The offered algorithms represent a basic set of methods for digital linearization of power amplifiers using the digital predistortion method. They are algorithms working on the principle of feedback system. Based on the feedback information, the inversion characteristic is then calculated, thereby reducing the nonlinear effects of the final stage on the transmitted signal.

Description of the technology

To transmit a radio signal, modern radio transmitters need a power amplifier. However, such devices are often non-linear, which causes distortion of the transmitted signal, hence more difficult reception and, above all, radiation into the surrounding frequency channels.

Therefore, the researchers of doc. Ing. Roman Maršálek, Ph.D. and Ing. Tomáš Gotthans, Ph.D. from the Faculty of Electrical Engineering and Communication Technologies have developed a set of algorithms that largely solves this problem. The essence of the solution lies in the use of baseband signal processing methods, where nonlinearities caused by the power amplifier are compensated by a digital system with adaptation, using feedback from the amplifier output. This results in a significant reduction of effects such as side-channel radiation or distortion in the transmitted band. The set of algorithms also includes routines for calculating the coefficients of the predistortion system in different ways (LS, LMS, RLS, Damped Newton method, QR decomposition,...).