We consider the problem of decoding block coded data, using a physical dynamical system. We sketch out a decompression algorithm for fractal block codes and then show how to implement a recurrent neural network using physically simple but highly-nonlinear, analog circuit models of neurons and synapses. The nonlinear system has many fixed points, but we have at our disposal a procedure to choose the parameters in such a way that only one solution, the desired solution, is stable. As a partial proof of the concept, we present experimental data from a small system a 16-neuron analog CMOS chip fabricated in a 2m analog p-well process. This chip operates in the subthreshold regime and, for each choice of parameters, converges to a unique stable state. Each state exhibits a qualitatively fractal shape.
[1]
Carver Mead,et al.
Analog VLSI and neural systems
,
1989
.
[2]
A. Jacquin.
A fractal theory of iterated Markov operators with applications to digital image coding
,
1989
.
[3]
Donald M. Monro,et al.
Fractal block coding of images
,
1992
.
[4]
M. Hirsch.
CONVERGENCE IN NEURAL NETS.
,
1987
.
[5]
Fernando J. Pineda,et al.
Dynamics and architecture for neural computation
,
1988,
J. Complex..
[6]
R. Keynes.
The ionic channels in excitable membranes.
,
1975,
Ciba Foundation symposium.