Behavioral Modeling and Predistortion of Dual-Band Power Amplifiers

Dual-band power amplifiers (PAs) are essential components in modern wireless communication systems. Compared to single band PAs, dual-band PAs enable efficient utilization of spectrum resources with lower capital and operational expenses. In modern applications, inherent nonlinearities in PAs are unavoidable and can significantly degrade signal quality causing significant spectrum regrowth. These nonlinearities, already problematic in the case of single band systems, become considerably worse in dual-band systems with the generation of a plethora of intermodulation and cross-modulation products. These are even more challenging since the dominant distortions products can vary depending on the carrier aggregation scenario and, most importantly, the spacing between the two bands. Digital predistortion (DPD) techniques are mature for single band PAs; however, there are still numerous challenges for their adoption in dual-band and multi-band systems. This paper provides a comprehensive review of dual-band digital predistorters, including polynomial based, hybrid, and neural networks based predistorters. The different models are presented using consistent formulations to better highlight their commonalities and differences. Finally, a discussion of the presented models is included along with some insight about potential areas of improvement. This review aims to provide a valuable resource for researchers and engineers working on the design and optimization of dual-band communication systems.