Applications of Enterprise Integration Patterns to Near-Real Time Radar Data Processing
summary
Experimental multistatic radar processing uses Spring Integration flows to stream range Doppler maps through service activators that call the MATLAB Compiler Runtime (single‑threaded), so processing scales by running multiple JVMs and swapping channel adapters for different deployments. A networked cyclic barrier implements cross‑node pipeline coordination, message channels reorder and buffer by sensor timestamps, and UDT congestion stats tune buffer sizes. End‑to‑end latency meets about 2 seconds or less in field tests, including a DHS experiment that detected low‑flying smugglers, with outputs delivered as KML or STANAG 4607.
AI-written summary · about
chapters (17)
- 0:00Intro
- 1:10Your Presenter
- 2:28What is MITRE?
- 5:34What is Radar?
- 9:04Simplest Case - Monostatic Radar
- 10:51Fusion of Multiple Monostatic Radars
- 12:27Trade-Offs
- 15:40It's a Data Reduction Problem
- 19:28Technical Challenges
- 24:30Pipeline Processing Steps
- 27:41Multiple Input/Output Types
- 29:30Development Environment
- 30:59Flexible Deployments
- 34:45Coordinated Processing
- 37:04Keeping Up
- 38:45This Stuff Works
- 40:11Thoughts
related talks
30mFive Challenges in Real-time Stream Processing and Five Solutions
08mReal-Time Data Systems: Speed, Scale, and Resilience
53mIntroduction to data streaming
38mRealtime Data Activation - Analytics, Intelligence & Decision Making
29mData Mesh Streaming Patterns: A Real-time Perspective
25mAccelerating Data Processing and Consumption for Building Modern Apps at the Edge
38mHow to run Real Time processing on Big Data
30mIntroduction to Data Streaming