Distributed Principal Component Analysis for Wireless Sensor

The Principal Component Analysis (PCA) is a data dimensionality reduction tech- nique well-suited for processing data from sensor networks. It can be applied to tasks like compression, event detection, and event recognition. This technique is based on a linear trans- form where the sensor measurements are projected on a set of principal components. When sensor measurements are correlated, a small set of principal components can explain most of the measurements variability. This allows to significantly decrease the amount of radio communication and of energy consumption. In this paper, we show that the power iteration method can be distributed in a sensor network in order to compute an approximation of the principal components. The proposed implementation relies on an aggregation service, which has recently been shown to provide a suitable framework for distributing the computation of a linear transform within a sensor network. We also extend this previous work by providing a detailed analysis of the computational, memory, and communication costs involved. A com- pression experiment involving real data validates the algorithm and illustrates the tradeoffs between accuracy and communication costs.

[1]  Franklin T. Luk,et al.  Principal Component Analysis for Distributed Data Sets with Updating , 2005, APPT.

[2]  David E. Culler,et al.  Telos: enabling ultra-low power wireless research , 2005, IPSN 2005. Fourth International Symposium on Information Processing in Sensor Networks, 2005..

[3]  Yu Hen Hu,et al.  Vehicle classification in distributed sensor networks , 2004, J. Parallel Distributed Comput..

[4]  Márk Jelasity,et al.  Asynchronous Distributed Power Iteration with Gossip-Based Normalization , 2007, Euro-Par.

[5]  Anna Scaglione,et al.  Precoding and Decoding Paradigms for Distributed Vector Data Compression , 2007, IEEE Transactions on Signal Processing.

[6]  Andrew R. Webb,et al.  Statistical Pattern Recognition , 1999 .

[7]  I. Jolliffe.  Principal Component Analysis , 2002 .

[8]  Kannan Ramchandran,et al.  Distributed source coding using syndromes (DISCUS): design and construction , 2003, IEEE Trans. Inf. Theory.

[9]  Ivan Stojmenović,et al.  Handbook of Sensor Networks: Algorithms and Architectures , 2005, Handbook of Sensor Networks.

[10]  E. Oja,et al.  Independent Component Analysis , 2001 .

[11]  Michael Gastpar,et al.  The Distributed Karhunen–Loève Transform , 2006, IEEE Transactions on Information Theory.

[12]  Leonidas J. Guibas,et al.  Wireless sensor networks - an information processing approach , 2004, The Morgan Kaufmann series in networking.

[13]  Deborah Estrin,et al.  Impact of network density on data aggregation in wireless sensor networks , 2002, Proceedings 22nd International Conference on Distributed Computing Systems.

[14]  Deborah Estrin,et al.  The impact of data aggregation in wireless sensor networks , 2002, Proceedings 22nd International Conference on Distributed Computing Systems Workshops.

[15]  Hillol Kargupta,et al.  Collective Principal Component Analysis from Distributed, Heterogeneous Data , 2000, PKDD.

[16]  David G. Stork,et al.  Pattern Classification , 1973 .

[17]  Yong Yao,et al.  The cougar approach to in-network query processing in sensor networks , 2002, SGMD.

[18]  Gianluca Bontempi,et al.  Unsupervised and supervised compression with principal component analysis in wireless sensor networks , 2007 .

[19]  Jack Dongarra,et al.  Templates for the Solution of Algebraic Eigenvalue Problems , 2000, Software, environments, tools.

[20]  Gianluca Bontempi,et al.  An adaptive modular approach to the mining of sensor network data , 2005 .

[21]  Geert Molenberghs,et al.  Transformation of non positive semidefinite correlation matrices , 1993 .

[22]  Yunfeng Zhang,et al.  Interactive sensor network data retrieval and management using principal components analysis transform , 2006 .

[23]  Wei Hong,et al.  Proceedings of the 5th Symposium on Operating Systems Design and Implementation Tag: a Tiny Aggregation Service for Ad-hoc Sensor Networks , 2022 .

[24]  Mani Srivastava,et al.  Energy-aware wireless microsensor networks , 2002, IEEE Signal Process. Mag..

[25]  Christophe Diot,et al.  Diagnosing network-wide traffic anomalies , 2004, SIGCOMM.

[26]  Wei Hong,et al.  TinyDB: an acquisitional query processing system for sensor networks , 2005, TODS.

[27]  Feng Zhao,et al.  Collaborative signal and information processing in microsensor networks , 2002, IEEE Signal Processing Magazine.

[28]  Ling Huang,et al.  In-Network PCA and Anomaly Detection , 2006, NIPS.

[29]  R.G. Baraniuk,et al.  Distributed Compressed Sensing of Jointly Sparse Signals , 2005, Conference Record of the Thirty-Ninth Asilomar Conference onSignals, Systems and Computers, 2005..

[30]  R. Wattenhofer,et al.  Dozer: Ultra-Low Power Data Gathering in Sensor Networks , 2007, 2007 6th International Symposium on Information Processing in Sensor Networks.

[31]  Zixiang Xiong,et al.  Distributed source coding for sensor networks , 2004, IEEE Signal Processing Magazine.

[32]  Ian F. Akyildiz,et al.  Wireless sensor networks: a survey , 2002, Comput. Networks.

[33]  Konstantinos Psounis,et al.  Modeling spatially-correlated sensor network data , 2004, 2004 First Annual IEEE Communications Society Conference on Sensor and Ad Hoc Communications and Networks, 2004. IEEE SECON 2004..

[34]  David L Donoho,et al.  Compressed sensing , 2006, IEEE Transactions on Information Theory.

[35]  Deborah Estrin,et al.  Building efficient wireless sensor networks with low-level naming , 2001, SOSP.

[36]  Gianluca Bontempi,et al.  New Routes from Minimal Approximation Error to Principal Components , 2008, Neural Processing Letters.

[37]  Mohamed F. Younis,et al.  A survey on routing protocols for wireless sensor networks , 2005, Ad Hoc Networks.

[38]  Gene H. Golub,et al.  Matrix computations , 1983 .