Proposal for Stochastic Bit Stream Processing Using Optoelectronic Smart Pixels: A Neural Network Architectural Case Study

A neural network architecture which uses stochastic processing techniques to perform the weighted input multiplication, summation, and thresholding processes of a neuron using the optimal amount of hardware is described. It will be argued that the advantage of this approach is that it will allow large neural networks to be fabricated with relatively small amounts of hardware. The architecture allows a choice to be made between the speed and accuracy of processing, as well as a choice of hardware. Implementations of a bit stream neuron using electronic, optoelectronic and optical hardware are developed and their capabilities are compared based on speed of processing and network size. The aim of this study is to investigate the capabilities of optical logic in distributed processing systems and specifically the use of the optical thyristor as logic elements. It is shown that optical processing and optical interconnection allows a simplification of the processing sequence and allows the parallelism of distributed systems to be utilized. Experimental results of a detector matrix which can statistically quantify the occupancy ratio of optical ones and zeros in a spatial pattern and that can be considered to implement the sum, thresholding, and sigmoid translation functions of a neuron are given.

[1]  John Shawe-Taylor,et al.  A Device for Generating Binary Sequences for Stochastic Computing , 1993 .

[2]  L.A. D'Asaro,et al.  4/spl times/4 arrays of FET-SEED embedded control 2/spl times/1 optoelectronic switching nodes with electrical fan-out , 1994, IEEE Photonics Technology Letters.

[3]  E. Watson.  Primitive Polynomials (Mod 2) , 1962 .

[4]  Parimal Pal Chaudhuri,et al.  A class of two-dimensional cellular automata and their applications in random pattern testing , 1994, J. Electron. Test..

[5]  I. Mazin,et al.  Theory , 1934 .

[6]  Hugo Thienpont,et al.  Interconnection issues for vertical-to-surface-transmission electrophotonic device (VSTEP) optoelectronic information processing systems , 1995, Photonics West.

[7]  A. Compagner,et al.  Maximum-length sequences, cellular automata, and random numbers , 1987 .

[8]  S. Wolfram.  Random sequence generation by cellular automata , 1986 .

[9]  Neal Zierler,et al.  Primitive Trinomials Whose Degree is a Mersenne Exponent , 1969, Inf. Control..

[10]  H. Bolouri,et al.  A RAM-based neural network architecture for wafer scale integration , 1995, Proceedings IEEE International Conference on Wafer Scale Integration (ICWSI).

[11]  Maarten Kuijk,et al.  Cascadable Thyristor Optoelectronic Switch Operating at 50Mbit/Sec with 7.2 Femtojoule External Optical Input Energy , 1995 .

[12]  Jacek M. Zurada,et al.  Introduction to artificial neural systems , 1992 .

[13]  P. Garda,et al.  2-D generation of random numbers by multimode fiber speckle for silicon arrays of processing elements , 1990 .

[14]  John Shawe-Taylor,et al.  Probabilistic Bit Stream Neural Chip: Theory , 1991 .

[15]  P. Heremans,et al.  Down-scaling differential pairs of depleted optical thyristors , 1995, IEEE Photonics Technology Letters.

[16]  S. Redner,et al.  Introduction To Percolation Theory , 2018 .

[17]  Brian R. Gaines,et al.  Stochastic Computing Systems , 1969 .

[18]  Hugo Thienpont,et al.  Optoelectronic programmable logic array which employs diffractive interconnections , 1995, Photonics West.

[19]  J. von Neumann,et al.  Probabilistic Logic and the Synthesis of Reliable Organisms from Unreliable Components , 1956 .

[20]  M. Yagyu,et al.  Design, fabrication and evaluation of a 5-inch wafer scale neural network LSI composed on 576 digital neurons , 1990, 1990 IJCNN International Joint Conference on Neural Networks.