Name
Title | ||
|---|---|---|
Adaptive Digital Predistortion Based on Hybrid Indirect Learning Structure with Variable Step-Size for Wideband HPA. |
Abstract | ||
|---|---|---|
Adaptive digital predistortion is one of the most promising linearization technique, which leads to more efficient and cost-effective high power amplifier (HPA). In this paper, a fast-converging variable step-size least mean square algorithm based on hybrid indirect learning structure is presented to provide an unbiased predistortion identification in the non-stationary signal condition for wideband HPA. By using a variable step-size strategy, the proposed identifying algorithm can adapt its learning rate according to the dynamic range of input signal. Thus, more stable learning convergence and lower steady-state mis-adjustment error are obtained simultaneously. A comprehensive theoretical analysis is presented, and some important results including learning stability, convergence behavior, and selection criteria for initialization parameter setting are derived. Both its superiority and robustness are well confirmed through extensive numerical simulations. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1007/s11277-017-4338-5 | Wireless Personal Communications |
Keywords | Field | DocType |
High power amplifier (HPA),Digital predistortion (DPD),Indirect learning structure,Variable step-size | Convergence (routing),Wideband,Dynamic range,Control theory,Computer science,Robustness (computer science),Initialization,Predistortion,Linearization,Amplifier | Journal |
Volume | Issue | ISSN |
96 | 2 | 0929-6212 |
Citations | PageRank | References |
0 | 0.34 | 8 |
Authors | ||
2 | ||