End-to-end Waveform Learning Through Joint Optimization of Pulse and Constellation Shaping

TL;DR

This paper introduces an end-to-end waveform design method that jointly optimizes pulse shaping and constellation geometry using neural networks, improving spectral efficiency and reducing interference without increasing transmitter complexity.

Contribution

It proposes a novel joint learning framework for waveform design that enhances spectral efficiency and outperforms traditional methods in spectral leakage and power ratios.

Findings

Achieves significantly lower adjacent channel leakage ratios (ACLRs).

Maintains competitive peak-to-average power ratios (PAPRs).

Preserves information rate on AWGN channels.

Abstract

As communication systems are foreseen to enable new services such as joint communication and sensing and utilize parts of the sub-THz spectrum, the design of novel waveforms that can support these emerging applications becomes increasingly challenging. We present in this work an end-to-end learning approach to design waveforms through joint learning of pulse shaping and constellation geometry, together with a neural network (NN)-based receiver. Optimization is performed to maximize an achievable information rate, while satisfying constraints on out-of-band emission and power envelope. Our results show that the proposed approach enables up to orders of magnitude smaller adjacent channel leakage ratios (ACLRs) with peak-to-average power ratios (PAPRs) competitive with traditional filters, without significant loss of information rate on an additive white Gaussian noise (AWGN) channel, and no additional complexity at the transmitter.