Open Access2D curtailed harmonic memory polynomial for reduced complexity in concurrent dual‐band modelling and digital predistortion with the second band at harmonic frequency
Author(s) -
Jaraut Praveen,
Rawat Meenakshi,
Ghannouchi Fadhel M.
Publication year - 2018
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2017.1376
Subject(s) - predistortion , intermodulation , amplifier , computer science , computational complexity theory , harmonic , modulation (music) , electronic engineering , interference (communication) , signal (programming language) , wideband , algorithm , telecommunications , physics , bandwidth (computing) , acoustics , engineering , channel (broadcasting) , programming language
Multi‐band transmitter systems are evolving to support the smooth transition from 4G to 5G communication systems. Moreover, recent developments of multi‐band and ultra‐wideband power amplifiers have led to a possible scenario where the second carrier signal is transmitted at the harmonic frequency of the first carrier signal. This results in harmonic interference from the first carrier signal as well as additional cross‐modulation and intermodulation distortion (IMD) components, which cannot be filtered out. The computational and memory requirements for digital predistortion (DPD) in such scenario increase drastically to include all interference terms. This study presents a novel two‐dimensional curtailed harmonic memory polynomial (2D‐CHMP) model to capture harmonic interferences, cross‐modulation and IMDs. The model complexity and memory requirement of 2D‐CHMP are very less as compared to the state‐of‐the‐art two‐dimensional harmonic memory polynomial (2D‐HMP) model. For proof‐of‐concept, it is shown with two different measurement setups that the proposed 2D‐CHMP DPD provides similar linearisation performances as compared to the 2D‐HMP DPD with less number of coefficients and computational complexity. As a study, it is shown that the proposed model can be further adapted to a low‐precision (low‐bit) environment by utilising principal component analysis.