Description
Book SynopsisOcean covers 70.8% of the Earth’s surface, and it plays an important role in supporting all life on Earth. Nonetheless, more than 80% of the ocean’s volume remains unmapped, unobserved and unexplored. In this regard, Underwater Sensor Networks (USNs), which offer ubiquitous computation, efficient communication and reliable control, are emerging as a promising solution to understand and explore the ocean. In order to support the application of USNs, accurate position information from sensor nodes is required to correctly analyze and interpret the data sampled. However, the openness and weak communication characteristics of USNs make underwater localization much more challenging in comparison to terrestrial sensor networks.
In this book, we focus on the localization problem in USNs, taking into account the unique characteristics of the underwater environment. This problem is of considerable importance, since fundamental guidance on the design and analysis of USN localization is very limited at present. To this end, we first introduce the network architecture of USNs and briefly review previous approaches to the localization of USNs. Then, the asynchronous clock, node mobility, stratification effect, privacy preserving and attack detection are considered respectively and corresponding localization schemes are developed. Lastly, the book’s rich implications provide guidance on the design of future USN localization schemes.
The results in this book reveal from a system perspective that underwater localization accuracy is closely related to the communication protocol and optimization estimator. Researchers, scientists and engineers in the field of USNs can benefit greatly from this book, which provides a wealth of information, useful methods and practical algorithms to help understand and explore the ocean.
Table of Contents
1. Introduction
1.1 Network Architecture of Underwater Sensor Networks
1.2 Prior Arts in Localization
1.3 Underwater Weak Communication Characteristics
2. Asynchronous Localization of Underwater Sensor Networks with Mobility Prediction
2.1 Introduction
2.2 System Modeling and Problem Formulation
2.3 Design of Asynchronous Localization Approach
2.4 Performance Analysis
2.4.1 Convergence
2.4.2 Cramer-Rao Lower Bound
2.5 Simulation
2.5.1 Simulation of Active Sensor Node
2.5.2 Simulation of Passive Sensor Node
2.6 Summary
References
3. Asynchronous Localization of Underwater Sensor Networks with Consensus-Based Unscented Kalman Filtering
3.1 Introduction
3.2 System Modeling and Problem Formulation
3.3 Design of Consensus-Based UKF Localization Approach
3.4 Performance Analysis
3.4.1 Observability Analysis
3.4.2 Convergence Conditions
3.4.3 Cramer-Rao Lower Bound
3.4.4 Computational Complexity Analysis
3.5 Simulation
3.6 Summary
Reference
4. Reinforcement Learning Based Asynchronous Localization of Underwater Sensor Networks
4.1 Introduction
4.2 System Modeling and Problem Formulation
4.3 Design of Reinforcement Learning Based Asynchronous Localization Approach
4.4 Performance Analysis
4.4.1 Convergence Conditions
4.4.2 Cramer-Rao Lower Bound
4.4.3 Computational Complexity Analysis
4.5 Simulation
4.5.1 Advantage of the RL-Based Localization Strategy
4.5.3 Simulation of Active Sensor Node
4.5.4 Simulation of Passive Sensor Node
4.6 Summary
Reference
5. Privacy Preserving Asynchronous Localization of Underwater Sensor Networks
5.1 Introduction
5.2 System Modeling and Problem Formulation
5.3 Design of Privacy Preserving Based Localization Approach
5.3.1 Design of Active Sensor Node Localization Strategy
5.3.2 Design of Ordinary Sensor Node Localization Strategy
5.4 Performance Analysis
5.4.1 Equivalence Analysis
5.4.1.1 Equivalence Analysis of Active Sensor Node
5.4.1.2 Equivalence Analysis of Ordinary Sensor Node
5.4.2 Level of Privacy Preservation
5.4.3 Communication Complexity Analysis
5.5 Simulation
5.5.1 Simulation of Active Sensor Node
5.5.2 Simulation of Ordinary Sensor Node
5.6 Summary
Reference
6. Privacy-Preserving Asynchronous Localization of Underwater Sensor Network with Attack Detection and Ray Compensation
6.1 Introduction
6.2 System Modeling and Problem Formulation
6.3 Design of Privacy-Preserving Localization Approach
6.4 Performance Analysis
6.4.1 Equivalence Analysis
6.4.2 Influencing Factors of Localization Errors
6.4.3 Privacy-Preserving Property
6.4.4 Tradeoff Between Privacy and Transmission Cost
6.5 Simulation
6.6 Summary
Reference
7. Deep Reinforcement Learning Based Privacy-Preserving Localization of Underwater Sensor Networks
7.1 Introduction
7.2 System Modeling and Problem Formulation
7.3 Design of Privacy-Preserving Localization Protocol
7.4 Design of DRL-based Localization Approach
7.4.1 Localization when All Data is Unlabeled
7.4.2 Localization when Labelled Data Occupies the Majority
7.4.3 Localization when Unlabeled Data Occupies the Majority
7.5 Performance Analysis
7.5.1 Equivalence of the Localization Protocols
7.5.2 Privacy Preservation Analysis
7.5.3 Global Optimum Analysis
7.5.4 Computation Complexity Analysis
7.6 Simulation
7.6.1 Advantage of the DRL-Based Localization Strategy
7.6.2 Simulation of Unsupervised DRL-Based Estimator
7.6.3 Simulation of Supervised DRL-Based Estimator
7.6.4 Simulation of Semisupervised DRL-Based Estimator
7.7 Summary
Reference
8. Conclusion and future perspective
