Recurrent Neural Equalizer Research Articles

Parameter Estimation of Recurrent Neural Equalizers Using the Derivative-Free Kalman Filter

For the last decade, recurrent neural networks (RNNs) have been commonly applied to communications channel equalization. The major problems of gradient-based learning techniques, employed to train recurrent neural networks are slow convergence rates and long training sequences. In high-speed communications system, short training symbols and fast convergence speed are essentially required. In this paper, the derivative-free Kalman filter, so called the unscented Kalman filter (UKF), for training a fully connected RNN is presented in a state-space formulation of the system. The main features of the proposed recurrent neural equalizer are fast convergence speed and good performance using relatively short training symbols without the derivative computation. Through experiments of nonlinear channel equalization, the performance of the RNN with a derivative-free Kalman filter is evaluated.

Decision Feedback Recurrent Neural Equalization With Fast Convergence Rate

Real-time recurrent learning (RTRL), commonly employed for training a fully connected recurrent neural network (RNN), has a drawback of slow convergence rate. In the light of this deficiency, a decision feedback recurrent neural equalizer (DFRNE) using the RTRL requires long training sequences to achieve good performance. In this paper, extended Kalman filter (EKF) algorithms based on the RTRL for the DFRNE are presented in state-space formulation of the system, in particular for complex-valued signal processing. The main features of global EKF and decoupled EKF algorithms are fast convergence and good tracking performance. Through nonlinear channel equalization, performance of the DFRNE with the EKF algorithms is evaluated and compared with that of the DFRNE with the RTRL.

Kalman Filter-Trained Recurrent Neural Equalizers for Time-Varying Channels

Recurrent neural networks (RNNs) have been successfully applied to communications channel equalization because of their modeling capability for nonlinear dynamic systems. Major problems of gradient-descent learning techniques commonly employed to train RNNs are slow convergence rates and long training sequences required for satisfactory performance. This paper presents decision-feedback equalizers using an RNN trained with Kalman filtering algorithms. The main features of the proposed recurrent neural equalizers, using the extended Kalman filter (EKF) and unscented Kalman filter (UKF), are fast convergence and good performance using relatively short training symbols. Experimental results for various time-varying channels are presented to evaluate the performance of the proposed approaches over a conventional recurrent neural equalizer.

칼만필터로 훈련되는 순환신경망을 이용한 시변채널 등화

Recurrent neural networks have been successfully applied to communications channel equalization. Major disadvantages of gradient-based learning algorithms commonly employed to train recurrent neural networks are slow convergence rates and long training sequences required for satisfactory performance. In a high-speed communications system, fast convergence speed and short training symbols are essential. We propose decision feedback equalizers using a recurrent neural network trained with Kalman filtering algorithms. The main features of the proposed recurrent neural equalizers, utilizing extended Kalman filter (EKF) and unscented Kalman filter (UKF), are fast convergence rates and good performance using relatively short training symbols. Experimental results for two time-varying channels are presented to evaluate the performance of the proposed approaches over a conventional recurrent neural equalizer.

Recurrent neural network adaptive equalizers based on data communication

In this paper, a decision feedback recurrent neural network equalizer and a modified real time recurrent learning algorithm are proposed, and an adaptive adjusting of the learning step is also brought forward. Then, a complex case is considered. A decision feedback complex recurrent neural network equalizer and a modified complex real time recurrent learning algorithm are proposed. Moreover, weights of decision feedback recurrent neural network equalizer under burst-interference conditions are analyzed, and two anti-burst-interference algorithms to prevent equalizer from out of working are presented, which are applied to both real and complex cases. The performance of the recurrent neural network equalizer is analyzed based on numerical results.

ON MODIFIED COMPLEX RECURRENT NEURAL NETWORK ADAPTIVE EQUALIZER

A new modified version called decision feedback complex recurrent neural network equalizer (DFCRNNE) is proposed with the study of complex recurrent neural network equalizer (CRNNE). Based on DFCRNNE, a modified real time recurrent learning (CRTRL) algorithm is developed. Simulation results show that DFCRNNE has better performance than CRNNE based on traditional CRTRL algorithm2 in complex nonlinear channels with severe intersymbol interference (ISI) and nonlinear distortion.

Comments on "A decision feedback recurrent neural equalizer as an infinite impulse response filter"

This correspondence corrects and clarifies some notation inconsistencies and the absence of some definitions in a previous paper by Ong et al. (see ibid., vol. 45, p. 2851-8, Nov. 1997) In addition, we show that the assumptions made in the derivation of the MSE/sup DFRNE/-MSE/sup RNE/ invalidate the final result given in Ong et al. and lead to an unfair comparison between the two equalizers. We present a simulation that best compares the two equalizers.

A decision feedback recurrent neural equalizer as an infinite impulse response filter

An adaptive decision feedback recurrent neural equalizer (DFRNE), which models a kind of an IIR structure, is proposed. Its performance is compared with the traditional linear and nonlinear equalizers with FIR structures for various communication channels. The small size and high performance of the DFRNE makes it suitable for high-speed channel equalization.

A decision feedback recurrent neural equalizer for digital communication

An important problem in high density digital magnetic recording system is that of channel equalization, that is, removal of distortions introduced by linear or nonlinear message corrupting mechanisms in the reconstruction of the original symbols. Severe nonlinear distortions in high density digital magnetic recording systems can make it difficult for conventional equalizers to reconstruct the originally recorded symbols. In this paper, we propose a Decision Feedback Recurrent Neural Equalizer (DFRNE) with a simple structure, which can recover the original symbols correctly under severe nonlinear distortion.