Behavioral Modeling And Predistortion Of Wideband Power Amplifier With Strong Nonlinearity

With the rapid development of mobile communication industry, the spectrum resource becomes more and more intense, in order to improve the utilization of spectrum, non constant envelope modulation modes are widely used in the modern communication systems, which put forward higher requirements of the linearity of power amplifiers (PA). In order to improve the efficiency of PAs, PAs should work in saturation state, which will make PAs show very serious nonlinearity. In addition, along with the bandwidth expansion of communication systems, the memory effect of PAs is also more and more obvious. In order to improve the communication quality, the nonlinearity and memory effect of PAs must be compensated. Digital predistortion technology has become the mainstream of the PA linearization techniques because of its excellent performance, easy realization, and low cost.This paper mainly studies the behavioral modeling and predistortion methods of PAs in wideband communication systems. The main works and innovation points are as follows.1. The principle and classification of predistortion technology are elaborated, the existing power amplifier behavior models are also introduced, and several criterions of behavior modeling and predistortion evaluation are introduced.2. This paper proposed an effective PMEC predistortion method of wideband PA with strong nonlinearity and memory effect. The PMEC modle was formed by the parallel connection of MP, envelope memory polynomial (EMP), and cross items between memory times (CIMT). The simulation and experimental results show that PMEC model achieves higher accuracy with less number of coefficients.3. Through deep analysis of the reason why traditional Volterra series model brings large number of coefficients, we reduced the quantity of memory time variable and proposed a new simplified Volterra series model. The experimental results show that the simplified Volterra series model has advantages in accuracy, linearity and complexity.4. Based on the depth analysis of the physical structure and function of PA, we simplified the equivalent discrete time model of PA using physical knowledge, and proposed the SPK model. The experimental results show that the SPK model achieves the best performance for different input signals or PAs.