基于微粒群优化LSSVM的室内指纹定位算法
|
摘要
为了降低利用最小二乘支持向量机(LSSVM)定位过程中参数选取对定位精度的影响,提出一种基于微粒群进行参数优化的室内指纹定位算法。该算法通过离线采集的RSSI数据训练最小二乘支持向量机,利用微粒群算法寻找并确定LSSVM全局最优参数,获得基于位置指纹的LSSVM定位模型。仿真结果表明,相对于传统LSSVM定位,PSO-LSSVM有效提高了定位准确度,并能在小样本情况下保持良好的定位精度。
The application of least square support vector machine(LSSVM)for indoor localization suffers from instability of accuracy as the parameters are chosen based on experience.In order to reduce the influence,we use particle swarm optimization to improve positioning accuracy.We obtain the training samples from offline to the LSSVM,and find the optimal LSSVM parameters in PSO.The obtained optimal parameters are then used in the positioning.Simulations show that compared with the original LSSVM position method,PSO-LSSVM method effectively improves localization accuracy and maintains good positioning accuracy in small sample.
The application of least square support vector machine(LSSVM)for indoor localization suffers from instability of accuracy as the parameters are chosen based on experience.In order to reduce the influence,we use particle swarm optimization to improve positioning accuracy.We obtain the training samples from offline to the LSSVM,and find the optimal LSSVM parameters in PSO.The obtained optimal parameters are then used in the positioning.Simulations show that compared with the original LSSVM position method,PSO-LSSVM method effectively improves localization accuracy and maintains good positioning accuracy in small sample.
引文
[1]冯亚超,贺康,杨红丽,等.一种无线传感器网络数据收集协议的研究与优化[J].传感技术学报,2014,27(3):355-360.
[2]AKYILDIZ W S,SANAKARASUBRAMANIAM Y,CAYICI E,et al.Wireless sensor networks:a survey[J].Computer Networks,2002,38(4):393-422.
[3]HAO Z,WANG S.Survey on localization algorithms for wireless sensor networks[J].Journal of Huazhong University of Science and Technology(Natural Science Edition),2008,36(1):224-227.
[4]LI X,ZHOU F,DU J.LDTS:a lightweight and dependable trust system for clustered wireless sensor network[J].IEEE Transactions on Information Forensics and Security,2013,99:1.
[5]CHONG C Y,KUMAR S P.Sensor networks:evolution,opportunities,and chanllenges[J].Proceedings of the IEEE,2003.
[6]汪炀.基于无线传感器网络定位技术研究[D].合肥:中国科技大学,2007.
[7]朱晓娟,孟祥瑞.基于加权质心的无线传感器网络动态节点定位算法[J].计算机工程与科学,2011,33(11):15-19.
[8]JAVIER R,VALENTíN B,ESCUDERO C J.Architecture for multi-technology real-time location systems[J].Sensors,2013,13(2):2220-2253.
[9]徐玉滨,邓志安,马琳.基于核直接判别分析和支持向量机回归的WLAN室内定位算法[J].电子与信息学报,2011,33(4):13-17.
[10]MAHFOUZ S,MOURAD-CHEHADE F,HONEINE P,et al.Target tracking using machine learning and Kalman filter in wireless sensor networks[J].Sensors Journal IEEE,2014,14(10):3715-3725.
[11]邱爱昆.基于机器学习的WiFi高精度定位研究[D].无锡:江南大学,2016.
[12]桑楠,袁兴中,周瑞.基于SVM分类和回归的WiFi室内定位方法[J].计算机应用研究,2014,31(6):52-53.
[13]赵银龙,安胜彪.基于核化K-means和SVM分类回归的WiFi室内定位算法[J].信息技术,2018(1):26-27.
[14]杨晋生,林振军.一种改进支持向量机的无线定位方法[J].应用科学学报,2017,35(6):685-692.
[15]FARJOW W,CHEHRI A,MOUFTAH H,et al.Support vector machines for indoor sensor localization[C].Wireless Communications&Networking Conference.IEEE,2011.
[16]CORTES C,VAPNIK V.Support-vector networks[J].Machine Learning,1995,20(3):273-297.
[17]张会清,石晓伟,邓贵华.基于BP神经网络和泰勒级数的室内定位算法研究[J].计算机科学,2012,39(4):32-35.
[18]孙斌,姚海涛.基于PSO优化LSSVM的短期风速预测[J].电力系统保护与控制,2012,40(5):85-89.
[19]马立新,王宏宇.基于非支配解的多目标粒子群无功优化[J].控制工程,2014,21(5):748-752.
[20]申静涛.基于RSSI的对数距离路径损耗模型研究[J].电子质量,2013(12):15-17.
[21]MEDEISIS A,KAJACKAS A.On the use of the universal Okumura-Hata propagation prediction model in rural areas[C].IEEE Vehicular Technology Conference.IEEE Xplore,2000.
[2]AKYILDIZ W S,SANAKARASUBRAMANIAM Y,CAYICI E,et al.Wireless sensor networks:a survey[J].Computer Networks,2002,38(4):393-422.
[3]HAO Z,WANG S.Survey on localization algorithms for wireless sensor networks[J].Journal of Huazhong University of Science and Technology(Natural Science Edition),2008,36(1):224-227.
[4]LI X,ZHOU F,DU J.LDTS:a lightweight and dependable trust system for clustered wireless sensor network[J].IEEE Transactions on Information Forensics and Security,2013,99:1.
[5]CHONG C Y,KUMAR S P.Sensor networks:evolution,opportunities,and chanllenges[J].Proceedings of the IEEE,2003.
[6]汪炀.基于无线传感器网络定位技术研究[D].合肥:中国科技大学,2007.
[7]朱晓娟,孟祥瑞.基于加权质心的无线传感器网络动态节点定位算法[J].计算机工程与科学,2011,33(11):15-19.
[8]JAVIER R,VALENTíN B,ESCUDERO C J.Architecture for multi-technology real-time location systems[J].Sensors,2013,13(2):2220-2253.
[9]徐玉滨,邓志安,马琳.基于核直接判别分析和支持向量机回归的WLAN室内定位算法[J].电子与信息学报,2011,33(4):13-17.
[10]MAHFOUZ S,MOURAD-CHEHADE F,HONEINE P,et al.Target tracking using machine learning and Kalman filter in wireless sensor networks[J].Sensors Journal IEEE,2014,14(10):3715-3725.
[11]邱爱昆.基于机器学习的WiFi高精度定位研究[D].无锡:江南大学,2016.
[12]桑楠,袁兴中,周瑞.基于SVM分类和回归的WiFi室内定位方法[J].计算机应用研究,2014,31(6):52-53.
[13]赵银龙,安胜彪.基于核化K-means和SVM分类回归的WiFi室内定位算法[J].信息技术,2018(1):26-27.
[14]杨晋生,林振军.一种改进支持向量机的无线定位方法[J].应用科学学报,2017,35(6):685-692.
[15]FARJOW W,CHEHRI A,MOUFTAH H,et al.Support vector machines for indoor sensor localization[C].Wireless Communications&Networking Conference.IEEE,2011.
[16]CORTES C,VAPNIK V.Support-vector networks[J].Machine Learning,1995,20(3):273-297.
[17]张会清,石晓伟,邓贵华.基于BP神经网络和泰勒级数的室内定位算法研究[J].计算机科学,2012,39(4):32-35.
[18]孙斌,姚海涛.基于PSO优化LSSVM的短期风速预测[J].电力系统保护与控制,2012,40(5):85-89.
[19]马立新,王宏宇.基于非支配解的多目标粒子群无功优化[J].控制工程,2014,21(5):748-752.
[20]申静涛.基于RSSI的对数距离路径损耗模型研究[J].电子质量,2013(12):15-17.
[21]MEDEISIS A,KAJACKAS A.On the use of the universal Okumura-Hata propagation prediction model in rural areas[C].IEEE Vehicular Technology Conference.IEEE Xplore,2000.