Nvidia corporation (20240265619). LEARNING DIGITAL TWINS OF RADIO ENVIRONMENTS simplified abstract

LEARNING DIGITAL TWINS OF RADIO ENVIRONMENTS

Organization Name

nvidia corporation

Inventor(s)

Faycal Ait Aoudia of Santa Clara CA (US)

Jakob Richard Hoydis of Courbevoie (FR)

Nikolaus Binder of Berin (DE)

Merlin Nimier-david of Zürich (CH)

Sebastian Cammerer of Berliin (DE)

Alexander Georg Keller of Berlin (DE)

Guillermo Anibal Marcus Martinez of Berlin (DE)

LEARNING DIGITAL TWINS OF RADIO ENVIRONMENTS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240265619 titled 'LEARNING DIGITAL TWINS OF RADIO ENVIRONMENTS

The patent application relates to learning digital twins of radio environments using differentiable ray tracing.

• Differentiable ray tracing refines scene geometry of the physical environment.
• It optimizes scene properties of objects and antennas in the scene.
• It learns and optimizes antenna patterns, array geometries, and positions of transmitters and receivers.
• The ray tracer simulates radio wave propagation to evaluate different configurations of scene geometry and radio devices.
• Scene geometry, properties, and antenna characteristics are computed by a differentiable parametric function like a neural network.
• Parameters of the parametric function are learned using differentiable ray tracing.

Potential Applications: - Wireless communication network optimization - Autonomous vehicle sensor simulation - Radar system design and testing

Problems Solved: - Improving radio wave propagation modeling accuracy - Enhancing the performance of wireless communication systems - Optimizing antenna configurations for better signal reception

Benefits: - Increased efficiency in designing radio environments - Cost savings through simulation-based optimization - Enhanced performance of wireless devices

Commercial Applications: "Optimizing Wireless Communication Systems Through Advanced Simulation Techniques"

Questions about the technology: 1. How does differentiable ray tracing improve the accuracy of radio wave propagation modeling? 2. What are the key advantages of using a differentiable parametric function like a neural network in this context?

Original Abstract Submitted

embodiments of the present disclosure relate to learning digital twins of radio environments. differentiable ray tracing may be used to refine the scene geometry of the physical environment, to learn or optimize the scene properties of objects in the scene, to learn or optimize the scene properties of antennas, and to learn or optimize antenna patterns, array geometries, and orientations and positions of transmitters and receivers. once scene properties have been learned or optimized, the differentiable ray tracer may further be used to simulate radio wave propagation to simulate the performance of different configurations of the scene geometry and radio devices, such as antennas. in an embodiment, one or more of the scene geometry, scene properties, and antenna characteristics are computed by a differentiable parametric function, such as a neural network, etc. and parameters of the differentiable parametric function are learned using the differentiable ray tracing.