doppler shifts [J]. IEEE signal processing letters, 2016, 23(9): 1160?1164.
[2] NOROOZI A, SEBT M A. Target localization in multistatic passive radar using SVD approach for eliminating the nuisance parameters [J]. IEEE transactions on aerospace and electronic systems, 2017, 53(4): 1660?1671.
[3] FISHLER E, HAIMOVICH A, BLUM R, et al. Spatial diversity in radars?models and detection performance [J]. IEEE tran?sactions on signal processing, 2006, 54(3): 823?838.
[4] GODRICH H, HAIMOVICH A M, BLUM R S. Target localisation techniques and tools for multiple?input multiple?output radar [J]. IET radar, sonar and navigation, 2015, 3(4): 314?327.
[5] YANG H, CHUN J. An improved algebraic solution for moving target localization in noncoherent MIMO radar systems [J]. IEEE transactions on signal processing, 2016, 64(1): 258?270.
[6] LI J, STOICA P. MIMO radar with colocated antennas [J]. IEEE signal processing magazine, 2017, 24(5): 106?114.
[7] NOROOZI A, SEBT M A. A new estimator for elliptic localization in distributed MIMO radar systems [C]// Proceedings of 2017 Iranian Conference on Electrical Engineering. Tehran: IEEE, 2017: 1615?1618.
[8] DIANAT M, TABAN M R, DIANAT J, et al. Target localization using least squares estimation for MIMO radars with widely separated antennas [J]. IEEE transactions on aerospace electronic systems, 2013, 49(4): 2730?2741.
[9] EINEMO M, SO H C. Weighted least squares algorithm for target localization in distributed MIMO radar [J]. Signal proces?sing, 2015,115(4): 144?150.
[10] NOROOZI A, SEBT M A. Target localization from bistatic range measurements in multi?transmitter multi?receiver passive radar [J]. IEEE signal processing letters, 2015, 22(12): 2445?2449.
[11] NOROOZI A, SEBT M A. Weighted least squares target location estimation in multi?transmitter multi?receiver passive radar using bistatic range measurements [J]. IET radar, sonar and navigation, 2016, 10(6): 1088?1097.
[12] GODRICH H, HAIMOVICH A M, BLUM R S. Target loca?lization accuracy gain in MIMO radar?based systems [J]. IEEE transactions on information theory, 2010, 56(6): 2783?2803.