Home > Research > Publications & Outputs > Recovering missing values from corrupted spatio...

Links

Text available via DOI:

View graph of relations

Recovering missing values from corrupted spatio-temporal sensory data via robust low-rank tensor completion

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published
  • Wenjie Ruan
  • Peipei Xu
  • Quan Z. Sheng
  • Nickolas J.G. Falkner
  • Xue Li
  • Wei Emma Zhang
Close
Publication date1/04/2017
Host publicationDatabase Systems for Advanced Applications
EditorsS. Candan, L. Chen, T. Pedersen, L. Chang, W. Hua
Place of PublicationCham
PublisherSpringer
Pages607-622
Number of pages16
ISBN (Electronic)9783319557533
ISBN (Print)9783319557526
<mark>Original language</mark>English
Event22nd International Conference on Database Systems for Advanced Applications, DASFAA 2017 - Suzhou, China
Duration: 27/03/201730/03/2017

Conference

Conference22nd International Conference on Database Systems for Advanced Applications, DASFAA 2017
Country/TerritoryChina
City Suzhou
Period27/03/1730/03/17

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer
Volume10177
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference22nd International Conference on Database Systems for Advanced Applications, DASFAA 2017
Country/TerritoryChina
City Suzhou
Period27/03/1730/03/17

Abstract

With the booming of the Internet of Things, tremendous amount of sensors have been installed in different geographic locations, generating massive sensory data with both time-stamps and geo-tags. Such type of data usually have shown complex spatio-temporal correlation and are easily missing in practice due to communication failure or data corruption. In this paper, we aim to tackle the challenge-how to accurately and efficiently recover the missing values for corrupted spatiotemporal sensory data. Specifically, we first formulate such sensor data as a high-dimensional tensor that can naturally preserve sensors’ both geographical and time information, thus we call spatio-temporal Tensor. Then we model the sensor data recovery as a low-rank robust tensor completion problem by exploiting its latent low-rank structure and sparse noise property. To solve this optimization problem, we design a highly efficient optimization method that combines the alternating direction method of multipliers and accelerated proximal gradient to minimize the tensor’s convex surrogate and noise’s ℓ1-norm. In addition to testing our method by a synthetic dataset, we also use passive RFID (radiofrequency identification) sensors to build a real-world sensor-array testbed, which generates overall 115,200 sensor readings for model evaluation. The experimental results demonstrate the accuracy and robustness of our approach.