Researchers at Northeastern University’s Institute for Intelligent Networked Systems (INSI) have demonstrated the first open-source prototype of a massive MIMO (mMIMO) AI-RAN system. The platform integrates hardware from AmpliTech Group with open-source software from OpenAirInterface (OAI) and NVIDIA GPU-accelerated intelligence.

The system uses an AmpliTech O-RAN Category B radio unit and NVIDIA AI Aerial software for layer 1 and layer 2 radio access network (RAN) functions. By combining these with OAI’s software stack, the researchers created a standards-compliant platform that operates entirely on open-source software.

Technical testing involved a two-stage precoding architecture. This combined a four-layer MIMO precoder, accelerated by NVIDIA’s platform, with a 64-antenna element beamformer implemented in the radio unit. The university reported sustained data throughput across multiple mobile devices, validating that high-performance architectures can be built without proprietary, closed-stack solutions.

Massive MIMO systems, which use large antenna arrays to serve multiple users through spatial multiplexing, have traditionally relied on vendor-specific implementations. This demonstration shows that the full stack, from the physical layer to the control plane, can be assembled from interoperable components.

“What this demonstration shows is that openness and performance are not trade-offs,” said Tommaso Melodia, Director of the Institute for Intelligent Networked Systems at Northeastern University. “You can have a fully open, reproducible system and still push the boundaries of what massive MIMO can deliver through algorithmic control.”