New Multiplier and Input Layer in Current Mode for Analog Artificial Neural Networks

Thanks to FDSOI (Fully Depleted Silicon On Insulator) technology, this paper presents a new implementation of a multiplier and the input layer of an Analog ANN (Artificial Neural Network). The circuit architecture is designed around the Multi-Layer Perceptron (MLP) algorithm with back-propagation. Traditional analog implementations of this algorithm typically rely on multipliers, which consume substantial area and power. Additionally, this architecture faces challenges with weight storage. To overcome these limitations, we propose using advanced current mirrors as multipliers, leveraging the efficiency and benefits of FDSOI technology. Using a similar approach, we have realized the input layer using current mirrors without digital-to-analog converters, reducing both silicon area and power consumption. This dual reduction will eventually enable us to implement a much larger number of neurons, thus increasing the complexity of the final artificial neural network.