OpenELINT Lab: Bringing EW/RADAR textbook equations to life
I've been working on something a little different lately, and it is finally at the point where I am ready to put it out into the world.
Introducing OpenELINT Lab, an open, browser-based educational environment for exploring radar, electronic warfare, electronic support, and ELINT concepts interactively.
Interactive demonstrations for Radar & Electronic Warfare
https://hongselectronics.github.io/OpenELINT/
The goal behind OpenELINT is pretty simple: take concepts that are normally presented as equations, diagrams, or static examples in textbooks and turn them into something you can actually interact with.
If I change the target range, what happens to received SNR? What does increasing jammer power do to burn-through range? How does PRF affect range and velocity ambiguity? What happens when a phased array places a null on an interferer? What does a DRFM-generated false target look like? How accurately can an ELINT receiver geolocate an emitter using AOA, TDOA, or FDOA?
Instead of just reading the equations, OpenELINT lets you turn the knobs and watch the physics move.
30 Interactive Modules
The current release contains 30 live modules organized around several major areas of radar and electronic warfare.
Radar Fundamentals
- Radar fundamentals and first-order relationships
- The radar range equation
- Phased-array beamforming and adaptive null steering
- Sidelobe cancellation
- Scan loss
- PPI, A-scope, B-scope, and range-Doppler displays
- Ambiguity functions and PRF tradeoffs
- Clutter
- Monopulse tracking
- Propagation and radar horizon effects
One of the simplest examples is the radar range equation. Instead of treating it as a formula to memorize, OpenELINT lets you change each term and immediately see how much leverage it actually provides.
Doubling transmitter power, for example, does not double detection range. Since received radar power falls with the fourth power of range, doubling transmit power increases detection range by only about 19 percent.
That relationship is obvious mathematically, but it becomes much more intuitive when you can move the transmitter-power slider and watch the detection threshold move in real time.
Detection & Engagement
- Radar-versus-jammer engagement geometry
- Burn-through analysis
- Range-height displays
- Receiver-chain noise and sensitivity
- Pulse compression and signal processing
- Multipath
- Detection probability and decision thresholds
- Receiver dynamic range
- Link budgets
- Monte Carlo uncertainty analysis
Electronic Countermeasures & Electronic Protection
- Range Gate Pull-Off (RGPO)
- Velocity Gate Pull-Off (VGPO)
- Coordinated range/velocity deception
- False-target generation
- Angle deception
- DRFM fidelity and phase quantization
- Counter-DRFM techniques
- Digital image synthesis
- DRFM design-space exploration
- I/Q imbalance and image suppression
- Radar Warning Receiver concepts
- Decoys and chaff
The DRFM section is one of the areas where an interactive visualization becomes particularly useful. Concepts such as gate pull-off, false-target generation, finite phase resolution, and electronic-protection responses can be difficult to communicate with a block diagram alone.
OpenELINT includes multiple DRFM sub-modes so the effects can be viewed from different perspectives. Where possible, the models are also checked against examples in the published literature. For example, the DRFM phase-quantization model reproduces published finite-bit spur levels rather than simply drawing a representative spectrum.
ELINT / ESM
This is probably the section most directly responsible for the project's name.
The ELINT/ES environment includes demonstrations covering:
- Radar signal interception
- Pulse Descriptor Words (PDWs)
- Emitter de-interleaving
- Emitter identification
- PRI analysis
- Intrapulse analysis
- Antenna scan analysis
- Energy detection
- Amplitude-comparison and interferometric direction finding
- Bartlett, Capon/MVDR, MUSIC, and Root-MUSIC direction finding
- AOA, TDOA, FDOA, and combined geolocation
- Cross-ambiguity processing
- GDOP and geolocation uncertainty
The ELINT section also illustrates a problem that is easy to overlook when looking only at the equations: geometry matters.
A highly accurate timing or direction measurement does not automatically produce a highly accurate emitter location. Receiver placement, baseline geometry, measurement error, and the selected geolocation technique all contribute to the final uncertainty.
Being able to move between AOA, TDOA, FDOA, and combined techniques makes it much easier to see where each method works well and where the geometry starts working against you.
Special Topics
There are also modules dealing with topics that don't fit neatly into a single radar or EW category, including:
- Low-observable / stealth concepts
- Electronic order of battle
- GNSS/GPS jamming and spoofing
- CRPA anti-jam concepts
- Editable radar and jammer system parameters
First-Principles Physics, Not Pre-Rendered Animations
One of the things I wanted to avoid was creating a collection of animations that merely look like radar displays.
The plots and readouts in OpenELINT are calculated live from the underlying models. Change a parameter and the calculations are run again.
The radar range equation, array factors, Friis noise calculations, jammer-to-signal relationships, detection thresholds, direction-finding algorithms, geolocation geometry, DRFM effects, and other demonstrations are implemented directly in the application.
That means the purpose of the tool is not just to show what something looks like. The more interesting part is exploring why it behaves that way.
Those relationships become much easier to internalize when you can change a slider and immediately see the resulting curve move.
Built Around Published Literature
I also wanted the models to be traceable back to published material rather than becoming another collection of unexplained equations copied around the Internet.
The project includes a substantial reference section identifying the textbooks, papers, technical reports, and application notes behind the individual demonstrations.
Some of the major references include work by authors such as Lee Harrison, Richard Poisel, Phillip Pace, Nicholas O'Donoughue, David Adamy, Richard Wiley, Sue Robertson, David Lynch, and others, along with the Naval Air Warfare Center Weapons Division's Electronic Warfare and Radar Systems Engineering Handbook.
Where practical, I have also tried to validate calculated results against examples and tables contained in the source material.
The goal is for the visualizations to remain tied to the underlying engineering. For example, DRFM phase-quantization results are checked against published values, and I/Q image-suppression behavior is compared against reference examples.
That is an important distinction to me. I want OpenELINT to be useful as a learning tool, but I also want someone to be able to ask, "Where did this model come from?" and have an answer.
A Browser Is the Lab Equipment
Another design goal was keeping the barrier to entry extremely low.
There is no MATLAB installation, Python environment, SDR hardware, license server, or specialized test equipment required to start experimenting with the concepts.
You just open the page.
The application runs directly in a modern web browser and has no external runtime dependencies. This also makes it useful for demonstrations in classrooms, presentations, workshops, or anywhere else where installing a specialized engineering environment would be inconvenient.
Editable Scenarios
OpenELINT also includes a shared systems database for radar and jammer parameters.
Parameters can be modified and the changes propagate through the applicable demonstrations. Users can create hypothetical systems, change radar or jammer characteristics, and see how those changes affect engagement geometry, scopes, link budgets, and other calculations.
System parameter sets can also be exported and imported as CSV files, making it possible to save example scenarios or build classroom exercises around a common dataset.
What OpenELINT Is Not
OpenELINT is an educational and illustrative tool.
The included system parameters are generic, notional, class-representative values intended for instruction and relative comparison. They are not intended to represent the measured performance of specific operational systems.
No [REDACTED] information is used or implied, and the application is not intended for operational planning or real-world targeting.
The interesting part of this project is the underlying physics and signal-processing concepts, not building a database of sensitive system capabilities.
Why I Built It
A recurring theme with this page has always been learning by doing. "You can just do things."
I can read the radar range equation in a textbook. I can read about burn-through, adaptive arrays, DRFM deception, pulse de-interleaving, MUSIC direction finding, or TDOA geolocation.
But I learn these subjects much more effectively when I have to implement them.
Building the visualization forces another level of understanding. Suddenly you have to decide exactly what every variable means, what assumptions the equation is making, what units are being used, what happens at the edge cases, and whether the result actually makes physical sense.
And then you start changing things.
What if the jammer is in a sidelobe?
What if I change the PRF?
What if the two emitters are only a few degrees apart?
What if the DRFM only has a few bits of phase resolution?
What if my TDOA receiver geometry is terrible?
That is where a static equation turns into engineering intuition.
Try OpenELINT Lab
The project is available now:
hongselectronics.github.io/OpenELINT/
The project is open and intended to continue evolving. There are plenty of additional radar, EW, SIGINT, and RF concepts that would benefit from interactive demonstrations, so I expect the lab to keep growing.
If you work in radar, RF, electronic warfare, signal processing, or you're simply trying to learn the subject, give it a try and let me know what you think.
Comments
Post a Comment