Int J Performability Eng ›› 2019, Vol. 15 ›› Issue (8): 2091-2099.

Models with Memory for RF Power Amplifier Behavioral Modeling

Xiang Chena, Yihan Xiaob,*, Hui Hana, Jie Changb, and Nan Tangb

1. a State Key Laboratory of Complex Electromagnetic Environment Effects on Electronics and Information System (CEMEE), Luoyang, 471003, China
b Harbin Engineering University, Harbin, 150001, China
• Submitted on  ;
• Contact: * E-mail address: xiaoyihan@hrbeu.edu.cn
• Supported by:
This work is supported by the National Natural Science Foundation of China (No. 61771154), the Fundamental Research Funds for the Central Universities (No. HEUCFG201830), and the State Key Laboratory of CEMEE (No. CEMEE2018K0104A).

Abstract: Power amplifiers are used to amplify signals to the required power and then transmit the signals through antennas. They are the key component of modern wireless communication systems. However, the power amplifier itself has non-linear characteristics and memory effect, especially when the input signal is a broadband signal and high frequency signal, which seriously affects the normal transmission of communication systems. In this paper, the RF power amplifier under test is an amplifier using a BLT53A transistor. We collect the input and output signals of the power amplifier for behavioral modeling. The behavior model only considers the input and output of the amplifier. We adopt a memory polynomial model based on the least squares method. In this paper, we evaluate the correctness of the model from many aspects including AM-AM curve, AM-PM curve, spectrum, gain compression, constellation diagram, and normalized mean square error. The results show that the model is effective.