.png){kind=icon}
返回
Compressive Sensing Based Distributed Data Storage for Mobile Crowdsensing
.png){kind=icon}
PRE
AI
DOI：10.1145/3498321.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Mobile crowdsensing systems typically operate centralized cloud storage management, and the environment data sensed by the participants are usually uploaded to certain central cloud servers. Instead, this article addresses the decentralized data storage problem in scenarioswhere cloud servers or network infrastructures do not work as expected and the sensing data have to be temporarily stored on the mobile devices carried by the participants. Considering that the sensing data are generally correlated, this article investigates a compressive distributed storage scheme for mobile crowdsensing. We notice a key observation: when a participant has a random walk in the target sensing area, his walking/sensing process can be considered as a random sampling for the entire area, although the activity of the participant may only have a local scope. We then propose an encoding algorithm based on compressive sensing theory. Each participant encodes the sensing data in their local trajectory, but the encoded CS measurement is capable of roughly reflecting the entire information of the whole area. While a participant stores a blurred global image of the target sensing area, the entire data can then be collaboratively stored by a certain number of participants. We further present a period-based data recovery algorithm to exploit the inter-period correlations, improving the recovery accuracy. Experimental results using real environmental data demonstrate the performance of the proposed compressive storage scheme. The test datasets and our source codes are available at https://github.com/siwangzhou/MCS-Storage.
Keyword:
Compressive sensing
distributed storage
mobile crowdsensing
wireless sensor network
期刊
IF:
4.7
论文数:
999
被引数:
2.0K
机构
引用论文
.png){kind=icon}
分享收藏
分享收藏Small-block sensing and larger-block recovery in block-based compressive sensing of images基于块的图像压缩感知中的小块感知和大块恢复
PREAI
分享收藏
查看更多内容