A dynamic coefficient polynomial predistorter based on direct learning architecture is proposed.Compared to the existing polynomial predistorter,on the one hand,the proposed predistorter based on thedirect learning architecture is more robust to initial conditions of the tap coefficients than that based on in-direct learning architecture;on the other hand,by using two polynomial coefficient combinations,differ-ent polynomial coefficient combination can be selected when the input signal amplitude changes,whicheffectively decreases the estimate error.This paper introduces the direct learning architecture and givesthe dynamic coefficient polynomial expression.A simplified nonlinear recursive least-squares(RLS)algo-rithm for polynomial coefficient estimation is also derived in detail.Computer simulations show that theproposed predistorter can attain 31 dB,28dB and 40dB spectrum suppression gain when our method is ap-plied to the traveling wave tube amplifier(TWTA),solid state power amplifier(SSPA)and polynomialpower amplifier(PA)model,respectively.
In memory polynomial predistorter design,the coefficient estimation algorithm based on normalized least mean square is sensitive to initialization parameters.A predistorter based on generalized normalized gradient descent algorithm is proposed.The merit of the GNGD algorithm is that its learning rate provides compensation for the independent assumptions in the derivation of NLMS,thus its stability is improved.Computer simulation shows that the proposed predistorter is very robust.It can overcome the sensitivity of initialization parameters and get a better linearization performance.
