← All projects

2026 · ECE 519A · Graduate course · Team of 2

5G Massive-MIMO Video Link with YOLOv10

A GPU-accelerated simulation of a full 5G NR physical layer — 64×4 massive MIMO with SVD precoding, QPSK/OFDM, and 3GPP fading channels — carrying live webcam video into real-time YOLOv10 object detection at the receiving edge.

Live demo window: YOLOv10 detection boxes over webcam video received through the simulated 64×4 massive-MIMO link, with a status bar reading DIST 2000 m, SNR 18.4 dB, beamforming gain 8.73, throughput 13.16 Mbps
The live demo — YOLOv10 detections on video received through the 64×4 massive-MIMO link at a simulated 2 km: SNR 18.4 dB, beamforming gain 8.73, 13.16 Mbps.

Overview

Edge AI needs high-resolution video, and 5G links fight path loss, multipath fading, and noise the whole way. This project quantifies how those physical-layer impairments actually affect a modern detector. We built the transmit/receive chain in Python — QPSK modulation, OFDM with cyclic prefix, SVD-based beamforming over a 64×4 antenna configuration, and Rayleigh / 3GPP Urban-Micro channel models — and accelerated the matrix-heavy parts with CuPy so the whole pipeline runs in real time on one GPU alongside the detector.

Live webcam frames are compressed adaptively based on the instantaneous SNR, pushed through the simulated channel, reconstructed, and handed to YOLOv10 — chosen for its NMS-free, low-latency design. The result is a live demo where you can watch detection confidence respond as simulated distance and channel conditions change, from a clean SISO baseline out to a 50 km massive-MIMO downlink.

Highlights

Python CuPy · GPU 5G NR · OFDM Massive MIMO SVD beamforming YOLOv10 OpenCV

Files & downloads