kaira.channels.BaseChannel
Inheritance diagram for BaseChannel
- class kaira.channels.BaseChannel(*args: Any, **kwargs: Any)[source]
-
Base abstract class for communication channel models.
In communications theory, a channel refers to the medium through which information is transmitted from a sender to a receiver. This class provides a foundation for implementing various channel models that simulate real-world effects like noise, fading, distortion, and interference.
All channel implementations should inherit from this base class and implement the forward method, which applies the channel effects to the input signal.
Channel models are implemented as PyTorch modules, allowing them to be: - Used in computational graphs - Combined with neural networks - Run on GPUs when available - Included in larger end-to-end communications system models
Methods
Initialize the base channel.
Transform input signal according to channel characteristics.
Get a dictionary of the channel's configuration.
Examples using
kaira.channels.BaseChannel
Modulation Schemes for Digital Communication Systems
Modulation Schemes for Digital Communication Systems
Sequential Model for Modular Neural Network Design
Sequential Model for Modular Neural Network Design- __init__(*args: Any, **kwargs: Any)[source]
Initialize the base channel.
- Parameters:
*args – Variable length argument list.
**kwargs – Arbitrary keyword arguments.
- abstract forward(x: Tensor, *args: Any, **kwargs: Any) Tensor[source]
Transform input signal according to channel characteristics.
This method defines how the channel transforms an input signal, which may include adding noise, applying fading, introducing hardware impairments, or other effects specific to the channel model.
- Parameters:
x (torch.Tensor) – The input signal.
*args – Additional positional arguments.
**kwargs – Additional keyword arguments.
- Returns:
The output signal after passing through the channel.
- Return type:
