Analog VLSI and neural systems
Analog VLSI and neural systems
Analog VLSI signal processing: why, where, and how?
Analog Integrated Circuits and Signal Processing - Joint special issue on analog VLSI computation
An Analogue Electronic Model of Ventral Cochlear Nucleus Neurons
MICRONEURO '96 Proceedings of the 5th International Conference on Microelectronics for Neural Networks and Fuzzy Systems
A Neuromorphic Sound Localizer for a Smart MEMS System
Analog Integrated Circuits and Signal Processing
Temporal order detection and coding in nervous systems
Neural Computation
| Hi-index | 0.00 |
This paper presents an electronic system thatextracts the periodicity of a sound. It uses threeanalog VLSI building blocks: a silicon cochlea, twoInner Hair Cell circuits and two spiking neuron chips.The silicon cochlea consists of a cascade of filterswhich delays (and filters) the input sound as it passesalong the cascade. The time delay added by each individual filter in the cochlea increasesexponentially with position of the filter along thecochlea. The frequency for which the time delaybetween two outputs that are n-stages apart correspondsto a phase delay of 2π therefore decreases exponentially along the cochlea. In the systempresented in this paper we create spike trains from theoutput of the cochlear filters and we compare theoutput of each filter with the output of a filter foursections earlier in the cascade. If the signalfrequency corresponds to the inverse of the time delaybetween these two filters, then the two spikes in the spike trains created at these two outputs willcoincide. Detecting these coincidences can yield veryselective filters, i.e., filters that respond only toa very narrow range of periodicities, but that at thesame time still respond after a few periods of theinput signal. This is an advantage over traditionalband-pass filters, where an increase in selectivityhas to be traded off against decrease in responsetime.