Sensor Management In ISR

This innovative resource is the first book that partitions the intelligence, surveillance and reconnaissance (ISR) sensor management process into partitioned functions that can be studied and optimized independently of each other through defined conceptual interfaces. The book explains the difference between situation information and sensor information and how to compute both. The information-based sensor management (IBSM) approach to real-time orchestrated resource management (ORM) of intelligence, surveillance, and reconnaissance (ISR) assets in the physical, cyber, and social domains are detailed. The integrating concept of mission value through use of goal lattice (GL) methodology is explored. Approaches to implementing real-time sensor management (SM) systems by applying advanced information-based approaches that consider contextual situation and optimization of diverse sensor capabilities for information-based objectives are also covered. These methods have applications in physical intelligence, surveillance, and reconnaissance (ISR), as well as in cyber, and social domains. Based on 30 years of research in developing a mission-valued approach to maximizing the transfer of information from real, cyber, and social environments into a mission-valued, probabilistic representation of that environment on which decision makers can formulate actions, this is the only book that addresses real-time management of ISR from a first principles approach (information theory), and how information theory can be applied to the design and development of ISR systems. Sensor Management in ISR Contents Preface 1 Introduction to Sensor Management 1.1 Motivation for Sensor Management in Intelligence, Surveillence, and Reconaissance 1.2 Sensor Management Versus Data Fusion 1.3 Sensor Management Is Motivated by the Need to Inform Situation Assessment 1.4 Sensor Management 1.5 Sensor Scheduling, Sensor Management, and Mission Management 1.6 Optimum Planning Versus Optimum Scheduling 1.7 Sensor Suite Viewed as a Constrained Communications Channel 1.8 Preliminaries 1.9 Road Map for the Sequel References 2 Historical Basis for Sensor Management 2.1 From Task-Specific Sensor to Heterogeneous Network 2.2 Integration of Frequency Diverse Radars 2.3 Integration of Modality Diverse Sensors During the Vietnam Era 2.4 Networks of Homogeneous Sensors 2.5 Network of Heterogeneous Sensors 2.6 Network-Centric Warfare: The Start of Modern Sensor Management References 3 Sensor Management Inherent Problems 3.1 Indirect Sensor Management Issues 3.2 Multidisciplinary Problem 3.3 Passive Sensor Issues 3.4 Active Sensor Issues 3.5 Virtual Sensor, Heterogeneous Sensor, and Pseudo-Sensor Issues 3.6 World Models 3.6.1 Physical Models 3.6.2 Context 3.6.3 Probabilistic Models 3.6.4 Social Network Models 3.7 Operational Issues 3.7.1 Myopic Scheduling 3.7.2 Sensor Management Objective Functions References 4 Sensor Management Related Problems 4.1 Introduction 4.2 Fusion-Related Issues 4.2.1 Common Frame of Reference and Merging of Data from Different Platforms 4.2.2 Data Association Coordinate System Errors 4.2.3 Data Pedigree 4.2.4 Data Veracity 4.2.5 Hard and Soft Data Fusion 4.3 Alternative Configurations for Search, Track, and Identification 4.4 Detection Criteria 4.5 Target Models 4.6 Scheduling Constraints 4.6.1 Deleterious Interaction of Sensors 4.6.2 Computational Constraints 4.6.3 Randomly Occurring Sensor Failures References 5 Theoretical Approaches to Sensor Management 5.1 Overview of Sensor Management Theories 5.2 Scheduling Approaches Versus Decision-Making Approaches 5.3 Decision Theoretic Approaches 5.4 Normative Decision Theoretic Approaches 5.5 Descriptive Decision Theoretic Approaches 5.6 Sensor Management Architecture-Based Approaches 5.6.1 Decentralized Management 5.6.2 Game Theory-Based Approaches 5.6.3 Market Theory-Based Approaches 5.6.4 Hybrid Approaches References 6 Artificial Intelligence for Sensor Management 6.1 Introduction 6.2 Resurgence of AI 6.3 Specific Mapping of AI Capabilities to IBSM Functions 6.4 Supervised Machine Learning 6.5 Unsupervised Machine Learning 6.6 Data Fusion 6.6.1 Implementing Data Fusion Within IBSM 6.7 Ontologies for Storage and Reasoning 6.8 Characterizing Uncertainty 6.9 Qualitative Reasoning 6.10 Distributed Cognition for Sensor Management 6.11 Levels of Autonomy 6.11.1 A Survey of Levels of Autonomy 6.11.2 Adaptability Enables Autonomy 6.11.3 Measuring Adaptability 6.11.4 Foreseeable Adaptation 6.11.5 Unforeseeable Adaptation 6.11.6 Adaptation Measurement 6.12 Measuring the Effectiveness of Autonomy 6.13 Control Models for Close Coordination of Sensor Platforms 6.14 Machine Learning 6.15 Explainable AI References 7 MQ-4C Triton: a Case Study 7.1 Overview of the Triton Broad Area Maritime Surveillance System 7.2 A Brief History of Triton 7.3 The Triton Sensor Payload 7.4 Operational Management of Triton 7.5 Doctrinal Guidance for Triton Operation 7.6 Sensor Management During a Notional Triton Sortie References 8 Information Theoretic Approach to Sensor Management 8.1 Overview of the IBSM 8.2 Data, Information, and Knowledge 8.3 Information Measures 8.3.1 Fisher Information 8.3.2 Kullback- Leibler Divergence (Also Known as Relative Entropy, (InFormation) Divergence) 8.3.3 Mutual Information (Also Known as Information Gain) 8.3.4 Csiszar-Rényi Generalized Information 8.3.5 Entropy 8.3.6 Knowledge 8.3.7 NIIRS Information 8.3.8 IBSM Information Measures 8.3.9 Sensor Information 8.3.10 Situation Information 8.4 Time Value of Information (TVI) 8.5 The IBSM Model 8.6 Collaboration Among IBSM-Managed Sensor Platforms 8.7 Benefits of the IBSM 8.8 Summary References 9 IBSM Optimization Criterion: Expected Information Value Rate 9.1 Global, Commensurate, Objective Function 9.1.1 EIVRsit Expected Situation Information Value Rate 9.1.2 EIVRsen Expected Sensor Information Value Rate 9.2 BNCO 9.3 Goal Lattice for Situation and Sensor Valuation 9.3.1 Goal Lattice Valuation 9.3.2 Goal Lattice Computation 9.3.3 Method and Apparatus of Measuring a Relative Utility for Each of Several Different Tasks Based on Identified System Goals 9.3.4 Goal Lattice Sensitivity 9.4 System Goal Lattice Examples 9.5 Collaboration Through Goal Lattices 9.6 Orchestrated Goal Lattice Engine References 10 IBSM Implementation Approaches 10.1 Introduction 10.2 Situation Information Expected Value Network 10.2.1 Nonmanaged Nodes 10.2.2 Situation Hypothesis Nodes 10.2.3 Managed Nodes 10.3 Dynamic Bayesian Networks and Situation Information 10.4 Sensor Selection and Control Functions 10.4.1 AFT 10.4.2 Information Instantiator 10.4.3 Merging AFT and the Bottom of the Goal Lattice 10.4.4 Temporal Constraints 10.5 Sensor Scheduler 10.6 Communications Manager 10.7 Situation Assessment Database (SADB) References 11 Future Technologies and Implications 11.1 Introduction 11.2 The IoT as a Sensor System 11.3 Cyber-Physical Systems 11.4 Fifth Generation (5G) Mobile Communication Networks 11.5 Smart Cities 11.6 Sensing-as-a-Service Business Models 11.7 Social Media as a Sensor 11.8 Summary References Acronyms and Abbreviations About the Author Index