Abstract
In this paper we explore a combination of above-and-subthreshold CMOS circuit techniques for the implementation of analog neuromorphic network processing. The implemented network embodies a blind signal separation algorithm performing an independent component analysis. It is essentially a continuous time recursive linear adaptive filter that uses a non-linear adaptation rule. Analog I/O interface, weight coefficients and adaptation blocks are all integrated on the chip. A test 2-neuron-2-synapse network as well as a small 5-neuron-20-synapse network were fabricated in a 2 micron n-well double-polysilicon, double-metal CMOS process. Circuit designs at the transistor level yield area efficient implementations for synapses and the adaptation blocks. We present experimental results from testing the systems with sinusoidal signals and noise and report on its performance as a blind separator of mixed speech signals.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
Keywords
Computer scienceCMOSNeuromorphic engineeringSubthreshold conductionChipComponent (thermodynamics)Filter (signal processing)Electronic engineeringTransistorArtificial neural networkArtificial intelligenceElectrical engineeringEngineeringTelecommunications
Affiliated Institutions
Related Publications
Reception of multiple co-channel digital signals using antenna arrays with applications to PCS
Antenna arrays can be used to increase system capacity in PCS networks by supporting multiple co-channel users per cell in receive and in transmit. We propose a novel approach f...
The Capacity of Linear Channels with Additive Gaussian Noise
The standard method of computing the mutual information between two stochastic processes with finite energy replaces the processes with their Fourier coefficients. This procedur...
Certain Factors Affecting Telegraph Speed<sup>1</sup>
This paper considers two fundamental factors entering into the maximum speed of transmission of intelligence by telegraph. These factors are signal shaping and choice of codes. ...
CSPNet: A New Backbone that can Enhance Learning Capability of CNN
Neural networks have enabled state-of-the-art approaches to achieve incredible results on computer vision tasks such as object detection. However, such success greatly relies on...
Publication Info
- Year
- 1995
- Type
- article
- Volume
- 42
- Issue
- 2
- Pages
- 65-77
- Citations
- 34
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
34
OpenAlex
Cite This
M.H. Cohen,
Andreas G. Andreou
(1995).
Analog CMOS integration and experimentation with an autoadaptive independent component analyzer.
IEEE Transactions on Circuits and Systems II Analog and Digital Signal Processing
, 42
(2)
, 65-77.
https://doi.org/10.1109/82.365346
Identifiers
- DOI
- 10.1109/82.365346