An enhanced active load-pull algorithm for faster convergence

This paper proposes a new active load-pull equation and discusses the use of behavioural models to add intelligence to an active, open loop load-pull algorithm. This improves the prediction of the required injected signal conditions to achieve a target load, minimising the iterations required and improving convergence reliability. The results were validated by performing load-pull measurements on an amplifier at 2.14 GHz using the Focus Microwaves Rapid-VT load-pull system. A reduction of 47% in the average number of iterations required was found when using a linear model based load-pull algorithm compared to a simple weighted algorithm. A further reduction was achieved in the compression region by increasing the behavioural model complexity to include the non-linear 3rd order conjugate term