A neural network MLSE receiver based on natural gradient descent: application to satellite communications

Abstract

The paper proposes a maximum likelihood sequence estimator (MLSE) receiver for satellite communications. The satellite channel model is composed of a nonlinear traveling wave tube (TWT) amplifier followed by a multipath propagation channel. The receiver is composed of a neural network channel estimator (NNCE) and a Viterbi detector. The natural gradient (NG) descent is used for training. Computer simulations show that the performance of our receiver is close to the ideal MLSE receiver in which the channel is perfectly known.

Advanced~neural network receiver design~to combat multiple channel~impairments

“…Neural networks can be used to solve different types of problems with their remarkable ability to derive meaning from complicated or imprecise data [6,7]. Neural networks (NNs) play important roles in many engineering areas such as control, biomedical, electronics engineering, and recently communication engineering areas [8]. They are generally used to approximate unknown nonlinear functions by using their universal approximation, learning, and adaptation abilities.…”

“…They are generally used to approximate unknown nonlinear functions by using their universal approximation, learning, and adaptation abilities. Active research has been done in NNs for communication systems [9,10] and several NN approaches have been proposed to design receivers [11][12][13][14][15]. In [11], despite using a large number of hidden neurons, the proposed NND (neural network demodulator) is shown to be suboptimal.…”

A Novel Efficient Cluster-Based MLSE Equalizer for Satellite Communication Channels with-QAM Signaling

“…3 The method exploits the nature of the TWT nonlinearity (dependence only on the modulus of the input signal) and appeals to the methodology of the 1D CBSE in order to provide a computationally cheap estimate of the cluster centers. Furthermore, the required training sequence is very short, compared to other previously used techniques (e.g., [31]).…”

“…The pulse shaping filter before the memoryless nonlinearity of the HPA is a square root raised cosine (SRRC) filter of sufficient bandwidth compared to the signal bandwidth. Therefore, ISI is introduced only by filters following the nonlinearity [11,31]. The adopted signaling scheme is the rectangular M-QAM.…”

Ανάπτυξη Μοντελοποίηση Και Προσομοίωση Προηγμένων Τεχνικών Εκτίμησης Και Βελτίωσης Της Επίδοσης Δορυφορικών Τηλεπικοινωνιακών Συστημάτων Σε Μη Γραμμικούς Διαύλους

“…Ôï øçöéáêü óÞìá ðïõ ìåôáäßäåôáé áðïôåëåß Ýíá ñåýìá äåäïìÝíùí u+jv, áíåîÜñôçôùí êáé éäáíéêÜ êáôáíåìçìÝíùí (i.i.d.). Ôï ößëôñï äéáìüñöùóçò ðáëìïý ðñéí ôç äß÷ùò ìíÞìç ìç ãñáììéêüôçôá ôïõ HPA åßíáé Ýíá ößëôñï ôåôñáãùíéêÞò ñßaeáò áíõøùìÝíïõ óõíçìéôüíïõ (Square Root Raised Cosine, SRRC) ìå åýñïò aeþíçò ìåãáëýôåñï áðü áõôü ôïõ óÞìáôïò.¸ôóé, ç ISI åéóÜãåôáé ìüíï áðü ôá ößëôñá ðïõ áêïëïõèïýí ôç ìç ãñáììéêüôçôá[30,48]. Ç óçìáôïäïóßá ãßíåôáé ìå äéáìüñöùóç ôåôñáãùíéêïý M-QAM.…”