Threat emulation and the scatterer method
What a synthetic scene has to get right
A target is not a single number to a radar. To stand in for a real one, a synthetic return has to carry several properties at once, and which ones matter most depends on what the system under test is trying to measure.
Range, as a time delay. Distance is read from the round-trip time of a pulse, so the emulated return is delayed to place the target where the scenario wants it.
Velocity, as a Doppler shift. A closing or receding target shifts the returned frequency, and the emulator applies that shift so the radar reads the intended speed.
Size, as amplitude and radar cross section. How strongly a target reflects sets the return strength, and a realistic scene varies it the way real objects do.
Angle, as a direction of arrival. A real target sits somewhere in the field of view, and a system that tracks direction has to be tested against returns that genuinely arrive from the right bearing. This is the property that most separates the two methods below.
Structure and environment. Extended targets glint and spread, the background adds clutter and multipath, and electronic attack adds deception. A credible scene includes these rather than one clean point.
Digital playback and physical scatterers
Digital methods capture the radar's transmitted pulse, hold it in memory, modify it, and retransmit it. This is the family built around digital radio frequency memory, or DRFM. Because the copy is shaped in software, these systems can place many false targets at once, sweep their range and Doppler, and reproduce deception techniques such as range-gate and velocity-gate pull-off on demand. They are compact, highly repeatable, and need no chamber when the signal is injected straight into the receiver chain.
The tradeoff is that direct injection can bypass the antenna. If the synthetic return enters after the aperture, the test exercises the receiver and the processor but not the antenna pattern or the way the system estimates the direction of a target.
Physical, scatterer-based methods work over the air. A controllable element receives the radar's actual transmitted signal and re-transmits a shaped echo back to it, so the radar sees a genuine reflection that has traveled through its own antenna. The public literature describes these as radar target simulators that generate virtual echoes perceived as real targets by the system under test, with each channel using one antenna to receive and another to re-radiate.
The distinctive property is angle. Because a scatterer occupies a real position, the direction a synthetic target appears to come from is set by where the scatterer physically sits. That makes the method well suited to testing antennas, monopulse and angle tracking, and seekers, which a bypassed injection cannot fully reach. A field of scatterers can build an extended target with a distributed radar cross section.
Its tradeoff is the mirror image of the digital one. The number of independent scatterers is bounded by the number of antenna channels, so dense scenes are approximated by clustering scattering points, and the setup needs a chamber or range and physical positioning rather than a software change.
In practice the two are often combined. A digital core can drive a physical scatterer front end, pairing programmable waveforms with real over-the-air angle. The families are complementary, not rivals.
Choosing between them
There is no method that is correct in general. The right one follows from the test objective. If the goal is to stress signal processing against many false targets and deception techniques, the flexibility of the digital approach is the point. If the goal is to validate how a system finds the direction of a target, the physical realism of a scatterer scene is the point. Cost, available facilities, and the band in use all weigh in.
Rogue River Tech indexes both approaches and the public work behind them. It does not rank them, and it does not treat either as the default.
Feeding the bench
Threat emulation is the supply side of the bench. It is how a realistic scene gets fed into the stages of test and evaluation that use real hardware, from hardware-in-the-loop through integrated and installed-system testing, before anything moves to a live range. The same progression that proves a radar applies to the scenes used to prove it.
For the system on the receiving end of these scenes, see hardware-in-the-loop radar testing.
Standards and spectrum
A scatterer front end radiates, and any open-air or installed test that transmits needs the right authorization. In the United States the path is the FCC Experimental Radio Service under 47 CFR Part 5, granted on a non-interference basis and coordinated with the National Telecommunications and Information Administration for shared federal bands. Under 47 CFR 5.7(g), an experimental license is not required when the device operates fully within an anechoic chamber or a Faraday cage. The hardware-in-the-loop explainer covers this in more detail.
FCC Part 5, Experimental Radio Service ↗ Spectrum reference Standards index
Primary public sources
The methods here rest on established, openly published work. Two examples, one from each family: