Production test-based classification of antennas using the feed-forward neural network

This paper deals with application of a feed-forward neural network for the production test of miniature antennas, where different 6-wire antennas with a ferrite core were tested. These devices under test were driven by a MOS H-bridge integrated in a dedicated ASIC controlled by non-overlapping clock signals. Tested devices were measured and seven significant parameters were evaluated and classified using the feed-forward neural network. The neural network was trained on a training data set that consisted of a uniform number of good and faulty antennas. Neural network training was realized through the error backpropagation algorithm.