Adaptive digital predistortion based on MC-FQRD-RLS algorithm using indirect learning architecture

Abstract

Multichannel fast QR decomposition recursive least-squares (MC-FQRD-RLS) algorithms are well known for their good numerical properties and low computational complexity. There are tow distinct ways to obtain the expended input vector, sequential-type method, and block-type one. The latter one, despite exhibiting a higher computational burden as compared to the former one, has some attractive features, e.g., suitability for parallel implementation. Predistortion techniques for linearizing power amplifier (PA) nonlinearity with indirect learning architecture (IDLA) are widely used. The benefit of the IDLA leaves unnecessary the assumption of a model for PA, corresponding parameters estimation and inverse construction. In this paper we present a new technique for predistortion using block-type MC-FQRD-RLS algorithm with IDLA, in which the predistorter is constructed by simplified volterra model. Simulations results verify that the proposed techniques have good convergence property.