A serial distance calculation circuit for the application in artificial neural networks and pattern recognition
Marta Kolasa, Tomasz Talaśka, Rafal Tomasz Dlugosz · 2017
The paper presents a novel circuit for the calculation of Manhattan distance between two vectors of signals, suitable for various machine learning algorithms realized at the transistor level. In Self-Organizing Artificial Neural Networks, for example, one of the basic operations is the calculation of a distance between input learning patterns and vectors of neuron weights. In pattern recognition two patterns are being compared in a similar way. The proposed digital circuit is a serial solution, whose calculation scheme relies on the accumulation of absolute values of differences between corresponding components of both vectors. This allows to stop the calculations, if necessary, at any component of the compared vectors. The circuit is has been realized in the CMOS 130 nm technology and verified by means of postlayout simulations.