A Distributed Training Method for Sparse Kernel Machine over WSN

doi:10.13190/j.jbupt.2016.03.014

JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM ›› 2016, Vol. 39 ›› Issue (3): 80-84.doi: 10.13190/j.jbupt.2016.03.014

• Papers • Previous Articles Next Articles

A Distributed Training Method for Sparse Kernel Machine over WSN

JI Xin-rong^1,2, HOU Cui-qin¹, HOU Yi-bin¹

1. Beijing Engineering Research Center for IOT Software and Systems, Beijing University of Technology, Beijing 100124, China;
2. School of Information and Electrical Engineering, Hebei University of Engineering, Hebei Handan 056038, China

Received:2015-09-09 Online:2016-06-28 Published:2016-06-27

Abstract

Abstract:

In wireless sensor network (WSN), the centralized learning method by transmitting all training samples scattered across different sensor nodes to a centralized data center to train classifier will significantly increase the communication cost. To decrease the communication cost in transmitting training samples, a distributed learning method for kernel minimum squared error (KMSE) by incorporating L1 regularized term was studied, which just relies on in-network processing between single-hop neighboring nodes. Each node obtains its local optimum sparse model by constructing the optimization problem of L1 regularized KMSE based on its local training samples and solving it using parallel projections and alternating the direction method of multipliers, then a consistent model is achieved on all nodes by using the global average consensus algorithm. For carrying out this method,a new distributed training algorithm for L1-regularized kernel minimum squared error based on parallel projections (L1-DKMSE-PP) was proposed. Simulations show that L1-DKMSE-PP can obtain almost the same prediction accuracy as that of the centralized counterpart and a sparser model, and more importantly, it can significantly reduce the communication cost.

Key words: wireless sensor network, kernel machine, distributed learning, L1-regularized, parallel projection, alternating direction method of multipliers

CLC Number:

TP181

JI Xin-rong, HOU Cui-qin, HOU Yi-bin. A Distributed Training Method for Sparse Kernel Machine over WSN[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2016, 39(3): 80-84.

[1]	YU Xiuwu, SHANG Yadong, LIU Yong. WSN Localization Algorithm Based on Temporary Best and Worst Centroid Opposite Cross Mutation and Sparrow Optimization [J]. Journal of Beijing University of Posts and Telecommunications, 2023, 46(1): 90-96.
[2]	LI Xilan, DUAN Jizhong. An Improved Sensitivity Encoding Reconstruction Algorithm Based on Sparse Transform Learning [J]. Journal of Beijing University of Posts and Telecommunications, 2022, 45(5): 97-102.
[3]	YU Xiuwu, QIN Xiaokun, LIU Yong. WSN Routing Algorithm Based on Firefly Algorithm to Optimize FCM [J]. Journal of Beijing University of Posts and Telecommunications, 2022, 45(2): 50-56.
[4]	YU Xiu-wu, HUANG Lu-ping, LIU Yong, YU Hao, LI Pei. Three-Dimensional DV-Hop Location Algorithm Based on Genetic-Tabu Search Optimization in WSN [J]. Journal of Beijing University of Posts and Telecommunications, 2021, 44(4): 75-81.
[5]	LIU Lin-lan, XIAO Ting-zhong, XIA Yang, SHU Jian. Link Quality Estimation Based on Extremely Fast Decision Tree [J]. Journal of Beijing University of Posts and Telecommunications, 2021, 44(3): 125-130.
[6]	YU Xiu-wu, LI Ying, LIU Yong, XIAO Ren-rong, YU Hao. Wireless Localization Algorithm of Adaptive Levy Whale Based on Mapping Curve [J]. Journal of Beijing University of Posts and Telecommunications, 2020, 43(5): 125-129,136.
[7]	YU Xiu-wu, ZHANG Ke, LIU Yong. Localization Algorithm Based on Penalty Function and Water Wave Optimization for WSN [J]. Journal of Beijing University of Posts and Telecommunications, 2020, 43(4): 106-112.
[8]	HU Zhong-dong, WANG Jun-ling, WANG Zhen-dong, ZENG Ting. Node Localization Algorithm of Wireless Sensor Networks for Complex Terrain in Mountain Areas [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2019, 42(5): 113-118.
[9]	PAN Lei-na, TAO Yang, XU Xiang-yang, WANG Jin. A Secure Clustering Routing Protocol Based on Trust and Balancing Energy Consumption [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2019, 42(3): 29-36.
[10]	QIAN Kai-guo, PU Chun-fen, WANG Yu-jian, SHEN Shi-kai. An Iteration Localization Algorithm in Wireless Sensor Network Based on Intersection Degree Ratio [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2019, 42(3): 98-105.
[11]	MA Jun-yan, ZHANG Te, WANG Jin. Task Transition Probability Based Anomaly Detection Method for Sensor Nodes [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2019, 42(3): 37-42.
[12]	YU Xiu-wu, XIAO Ren-rong, LIU Yong, GUO Qian, YU Hao. Signal Strength Localization Algorithm Based on Rank Filter and Fibonacci Tree [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2019, 42(2): 7-12.
[13]	LIU Zhe. Adaptive Data Sampling Rate Adjustment Based on Large Deviation Theory in Wireless Sensor Networks [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2018, 41(6): 129-136.
[14]	YU Xiu-wu, LIU Qin, LI Xiang-yang, ZHANG Ke, XIAO Ren-rong. Information Fusion Algorithm Based on Improved Ant Colony Optimization BP Neural Network in WSN [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2018, 41(4): 91-96.
[15]	XUE Han-han, WANG Bai, ZHANG Lei, HUANG Hai. A Self-Organized Clustering Algorithm Based on Computation and Transmission Node for WSN [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2018, 41(3): 101-106.

A Distributed Training Method for Sparse Kernel Machine over WSN

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

Comments