Analog LSI Implementation of an Auto-Adaptive Network for Real-Time Separation of Independent Signals
Marc H. Cohen, Philippe O. Pouliquen, Andreas G. Andreou · Neural Information Processing Systems · 1991
We present experimental data from an analog CMOS LSI chip that implements the Herault-Jutten adaptive neural network. Testing procedures and results in time and frequency-domain are described. These include weight convergence trajectories, extraction of a signal in noise, and separation of statistically complex signals such as speech.