Why Nano-DSP Will be Fan-In Constrained

This paper studies for the first time the performance of von-Neumann multiplexing (vN-MUX) when stuck at fault model is considered. In this study, vN-MUX is applied to majority (MAJ) gates of small fan-ins (¿ = 3, 5, 7, and 9), and respectively the corresponding redundancy factors (R = 6, 10, 14, and 18). This study is extremely important for a deeper understanding of vN-MUX, especially when considering the unreliable behavior of future nano-devices. The analysis confirms and enhances on well-known theoretical results, and is exact as being obtained using Bayesian network. Finally, the extension to device level will allow us to characterize vN-MUX with respect device failures for the first time ever. The results are very timely and are explaining a strange (non-linear) behavior of vN-MUX that was first reported two years ago (based on extensive Monte Carlo simulations).

[1]  Valeriu Beiu,et al.  Deeper Sparsely Nets can be Optimal , 1998, Neural Processing Letters.

[2]  Valeriu Beiu,et al.  What von Neumann Did Not Say About Multiplexing Beyond Gate Failures - The Gory Details , 2007, IWANN.

[3]  P. Jonker,et al.  A defect-?and fault-tolerant architecture for nanocomputers , 2003 .

[4]  Jie Han,et al.  A system architecture solution for unreliable nanoelectronic devices , 2002 .

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

[6]  A. S. Sadek,et al.  Fault-tolerant techniques for nanocomputers , 2002 .

[7]  U. Ruckert,et al.  On nanoelectronic architectural challenges and solutions , 2004, 4th IEEE Conference on Nanotechnology, 2004..

[8]  Leonard J. Schulman,et al.  On the maximum tolerable noise of k-input gates for reliable computation by formulas , 2003, IEEE Trans. Inf. Theory.

[9]  J.A.B. Fortes,et al.  Bifurcations and fundamental error bounds for fault-tolerant computations , 2005, IEEE Transactions on Nanotechnology.

[10]  A. S. Sadek,et al.  Parallel information and computation with restitution for noise-tolerant nanoscale logic networks , 2003 .

[11]  Don R. Hush,et al.  On the node complexity of neural networks , 1994, Neural Networks.

[12]  Cristian Constantinescu,et al.  Trends and Challenges in VLSI Circuit Reliability , 2003, IEEE Micro.

[13]  Finn V. Jensen,et al.  Bayesian Networks and Decision Graphs , 2001, Statistics for Engineering and Information Science.

[14]  S. Roy,et al.  Majority multiplexing-economical redundant fault-tolerant designs for nanoarchitectures , 2005, IEEE Transactions on Nanotechnology.

[15]  Snorre Aunet,et al.  Real-Time Reconfigurable Linear Threshold Elements and Some Applications to Neural Hardware , 2003, ICES.