A Committee Machine Neural Network for Dynamic and its Inverse Modeling of Distortions and Impairments in Wireless Transmitters

Abstract

The tradeoff between bandwidth efficiency and power efficiency with nonlinear distortion introduced by high power amplifier (HPA) in a modern communication system, at a high bit rate has become a prominent problem. Digital predistortion (DPD), the most cost-effective and yet lesser complex method is a widely used linearization method. In this method, the dynamic nonlinear characteristics of the PA are used to obtain an inverse transfer function to abolish the nonlinearity. The development of a good DPD technique depends on its ability to model an accurate PA behavior and its inverse behavior. Recently, several neural network-based predistortion linearizers had been proposed, and these linearizers are effectively proving their worth, but these methods report inadequate generalization performance in the presence of system imperfections like DC offset, IQ imbalance, and both. The present paper proposes a Mixture of Experts (MOE) based committee machine neural network for modeling the PA/transmitter system transfer function and its inverse transfer function under different system conditions. MOE models encompass a family of modular neural network architectures having several experts network and a single gating network, and targeting to divide a complex problem into simple subtasks. An algorithm based on a maximum likelihood criterion is considered for the MOE network training. The results obtained with the proposed model show good generalization performance with nonlinear HPA and with memory effect in the presence of system imperfections.

KEYWORDS:

• HPA nonlinearity
• Saleh model
• Class AB-PA
• Predistorter
• Linearization
• MOE

Additional information

Notes on contributors

Manoj Bhatt

Manoj Bhatt is currently pursuing PhD from Govind Ballabh Pant University of Agriculture & Technology, Pantnagar, Uttarakhand. He has done BTech from DBIT, Dehradun, India in 2010 and MTech (Digital Signal Processing) from Govind Ballabh Pant Institute of Engineering & Technology, Pauri Garhwal, Uttarakhand, India in 2012. His research interest includes communication systems, signal processing, predistortion techniques, and neural networks.

Meenakshi Rawat

Meenakshi Rawat is currently associate professor in IIT Roorkee, India. She received her PhD degrees from the University of Calgary, Calgary, AB, Canada, in 2012. From 2012 to 2013, she was a post-doctoral research fellow with the University of Calgary. From 2013 to 2014, she was a post-doctoral project researcher/scientist with The Ohio State University, Columbus, OH, USA. Dr Rawat was the workshop co-chair for the Automatic RF Techniques Group (ARFTG)−82 and the Session co-chair for millimeter-wave and terahertz designs for iMARC-2014, Bangalore, India. She is the chair for Women in microwave session in iMARC 2019 (IIT Bombay) and iMARC 2020 (held in IIT Kanpur). She is author of more than 75 publications in journals and conferences, two patents (applied), and one book chapter. Her area of expertise is software-defined radios and baseband/digital techniques for spectrum and power-efficient communication, filterless transmitter prototyping for ultra-wideband communication, digital/analog/hybrid predistortion techniques to mitigate distortion due to non-ideal RF components (power amplifier, mixers, multipliers, etc.) in low-cost transmitters, RF-DSP co-design techniques to reduce the interference in massive MIMO systems and baseband techniques for realizing mmWave communication using the commercially available platform. E-mail: [email protected]

Sanjay Mathur

Sanjay Mathur is professor in the Department of Electronics and Communication Engineering, Govind Ballabh Pant University of Agriculture and Technology, Pantnagar. He is presently head of the Department also. He received his BE degree from Madan Mohan Malaviya Engineering College (presently MMM University of Technology), Gorakhpur, India in 1985, ME degree from the University of Roorkee (presently IIT Roorkee), India in 1996, and PhD from Govind Ballabh Pant University of Agriculture and Technology, Pantnagar, India in 2009. His research interests include communication systems, signal processing, neural networks and smart antennas. He is a Fellow of the Institution of Engineers (India) and Life member of Indian Society of Technical Education. Dr Mathur is also chairman of Pantnagar Local Centre of The Institute of Engineers (India) for 2018-2021. E-mail: [email protected]