Initial Technology Exploration Summary The project began by evaluating multiple 5G signal processing approaches, ultimately leading to the breakthrough discovery of OpenAirInterface5G as the optimal foundation.
Python Library Assessment
Sionna and toolkit5G proved excellent for simulation but inadequate for real-time RF processing. Daniel Estévez's implementation provided valuable insights but contained hardcoded parameters (PCI, SCS, SFN) that limited general-purpose application. MATLAB's 5G Toolbox offered comprehensive capabilities but was financially impractical for open-source development.
OpenAirInterface5G Breakthrough
OAI emerged as the game-changer, providing a production-ready, standards-compliant 5G stack with advanced features like CSI analysis and blind synchronization. The RFSimulator enabled hardware-free development, while the Replay Node facilitated testing with captured IQ samples, though format compatibility challenges persisted.
Implementation Architecture
Function mapping revealed OAI's logical organization mirroring the 5G protocol stack, enabling systematic extraction of critical functions like nr_initial_sync() and polar_decoder(). The six-stage GNU Radio workflow balanced native blocks (FFT, correlation, QPSK demodulation) with custom 5G-specific implementations, maximizing performance while minimizing development complexity.
Practical Validation
PSS implementation using pss_nr.py successfully integrated OAI's proven algorithms into GNU Radio Python blocks, demonstrating effective PSS correlation and NID2 detection. Despite initial replay node timing alignment issues, the approach validated the feasibility of bridging OAI's mature C implementation with GNU Radio's flexible flow-graph architecture for comprehensive 5G cell scanning capabilities.