【Ray tracing with NeRF】WINERT

标签：Phi right tracing boldsymbol left Theta WINERT NeRF

WINERT: TOWARDS NEURAL RAY TRACING FOR WIRELESS CHANNEL MODELLING AND DIFFERENTIABLE SIMULATIONS

1. Overview

The task of channel modeling is to predict channel attributes (\(a_{k}(t)\), \(\tau_{k}(t)\), \(\Theta_{k}(t)\)) for a given environment map

\[h(t, \boldsymbol{\Theta}, \boldsymbol{\Phi})=\sum_{k} a_{k}(t) \delta\left(t-\tau_{k}(t)\right) \delta\left(\boldsymbol{\Theta}-\boldsymbol{\Theta}_{k}(t)\right) \delta\left(\boldsymbol{\Phi}-\mathbf{\Phi}_{k}(t)\right) \]

2. Renderer

2.1 Model

2.2 Dataset

3. Result

4. Summary

1. 没有考虑极化，使用全向天线
2. 考虑的场景较为简单
3. 没有单独考虑散射和衍射等更复杂的交互
4. 使用射线追踪仿真软件产生数据，没有使用实测数据

标签：Phi,right,tracing,boldsymbol,left,Theta,WINERT,NeRF
From： https://www.cnblogs.com/Tarjan-Zeng/p/18595179