Benchmarking and Interpreting End-to-End Learning of MIMO and Multi-User Communication

Abstract

End-to-end autoencoder (AE) learning has the potential of exceeding the performance of human-engineered transceivers and encoding schemes, without a priori knowledge of communication-theoretic principles. In this work, we aim to understand to what extent and for which scenarios this claim holds true when comparing with fair benchmarks. Our particular focus is on memoryless multiple-input multiple-output (MIMO) and multi-user (MU) systems. Four case studies are considered: two point-to-point (closed-loop and open-loop MIMO) and two MU scenarios (MIMO broadcast and interference channels).For the point-to-point scenarios, we explain some of the performance gains observed in prior work through the selection of improved baseline schemes that include geometric shaping as well as bit and power allocation. For the MIMO broadcast channel, we demonstrate the feasibility of a novel AE method with centralized learning and decentralized execution. Interestingly, the learned scheme performs close to nonlinear vector-perturbation precoding and significantly outperforms conventional zero-forcing. Lastly, we highlight potential pitfalls when interpreting learned communication schemes. In particular, we show that the AE for the considered interference channel learns to avoid interference, albeit in a rotated reference frame. After de-rotating the learned signal constellation of each user, the resulting scheme corresponds to conventional time sharing with geometric shaping.

Bit Error and Block Error Rate Training for ML-Assisted Communication

“…Remark. A popular loss for learning end-to-end communication systems is the categorical CE (CCE) between the transmitted and guessed message [8][9][10]. By identifying messages with the blocks of a block code, the CCE can be seen as a loss that optimizes the BLER.…”

“…The availability of software frameworks, such as TensorFlow [2] and, recently, NVIDIA Sionna [3], has made implementation and training of MLassisted communication systems convenient. Existing results in ML-assisted communication systems range from the atomistic improvement of data detectors (e.g., using deep unfolding) [4][5][6][7] to model-free learning of end-to-end communication systems [8][9][10]. Quite surprisingly, only little attention has been devoted to the question of how ML-assisted communication systems should be trained.…”

“…1 and 3 that minimizing the coded BER is a category mistake 3. When we speak of the BCE between vectors, we mean the sum of binary CEs between the individual components as defined in (3), and not the categorical CE between the bit-vector and its estimate (as used, e.g., in[8][9][10]).…”