Digital Pre Distortion Linearization of Power Amplifier using Neural Network

Power amplifiers lie at the front end of most radio frequency systems and are critical in ensuring the appropriate range of wireless systems. The persistent search of enhanced efficiency and linearity in power amplifiers has become a pivotal focus in modern wireless communication systems, driven by the escalating demands for higher data rates and improved signal quality. Digital pre-distortion has emerged as a well-known technique to address the inherent nonlinearities exhibited by power amplifiers, Digital pre-distortion has shown to be advantageous in systems employing higher order modulation formats. Digital pre-distortion enables power amplifiers to operate closer to their saturation point, thereby maximizing power efficiency without sacrificing signal integrity. In this paper, neural network technique has been used to model a digital pre-distorter for linearization of power amplifier used in 5G communication system. To compare the performance of proposed digital pre-distorter, cross-term memory polynomial based digital pre-distorter has also been modeled. Results show that proposed digital pre-distorter modeled using neural network has better performance. The paper is organized as follows: section I presents introduction, concept of digital pre-distortion is presented in section II, section III is concept of augmented real-valued time-delay neural network, proposed power amplifier linearization technique is presented in section IV and section V presents conclusion.