Proceedings Article | 7 December 2011
KEYWORDS: Curium, Quantization, Analog electronics, Transistors, Binary data, Logic, Mirrors, Image processing, Optoelectronics, Neural networks
The paper considers results of designing and modeling analogue-digital converters (ADC) based on current mirrors for
the optical systems and neural networks with parallel inputs-outputs. Such ADC, named us multichannel analog-todigital
converters based on current mirrors (M ADC CM). Compared with usual converters, for example, reading, a bitby-
bit equilibration, and so forth, have a number of advantages: high speed and reliability, simplicity, small power
consumption, the big degree of integration in linear and matrix structures. The considered aspects of designing of
M_ADC CM in binary codes. Base digit cells (ABC) of such M_ADC CM, series-pipelined are connected in structures,
consist from 20-30 CMOS the transistors, one photodiode, have low (1,5-3,5) supply voltage, work in current modes
with the maximum values of currents (10-40)μA. Therefore such new principles of realization high-speed low-discharge
M_ADC CM have allowed, as have shown modeling experiments, to reach time of transformation less than 20-30 nS at
5-6 bits of a binary code and the general power 1-5 mW. The quantity easily cascadable ABC depends on wordlength
ADC, and makes n, and provides quantity of levels of quantization equal N=2n. Such simple enough on structure M
ADC CM, having low power consumption ≤ 3 ÷ 5mW, supply voltage (3-7)V, is provided at the same time with good
dynamic characteristics (frequency of digitization even for 1,5μm
or
0,35
μm- CMOS-technologies has made 40 MHz,
and can be increased 10 times) and accuracy (Δquantization 156,25nA
for
I
max10μA) characteristics is show. The
range can be transformed optical signals, taking into account sensitivity of modern photodetectors makes 20-200 μW in
such ADC. M_ADC CM open new prospects for realization linear and matrix (with picture operands) micro
photoelectronic structures which are necessary for neural networks, digital optoelectronic processors, neurofuzzy
controllers, and so forth.
