https://gustybear.github.io/tags/testbed/TestbedHugo Blox Builder (https://hugoblox.com)en-usMon, 01 Sep 2025 04:14:54 -0800https://gustybear.github.io/media/logo_hu_d0a0b1783c391ac0.pnghttps://gustybear.github.io/tags/testbed/https://gustybear.github.io/facility/testbed_airan/Mon, 01 Sep 2025 04:14:54 -0800https://gustybear.github.io/facility/testbed_airan/<h1 id="executive-summary">Executive Summary</h1> <p>This lab-scale mmWave AI-Based RAN testbed integrates <strong>OAIBox</strong>, <strong>NVIDIA Aerial RAN</strong>, <strong>NI USRP X410</strong>, <strong>TMYTEK mmWave beamformers</strong>, and a high-fidelity <strong>Digital Twin pipeline</strong> using <strong>Remcom Wireless InSite</strong> and <strong>ANSYS HFSS SBR+</strong> to create a flexible, programmable, and AI-native 5G/6G research environment. It enables real-time prototyping of mmWave physical layers, AI-driven beam management, hybrid beamforming, AI-enhanced MAC scheduling, and joint communication–sensing (ISAC) experiments. The platform supports end-to-end 5G NR PHY/MAC stacks, GPU-accelerated baseband processing, and mmWave RF front-ends for high-bandwidth OTA testing, while the Digital Twin provides physics-accurate ray-tracing, EM-based antenna modeling, and virtual–to–real co-simulation for channel prediction, beam optimization, and AI dataset generation.</p> <h1 id="core-components">Core Components</h1> <h2 id="oaibox--openairinterface-ran-framework">OAIBox – OpenAirInterface RAN Framework</h2> <p>The testbed uses <strong>OAIBox</strong> as a compact and modular implementation of the full OAI RAN stack. It provides:</p> <ul> <li>Support for 5G SA/NSA gNB and UE</li> <li>Flexible PHY–MAC integration</li> <li>Customizable scheduling, HARQ, and protocol features</li> <li>Real-time experimentation with RAN procedures and RRC signaling</li> </ul> <p>OAIBox acts as the protocol and control anchor of the testbed.</p> <h2 id="nvidia-aerial-ran-cuphy--cumac">NVIDIA Aerial RAN (cuPHY + cuMAC)</h2> <p>The <strong>NVIDIA Aerial</strong> platform provides GPU-accelerated baseband processing and AI-native PHY/MAC capabilities:</p> <ul> <li><strong>cuPHY</strong> for NR physical-layer DSP on GPUs</li> <li><strong>cuMAC</strong> for dynamic MAC scheduling on GPU</li> <li>TensorRT for real-time neural inference</li> <li>Support for multi-cell, multi-user, and high-throughput pipelines</li> </ul> <p>Aerial enables experiments in:</p> <ul> <li>AI-driven beam selection and prediction</li> <li>Neural channel estimation</li> <li>Predictive link adaptation and blockage detection</li> </ul> <h2 id="ni-usrp-x410--wideband-software-defined-radio">NI USRP X410 – Wideband Software-Defined Radio</h2> <p>The <strong>USRP X410</strong> serves as the flexible transceiver frontend with:</p> <ul> <li>Up to 400 MHz instantaneous bandwidth</li> <li>Four synchronized TX/RX channels</li> <li>10/1588 PTP synchronization</li> <li>Digital IF for integration with NVIDIA Aerial</li> </ul> <p>It supports:</p> <ul> <li>mmWave IF/baseband experimentation</li> <li>Real-time CSI acquisition</li> <li>Multi-subarray MIMO and wideband waveform prototyping</li> </ul> <h2 id="tmytek-mmwave-beamformers-bbox-ud-box-beamform-modules">TMYTEK mmWave Beamformers (BBox, UD-Box, Beamform Modules)</h2> <p>TMYTEK hardware provides programmable mmWave RF front-ends:</p> <ul> <li><strong>UD-Box</strong> for 24–32 GHz up/down-conversion</li> <li><strong>BBox One / BBox Lite</strong> beamforming arrays</li> <li>API-driven phase/gain control</li> <li>Rapid beam steering and codebook-based operation</li> </ul> <p>These modules enable:</p> <ul> <li>Hybrid or analog beamforming</li> <li>Electronic steering up to ±60°</li> <li>Multi-beam and multi-focus mmWave experimentation</li> </ul> <h2 id="digital-twin">Digital Twin</h2> <p>A full <strong>Digital Twin framework</strong> integrates high-fidelity electromagnetic simulation with the physical testbed. This enables reproducible channel modeling, data augmentation, and virtual-to-real RAN optimization.</p> <h3 id="remcom-wireless-insite--ray-tracing-propagation">Remcom Wireless InSite – Ray Tracing Propagation</h3> <p>Wireless InSite provides a large-scale propagation environment supporting:</p> <ul> <li>GPU-accelerated 3D ray tracing</li> <li>Detailed mmWave diffraction, reflection, and scattering</li> <li>Urban, indoor, and open-field scenario modeling</li> <li>Material-dependent loss and blockage effects</li> <li>Beam-level channel prediction</li> </ul> <h3 id="ansys-hfss-sbr--full-wave-em-modeling">ANSYS HFSS SBR+ – Full-Wave EM Modeling</h3> <p>HFSS SBR+ enables full-wave modeling of:</p> <ul> <li><strong>Antenna arrays</strong>, including TMYTEK beamformers</li> <li><strong>Reflectarrays</strong>, metasurfaces, and RIS</li> <li><strong>Realistic gain patterns</strong> for hybrid beamforming</li> <li>Complex EM interactions under mmWave frequencies</li> </ul> <h1 id="capabilities">Capabilities</h1> <h2 id="ai-enhanced-ran-intelligence">AI-Enhanced RAN Intelligence</h2> <ul> <li>Neural beam prediction and tracking</li> <li>AI-based MAC scheduling</li> <li>CSI-driven link adaptation models</li> <li>Blockage prediction and proactive beam switching</li> </ul> <h2 id="mmwave-phymac-research">mmWave PHY/MAC Research</h2> <ul> <li>Hybrid and digital beamforming</li> <li>Channel sounding and dataset generation</li> <li>Evaluation of mobility, rotation, and blockage</li> <li>5G NR waveform prototyping</li> </ul> <h2 id="flexible-ran-architecture">Flexible RAN Architecture</h2> <ul> <li>O-RAN 7.2 split between OAIBox and Aerial</li> <li>Multi-RU and multi-sector emulation</li> <li>Edge-cloud cooperative intelligence</li> </ul> <h2 id="isac-integrated-sensing-and-communication-extensions">ISAC (Integrated Sensing and Communication) Extensions</h2> <ul> <li>Joint radar–communication waveform experiments</li> <li>2D/3D angle estimation</li> <li>Passive sensing with GPU-accelerated FFT pipelines</li> </ul> <h2 id="digital-twindriven-insights">Digital Twin–Driven Insights</h2> <ul> <li>Predictive channel statistics and blockage maps</li> <li>Virtual scenario pre-testing</li> <li>Dataset augmentation for AI training</li> <li>Virtual beam codebook optimization</li> </ul>https://gustybear.github.io/facility/testbed_holosmartwe/Mon, 19 May 2025 04:14:54 -0800https://gustybear.github.io/facility/testbed_holosmartwe/<h1 id="executive-summary">Executive Summary</h1> <p>This lab-scale <strong>Holographic Beamforming Surface–Enabled Smart Wireless Environment testbed</strong> enables controlled experimentation on environment-aware and RIS-assisted mmWave propagation. Operating at <strong>28 GHz</strong>, the platform integrates a programmable <strong>reconfigurable intelligent surface (RIS)</strong>, <strong>active mmWave beamforming</strong>, and a flexible <strong>RF measurement backend</strong> to dynamically manipulate electromagnetic wavefronts at both the transmitter and environmental levels. The testbed supports experimental research on holographic beamforming, surface-assisted channel shaping, and smart wireless environments for future <strong>5G/6G</strong> and <strong>ISAC</strong> systems.</p> <h1 id="core-components">Core Components</h1> <h2 id="reconfigurable-intelligent-surface-ris">Reconfigurable Intelligent Surface (RIS)</h2> <p>A programmable <strong>RIS</strong> functions as a spatial electromagnetic aperture, enabling dynamic control of reflection phase and amplitude for wavefront shaping, beam redirection, and holographic focusing.</p> <h2 id="tmytek-bbox-one-mmwave-beamformer">TMYTEK BBox One mmWave Beamformer</h2> <p>The <strong>TMYTEK BBox One</strong> provides programmable beamforming at <strong>28 GHz</strong> with electronic phase and gain control, supporting rapid beam steering and codebook-based operation.</p> <h2 id="directive-mmwave-illumination">Directive mmWave Illumination</h2> <p>A <strong>28 GHz horn antenna</strong> delivers stable, high-gain illumination of the RIS and measurement region, enabling repeatable and well-characterized propagation experiments.</p> <h2 id="rf-switching-and-measurement">RF Switching and Measurement</h2> <p>A <strong>custom RF switch matrix</strong> and <strong>Keysight FieldFox microwave analyzer</strong> support flexible signal routing and wideband channel measurement for systematic RIS-assisted characterization.</p> <h2 id="spatial-probing">Spatial Probing</h2> <p>Calibrated <strong>monopole antennas</strong> mounted on precision positioning stages enable spatial channel sampling and beam profiling.</p> <h1 id="capabilities">Capabilities</h1> <h2 id="holographic-and-ris-assisted-beamforming">Holographic and RIS-Assisted Beamforming</h2> <ul> <li>Programmable wavefront synthesis and spatial focusing</li> <li>Environment-level beam steering and reflection control</li> </ul> <h2 id="mmwave-channel-characterization">mmWave Channel Characterization</h2> <ul> <li>Wideband channel response measurement</li> <li>Spatial field mapping and repeatable benchmarking</li> </ul> <h2 id="smart-wireless-environment-research">Smart Wireless Environment Research</h2> <ul> <li>Environment-aware propagation control</li> <li>Foundations for AI-driven and surface-assisted wireless systems</li> </ul> <h2 id="isac-oriented-extensions">ISAC-Oriented Extensions</h2> <ul> <li>Surface-assisted sensing and localization</li> <li>Joint communication–environment manipulation</li> </ul>