Advanced neural network receiver design to combat multiple channel impairments

In communication systems, the channel noise is usually assumed to be white and Gaussian distributed. Therefore, an optimum receiver structure designed for the additive white Gaussian noise (AWGN) channel is employed in applications. However, in wireless communication systems, noise is often caused by strong interferences. Moreover, there are other effects such as phase offset that degrade the performance of the receiver. Designing the optimum receiver for different channel models is difficult and not reasonable because channel model and channel statistics are not known at the receiver. In this paper, we propose a neural network-based approach to demodulate the transmitted signal over unknown channels. Naturally, the collection of the training data, design and training of the neural network, and finally reconfiguration of the system according to the designed neural network are implemented on software-defined digital signal processing facilities. In particular, we show that the proposed receiver is capable of jointly canceling the strong interferences and phase offset. Simulation results in various signal environments are presented to illustrate the performance of the proposed system. It is shown that the proposed approach has the same performance as the correlation demodulator structure for AWGN channels, while it has a clear advantage for unknown channel models. Moreover, it is shown that the neural network-based receiver may be used for channel estimation and equalization over Rayleigh channels. Numerical results indicate that the performance of the proposed receiver is very close to the Rayleigh theoretical bound.

PDF

___

[1] Ulversoy T. Software defined radio: challenges and opportunities. IEEE Commun Surv Tut 2010; 12: 531-550.
[2] Luiz Garcia Reis A, Barros AF, Gusso Lenzi K, Pedroso Meloni LG, Barbin SE. Introduction to the software-defined radio approach. IEEE Lat Am T 2012; 10: 1156-1161.
[3] Amini M, Balarastaghi E. Universal neural network demodulator for software defined radio. International Journal of Machine Learning and Computing 2011; 1: 305-310.
[4] Zhang W, Yao D, Yang M. Implementation of LTE-R transceiver and the performance with WINNER D2a channel model. In: IEEE 5th International Symposium on Microwave, Antenna, Propagation and EMC Technologies for Wireless Communications; 2931 October 2013; Chengdu, China. New York, NY, USA: IEEE. pp. 704-708.
[5] Patnaik AA, Dimitrios E, Mishra RK, Lyke JC. Applications of neural networks in wireless communications. IEEE Antenn Propag M 2004; 46: 130-137.
[6] Dalkılı¸c TE, HancıBY, Apaydın A. Fuzzy adaptive neural network approach to path loss prediction in urban areas at GSM-900 band. Turk J Electr Eng Co 2010; 18: 1077-1094.
[7] T¨urker N, G¨une¸s F, Yıldırım T. Artificial neural design of microstrip antennas. Turk J Electr Eng Co 2006; 14: 445-453.
[8] Ibnkahla M, Yuan J. A neural network MLSE receiver based on natural gradient descent: application to satellite communications. In: IEEE 2003 International Symposium on Signal Processing and its Applications; 14 July 2003; Paris, France. New York, NY, USA: IEEE. pp. 33-36.
[9] Ibnkahla M, Yuan J. A neural network MLSE receiver based on natural gradient descent: application to satellite communications. EURASIP J Appl Sig P 2004; 16: 2580-2591.
[10] Sacchi C, Gera G, Regazzoni CS. Neural network-based techniques for efficient detection of variable-bit-rate signals in MC-CDMA systems working over LEO satellite networks. Signal Process 2005; 85: 505-522.
[11] Veciana GD, Zakhor A. Neural net-based continuous phase modulation receivers. IEEE T Commun 1992; 40: 1396- 1408.
[12] Nakayama K, Imai K. Neural demodulator for amplitude shift keying signals. In: IEEE 1994 World Congress on Computational Intelligence; 27 June2 July 1994; Orlando, FL, USA. New York, NY, USA: IEEE. pp. 3909-3914.
[13] Ohnishi K, Nakayama K. Neural demodulator for quadrature amplitude modulation signals. In: IEEE 1996 International Conference on Neural Networks; 36 June 1996; Washington, DC, USA. New York, NY, USA: IEEE. pp.1933-1938.
[14] Li M, Zhong HS, Li M. Neural network demodulator for frequency shift keying. In: IEEE 2008 International Conference on Computer Science and Software Engineering; 1214 December 2008; Wuhan, China. New York, NY, USA: IEEE. pp. 843-846.
[15] Amini MR, Balarastaghi E. Improving ANN BFSK demodulator performance with training data sequence sent by transmitter. In: IEEE 2010 Second International Conference on Machine Learning and Computing; 911 February 2010; Bangalore, India. New York, NY, USA: IEEE. pp. 276-281.
[16] Proakis J, Salehi M. Digital Communications. 5th ed. New York, NY, USA: McGraw-Hill Science, 2007.
[17] Gomes HC, Carvalho NB. Interference cancellation: new configuration technique for cancellation of strong interferences from adjacent frequency bands. In: IEEE 2008 Workshop on Integrated Nonlinear Microwave and Millimetre-Wave Circuits; 2425 November 2008; Malaga, Spain. New York, NY, USA: IEEE. pp. 65-68.
[18] Altınkaya MA, Sankur B, Anarım E. Performance of prefiltered model-based frequency estimators. Turk J Electr Eng Co 1998; 6: 23-35.
[19] Imad R, Houcke S, Ghogho M. Effective techniques for blind phase offset estimation in LDPC coded systems. In: IEEE 2009 Workshop on Signal Processing Advances in Wireless Communications; 2124 June 2009; Perugia, Italy. New York, NY, USA: IEEE. pp. 638-642.
[20] Onder M, Akan A, Dogan H. Neural network based receiver design for software defined radio over unknown channels. In: IEEE 2013 International Conference on Electrical and Electronics Engineering; 2830 November 2013; Bursa, Turkey. New York, NY, USA: IEEE. pp. 297-300.
[21] Seyman MN, Ta¸spınar N. Channel estimation based on neural network in space time block coded MIMO-OFDM System. Digit Signal Process 2013; 23: 275-280.