Model-Driven Deep Learning for MIMO Detection

Abstract

In this paper, we investigate the model-driven deep learning (DL) for joint MIMO channel estimation and signal detection (JCESD), where signal detection considers channel estimation error and channel statistics while channel estimation is refined by detected data and takes the signal detection error into consideration. In particular, the MIMO signal detector is specially designed by unfolding an iterative algorithm and adding some trainable parameters. Since the number of trainable parameters is much fewer than the data-driven DL based signal detector, the model-driven DL based MIMO signal detector can be rapidly trained with a much smaller data set. Furthermore, the proposed signal detector can be extended to soft-input soft-output detection easily. Based on numerical results, the model-driven DL based JCESD scheme significantly improves the performance of the corresponding traditional iterative detector and the signal detector exhibits superior robustness to signal-to-noise ratio (SNR) and channel correlation mismatches. Recently, it has been applied in physical layer communications [10]-[12], such as channel estimation [13]-[15], CSI feedback [16], signal detection [17]-[25], and channel coding [26], [27]. In particular, a five-layer fully connected deep neural network (DNN) is embedded into an orthogonal frequency-division multiplexing (OFDM) system for joint channel estimation and signal detection (JCESD) by treating the receiver as a black box and without exploiting domain knowledge [17]. However, training such a black-box-based network requires a lot of 1 A matrix A = UΣV is unitarily-invariant if U, Σ and V are mutually independent, and U, V are Haar-distributed. The independent and identically distributed (i.i.d.) Gaussian matrix is a typical unitarily-invariant matrix.

Atrial Fibrillation Identification With PPG Signals Using a Combination of Time-Frequency Analysis and Deep Learning

“…In recent years, with the rapid development of hardware computing power, deep learning technology has been applied to natural speech recognition, image processing, medical devices and many other fields [32]- [35]. In this study, we propose a new method of automatic identification of AF based on PPG.…”

Deep Soft Interference Cancellation for MIMO Detection

“…However, these methods assume a linear channel model of the form (2). Furthermore, these previous receivers typically require CSI, obtained either from a-priori knowledge or via channel estimation as in [34]. Unlike these previous receivers, DeepSIC, which is also based on a model-based algorithm, is independent of the channel model, and can efficiently learn to detect in a wide variety of channel conditions, ranging from linear Gaussian channels to non-linear Poisson channels, as numerically demonstrated in Section IV.…”

“…However, these previously proposed receivers all assume a linear channel with Gaussian noise, in which CSI is either available [30]- [33], [35] or estimated from pilots [34]. Consequently, these methods thus do not capture the potential of ML in being model independent, and are applicable only under specific channel setups.…”

Complex Trainable Ista for Linear and Nonlinear Inverse Problems

“…The first update rule (11) is a gradient step using the Wirtinger derivative whose step size is a trainable parameter β t (> 0). In the second update rule (12) called a shrinkage step, an estimate is updated by a trainable divergence free (DF)-like function [21,22] with shrinkage function η(·) to reflect prior information on x and trainable parameters {λ t , C t , D t }.…”