Analog AI at the antenna.
AML200 is the first analog AI chip to classify gigahertz RF signals before the ADC. Provides the efficiency to push RF intelligence to the edge for defense and commercial applications that need signal classification and intelligence.
Digital RF processing is too power-hungry for the edge.
Wideband RF means fast, expensive ADCs and heavy downstream digital processing. This power and latency burden has kept real-time RF intelligence tethered to racks and vehicles.
Classify the signal before it's ever digitized.
How it works
- SenseAML200 sits between the antenna array and the RF SoC.
- ClassifyIt classifies wideband RF in the analog domain, before the ADC.
- ReduceCompute and power drop across the whole receiver chain.
- ReachLow-power EW, SATCOM, 5G/6G networks, and sensing the digital approach can't serve.
One core, across the RF mission.
The same pre-ADC classification serves defense and commercial RF from electronic warfare to the 5G/6G network edge. Click any concept to enlarge.
EW & SIGINT
Reject jamming, fingerprint emitters.
Radar
Echoes to target class, in real time.
SATCOM
Prioritize uplink, filter the rest.
5G / 6G
Classify at the radio unit and the edge.
Illustrative application concepts. AML200 is in development; samples targeted for 2027.
Building at the RF edge? Let's compare notes.
AML200 is in development with samples targeted for 2027. If RF SWaP is your bottleneck, we'd like to understand your requirements.