A hybrid heuristic design technique for real-time matching optimization for wearable near-field ambient RF energy harvesters

In this paper, a novel real-time active matching circuit design process based on preliminary measurements and a hybridization of a genetic algorithm and a data mining method is discussed. As a result, our proposed matching circuit can potentially have higher dc output power at 92.0 % and 69.6 %of potential load combinations with a maximum matching performance improvement of 21.4 % and 37.6 % compared to conventional matching methods using a dc-dc converter and a fixed passive matching circuit, respectively. After the reduction of matching circuit variable choices utilizing the clustering method, it is possible to achieve a satisfactory matching in practically very short times in the order of 1 ms.