newsfilter
← Latest News
PR Newswire·

Northeastern University Demonstrates First Open-Source Prototype of a Massive MIMO AI-RAN System

AMPG
  • Northeastern University's Institute for Intelligent Networked Systems (INSI) Open6G OTIC demonstrated the first open-source prototype of a massive MIMO AI-RAN system.
  • The system integrates AmpliTech Group's ORAN compliant massive MIMO hardware with OpenAirInterface (OAI) open-source software & NVIDIA GPU-accelerated RAN intelligence.
  • The prototype combines an Amplitech mMIMO O-RAN Category B radio unit, NVIDIA AI Aerial for layer 1 & 2 RAN, & OAI's L2+ stack into a single cohesive platform using exclusively open-source software.
  • A two-stage precoding architecture was showcased, featuring a 4-layer CSI-feedback-based MIMO precoder accelerated on NVIDIA AI Aerial & a 64-antenna element codebook-based beamformer in the CAT-B O-RU.
  • The demonstration validated sustained throughput across multiple user equipment under mobility conditions, proving high-performance AI-RAN architecture can exist without proprietary, closed-stack solutions.
  • INSI certified the AmpliTech Group mMIMO CAT-B radio unit for O-RAN conformance to ensure interoperability within the open ecosystem.
  • Tommaso Melodia, Director of INSI, stated the results show that openness & performance are not trade-offs in AI-RAN development.
  • Irfan Ghauri of OpenAirInterface highlighted the significance of integrating OAI's CU/DU stack with NVIDIA AI Aerial & Amplitech hardware on open foundations.
  • Fawad Maqbool, CEO of AmpliTech, noted the integration validates that high-capacity massive MIMO performance & true Open RAN openness can coexist in disaggregated network environments.
  • The INSI team led system integration, testbed configuration, & validation measurements to provide a reproducible reference implementation for academic & industry researchers.
  • This milestone aligns with growing momentum for Open RAN & AI-native wireless systems essential for future 5G & 6G deployments.