Title :
Non-Binary Analog-to-Digital Converter Based on Amoeba-Inspired Neural Network
Author :
Ishida, Uichi ; Yamazaki, Yusuke ; Waho, Takao
Author_Institution :
Dept. of Inf. & Commun. Sci., Sophia Univ., Tokyo, Japan
Abstract :
An analog-to-digital converter (ADC) based on neural networks is proposed, and the feasibility of using no binary coding is discussed with circuit simulation. An amoeba-inspired computing technique is used to construct the present ADC, where switched-capacitor circuits are used as unit neurons. Dummy units are also added to improve the stability of circuit operation. For an ADC with a radix of 2, large quantization errors were observed due to the local minima. It was found that introducing a radix smaller than 2 effectively reduced the quantization error. Low-power operation can be expected by using a dynamic analog circuit technique in the present neuro-ADC.
Keywords :
circuit simulation; circuit stability; low-power electronics; neural chips; quantisation (signal); switched current circuits; amoeba-inspired computing technique; amoeba-inspired neural network; circuit operation stability improvement; circuit simulation; dummy units; dynamic analog circuit technique; local minima; low-power operation; neural networks; neuro-ADC; nonbinary analog-to-digital converter; nonbinary coding; quantization error reduction; radix; switched-capacitor circuits; unit neurons; Analog-digital conversion; Biological neural networks; Circuits and systems; Feedback loop; Neurons; Quantization (signal); analog-to-digital converter; neural network; non-binary;
Conference_Titel :
Multiple-Valued Logic (ISMVL), 2015 IEEE International Symposium on
Conference_Location :
Waterloo, ON
DOI :
10.1109/ISMVL.2015.13
