本页翻译不是最新的。点击此处可查看最新英文版本。

AI 功能

支持无线应用深度学习工作流的功能

您可以在包含神经网络的无线通信系统中使用 Deep Learning Toolbox™ 功能。训练期间，在神经网络中使用函数需要 dlarray (Deep Learning Toolbox) 支持。有关支持 dlarray (Deep Learning Toolbox) 对象的无线通信领域特定函数集，请参阅Wireless Communications (Deep Learning Toolbox)。

函数

dlarray 支持

`awgn`	Add white Gaussian noise to signal
`bit2int`	Convert bits to integers (自 R2021b 起)
`genqammod`	General quadrature amplitude modulation (QAM)
`ofdmmod`	Modulate using OFDM method
`ofdmdemod`	Demodulate using OFDM method
`ofdmChannelResponse`	Calculate OFDM channel response (自 R2023a 起)
`ofdmEqualize`	Equalize OFDM signals (自 R2022b 起)

对象

dlarray 支持

`comm.ChannelFilter`	Filter signal using multipath gains at specified path delays
`comm.MIMOChannel`	Filter input signal through MIMO multipath fading channel

数据导入和预处理

`dpdPreprocessor`	Digital predistortion (DPD) data input preprocessor for AI algorithms (自 R2024b 起)
`signalDatastore`	Datastore for collection of signals

主题

Cosine Similarity As a Channel Estimate Quality Metric
Use the cosine similarity metric to compare two vectors. (自 R2024b 起)
Normalized Mean Squared Error as a Distance Measure
Use the normalized mean squared error (NMSE) as a loss function for training a neural network in a wireless communications application. (自 R2025a 起)

相关信息

使用 MATLAB Coder 进行深度学习的前提条件 (MATLAB Coder)
Domain-Specific Functions with dlarray Support (Deep Learning Toolbox)
Datastores for Deep Learning (Deep Learning Toolbox)