In House Short Course Content

• SIGNAL AND DATA PROCESSING OVERVIEW
• Sensor Processing Functions
• Signal Processing
• Tracking: Data Association and Filtering
• Target Typing or Classification
• Situation Assessment and Sensor Tasking
• Definition of Terms
• Signal Processing
• TDI and Matched Filter
• Type I and II Errors
• Signal Processing Stages and Functions
• Alternative Signal Processing Approaches

• MULTIPLE FRAME SIGNAL PROCESSING
• Multiple Frame Processing: Advantages and Disadvantages
• Alternative Approaches and Comparisons
• Background Subtraction
• Detection in State Space
• The Optimal Matched Filters
• Track Before Detect
• Recent Advances

• DATA ASSOCIATION (CORRELATION) AND FILTERING ISSUES
• Critical Issues
• Complicating Factors and Misassociations
• Sparse Vs Dense Environments
• Basic Functions
• Multiple Sensor Approaches
• Impact of Algorithm Selection
• Performance Vs Required Resources
• System Decomposition Methodology
• Current Challenges in Algorithm Development
• Practical Single and Multiple Target Tracking
• Track Initiation and Termination
• Track Maintenance and Sensor Fusion
• Major Classes of Tracking Problems

• ESTIMATION METHODS FOR TARGET TRACKING
• Pertinent Probabilistic Methods
• Parameter Estimation and Kalman Filtering Methods
• Error Analysis with and without Model Errors

• DECISION AND TARGET IDENTIFICATION METHODS
• Pertinent Pattern Recognition Methods
• Linear and Quadratic Discriminant Scores
• Markov Chain Techniques

• ISOLATED TARGET TRACKING WITH FALSE SIGNALS
• Major Aspects of Track Processing
• Simple and Complex Algorithms
• Classical Nearest Neighbor
• Gated Optimal
• Probabilistic Data Association Filter (PDAF)
• Multiple Frame Association Approaches
• Processing Maneuvering Targets: IMM and Alternative Algorithms
• Impact of Dense Environment on Tracking Performance for Various Algorithms
• Characteristic of Performance Vs Required Resources

• GATE ANALYSIS
• Gating Approaches
• Traditional Vs A Two-Stage Gating
• Gate Sizing
• Gate Search: Potentially Resource Intensive Function
• Predicting the Number of Detections in a Gate
• Relative Observation Density As a Predictor of Sensor Effectiveness, Processing Complexity and Performance

• OPTIMAL MULTIPLE TARGET TRACKING METHODS
• Solutions for Various Optimization Criteria
• Minimum Mean Square Error (MMSE)
• Maximum A Posteriori Probability (MAP)
• Simultaneous MAP (JMAP)
• Basis Free MMSE
• Comparing Alternate Methods
• A "Simple" Two Target Example
• Computation of Two Optimal Solutions
• Evaluation of Alternative Solutions
• Unique Multiple Target Tracking Phenomena
• Pertinent Likelihood Formulas

• PRACTICAL MULTIPLE TARGET TRACKING APPROACHES
• Dangers of Independent Track Processing
• The Assignment Approach
• Algorithms and Algorithm Architecture
• Optimal and Sub-Optimal Assignment
• Important Sparse Matrix Considerations
• Performance Results Vs Target Density
• Track Initiation Approaches
• Morefield's Approach
• Alternative Algorithms: Advantages and Disadvantages
• Joint Probabilistic Data Association (JPDA)
• Reid's Multiple Hypothesis Tracking (MHT)
• Forming the Multiple Hypothesis Tree
• Combining, Trimming and Clustering
• Approach Comparison and Algorithm Selection

• MULTIPLE SENSOR TRACKING AND SENSOR DATA FUSION
• Alternative Multiple Sensor Approaches
• Comparing Track Initiation Methods
• Ambiguities and Number of Frames Needed
• Multiple Sensor Track Maintenance
• Track-to-Track Association Approaches
• Impact of Preferential Tracking
• Sensor Data Fusion
• Two-Source Processing
• N-Source Data Fusion
• Alternative Algorithm Architectures
• Centralized, Distributed, Hierarchical and Hybrid Fusion Approaches
• Recent Developments in Track Fusion and Tracklets
• Decorrelation of Sensor Tracks
• Tracklets from Shadow Tracker
• Characteristics of Approaches and Selection Criteria

• PERFORMANCE EVALUATION OF MULTIPLE TARGET TRACKING
• Ambiguities caused by Misassociation or Unresolved Closely Spaced Objects
• A Two Step Methodology
• Alternative Assignment Approaches
• Criteria for Selecting an Approach
• A Proposed Methodology
• Measures of Performance
• Critical Performance Evaluation Parameters

• MULTIPLE SENSOR TRACKING OF CLUSTERS OR SMALL TARGETS
• The Challenge of and Needs for Cluster Tracking
• A Powerful Method for Tracking Clusters or Small Extended Objects
• Model and Parameter Selection
• Centroid and Extent Tracking Algorithms
• Performance Evaluation and Recent Results

• ALGORITHM SIZING: PREDICTING REQUIRED RESOURCES
• Filter Sizing and Tradeoffs
• Sizing Track Maintenance Processing
• An Analytical Sizing Method
• Predicting Required Thru-Put and Memory
• Sizing Track Initiation Processing
• Number of Candidate Tracks Per Real and False Track Initiators
• Predicting Required Resources

• BIBLIOGRAPHY

  Return To Course Menu

Copyright 1995 O.E. Drummond