عنوان

When compressive sensing meets mobile crowdsensing /

9789811377761

9811377766

9789811377754

9811377758

When compressive sensing meets mobile crowdsensing /

[Book]

Linghe Kong, Bowen Wang and Guihai Chen.

Singapore :

Springer,

[2019]

©2019

1 online resource

Includes bibliographical references.

Intro; Preface; Acknowledgements; Contents; 1 Introduction; 1.1 Mobile Crowdsensing Overview; 1.2 Compressive Sensing Overview; 1.3 Organization; References; 2 Mobile Crowdsensing; 2.1 Background; 2.2 Data Quality Problem in Mobile Crowdsensing; 2.3 Existing Data Quality Solutions in Mobile Crowdsensing; 2.3.1 Quality-Aware Incentive Mechanisms; 2.3.2 Quality-Driven Participator Selection Mechanisms; 2.3.3 Authentication Mechanisms; 2.3.4 Quality-Driven Task Allocation Mechanisms; 2.3.5 Lightweight Preprocessing Strategies; 2.3.6 Outlier Detection and Data Correction; 2.4 Summary; References

3 Compressive Sensing3.1 Background; 3.2 Conventional Compressive Sensing; 3.2.1 Sparsity and Compressible Signals; 3.2.2 Sampling; 3.2.3 Reconstruction; 3.3 Compressive Sensing for Matrix Completion; 3.3.1 From Vectors to Matrices; 3.3.2 Sparsity in Matrices; 3.3.3 Sampling in Matrices; 3.3.4 Reconstruction in Matrices; 3.4 Summary; References; 4 Basic Compressive Sensing for Data Reconstruction; 4.1 Background; 4.2 Problem Statement; 4.2.1 Missing Data Problem in Mobile Crowdsensing; 4.2.2 Sparse and Uneven Data Distribution; 4.3 Basic Compressive Sensing Algorithm

4.3.1 Revealing Hidden Structure4.3.2 Missing Data Reconstruction; 4.3.3 Design Optimizations; 4.4 Experiments and Analysis; 4.4.1 Methodology and Experimental Setup; 4.4.2 Compared Algorithms; 4.4.3 Results; 4.5 Improvements of Compressive Sensing; 4.6 Summary; References; 5 Iterative Compressive Sensing for Fault Detection; 5.1 Background; 5.2 Problem Statement; 5.3 Iterative Compressive Sensing; 5.3.1 Overview; 5.3.2 Optimized Local Median Method; 5.3.3 Time Series and Compressive Sensing; 5.3.4 Discussion; 5.4 Evaluation; 5.4.1 Evaluation Settings

5.4.2 Performance in Faulty Data Detection5.4.3 Performance in Missing Value Reconstruction; 5.4.4 Impact of Faulty and Missing Data in Velocity; 5.4.5 Convergence; 5.5 Summary; References; 6 Homogeneous Compressive Sensing for Privacy Preservation; 6.1 Background; 6.2 Problem Statement; 6.2.1 Trajectory Recovery Model; 6.2.2 User Models and Adversary Models; 6.2.3 Accuracy and Privacy Problem; 6.3 Homogeneous Compressive Sensing Scheme; 6.3.1 Trace Preparation and Validation; 6.3.2 Overview; 6.3.3 Encrypt the Sensed Trajectories at Individual Users

6.3.4 Recover the Encrypted Trajectories at the Server6.3.5 Decrypting the Recovered Trajectories at Individual Users; 6.4 Theoretical Analysis; 6.4.1 Accuracy Analysis; 6.4.2 Privacy Preservation against Eavesdroppers; 6.4.3 Privacy Preservation Against Stalkers; 6.4.4 Complexity Analysis; 6.4.5 Design Discussion; 6.5 Performance Evaluation; 6.5.1 Simulation Settings; 6.5.2 Performance Analysis; 6.5.3 Illustrative Results; 6.6 Summary; References; 7 Converted Compressive Sensing for Multidimensional Data; 7.1 Background; 7.2 Problem Statement; 7.2.1 Preliminary

0

8

8

8

8

This book provides a comprehensive introduction to applying compressive sensing to improve data quality in the context of mobile crowdsensing. It addresses the following main topics: recovering missing data, efficiently collecting data, preserving user privacy, and detecting false data. Mobile crowdsensing, as an emerging sensing paradigm, enables the masses to take part in data collection tasks with the aid of powerful mobile devices. However, mobile crowdsensing platforms have yet to be widely adopted in practice, the major concern being the quality of the data collected. There are numerous causes: some locations may generate redundant data, while others may not be covered at all, since the participants are rarely systematically coordinated; privacy is a concern for some people, who dont wish to share their real-time locations, and therefore some key information may be missing; further, some participants may upload fake data in order to fraudulently gain rewards. To address these problematic aspects, compressive sensing, which works by accurately recovering a sparse signal using very few samples, has proven to offer an effective solution.

Springer Nature

com.springer.onix.9789811377761

When compressive sensing meets mobile crowdsensing.

9789811377754

Data mining.

Mobile communication systems.

Multisensor data fusion.

Sensor networks.

Data mining.

Mobile communication systems.

Multisensor data fusion.

Sensor networks.

TECHNOLOGY & ENGINEERING-- Technical & Manufacturing Industries & Trades.

TEC-- 040000

Kong, Linghe

Chen, Guihai

Wang, Bowen

20200823230904.0

pn

[Book]

Y