Data Recovery in Wireless Sensor Networks Based on Low Rank and Short-Term Stability

The most effective way to reduce the energy consumption of energy-limited wireless sensor networks is to reduce the amount of data collected. However, this will increase the difficulty of data recovery. At the same time, most of the data collection and recovery algorithms based on matrix completion...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2019, Vol.7, p.166796-166802
Main Authors: Liu, Xiaochao, Sun, Guiling, Li, Zhouzhou, Zheng, Bowen
Format: Article
Language:English
Subjects:
Algorithms
Data collection
Data recovery
Energy consumption
Energy limitation
low rank matrix completion
Mathematical analysis
Mathematical model
Mathematical models
Matrix decomposition
Matrix methods
Optimization
Regularization
Sensors
short-term stability
Stability
Stability analysis
Temperature sensors
Thermal stability
Wireless networks
Wireless sensor networks
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The most effective way to reduce the energy consumption of energy-limited wireless sensor networks is to reduce the amount of data collected. However, this will increase the difficulty of data recovery. At the same time, most of the data collection and recovery algorithms based on matrix completion are optimized by matrix decomposition. Therefore, the mathematical model and corresponding optimization algorithm will be more complicated and take a lot of running time. In order to deal with the above problems, we propose a data recovery method based on low rank and short-term stability in wireless sensor networks in this paper. The low-rank and short-term stability features of the sensor data are fused in the nuclear norm regularization minimization model, and the proposed method is used to optimize the model. The simulation results show that the mathematical model and data recovery method constructed in this paper outperform the state-of-the-art methods in terms of recovery performance and reconstruction accuracy.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2019.2953794