|
For Full-Text PDF, please login, if you are a member of IEICE,
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
|
Multicarrier Power Amplifier Linearization Based on Artificial Intelligent Methods
Masoud FAROKHI Mahmoud KAMAREI S. Hamaidreza JAMALI
Publication
IEICE TRANSACTIONS on Electronics
Vol.E88-C
No.4
pp.744-752 Publication Date: 2005/04/01 Online ISSN:
DOI: 10.1093/ietele/e88-c.4.744 Print ISSN: 0916-8516 Type of Manuscript: PAPER Category: Electronic Circuits Keyword: multi-carrier power amplifiers, neuro-fuzzy controller, neural network, perdistortion, power amplifier linearization, artificial intelligent techniques,
Full Text: PDF(623.4KB)>>
Summary:
This paper presents two new intelligent methods to linearize the Multi-Carrier Power Amplifiers (MCPA). One of the them is based on the Neuro-Fuzzy controller while the other uses two small neural networks as a polar predistorter. Neuro-Fuzzy controllers are not model based, and hence, have ability to control the nonlinear systems with undetermined parameters. Both methods are adaptive, low complex, and can be implemented in base-band part of the communication systems. The performance of the linearizers is obtained via simulation. The simulation is performed for three different scenarios; namely, a multi-carrier amplifier for GSM with four channels, a CDMA amplifier and a multi-carrier amplifier with two tones. The simulation results show that Neuro-Fuzzy Controller (NFC) and Neural Network Polar Predistorter (NNPP) have higher efficiencies so that reduce IMD3 by more than 42 and 32 dB, respectively. The practical implementation aspects of these methods are also discussed in this paper.
|
|
|