Talk Abstract
A Simple, Powerful Build Target for Nanoelectronics
Nicholas J. Macias and Lisa J.K. Durbeck, Cell Matrix Corporation
 
talk given by Macias at Nanospace 2002, The Fifth International Conference on Integrated Nano/Micro/Biotechnology for Space and Medical and Commercial Applications, held in Galveston Island, Texas, U.S.A., June 27 2002.
 
This conference was sponsored by the Institute for Advanced Interdisciplinary Research, Rice University, NASA, The Houston Technology Center, University of Texas Medical Branch, NanoTex Foundation, Inc., University of Houston, Texas A&M University, The National Space Biomedical Research Institute
We would like to make nanotechnology researchers working on nanoelectronics aware of the Cell Matrix computing architecture, because this technology can catalyze rapid progress toward bringing nanotechnology to market. Our technology permits the construction of a simple physical structure that is then electronically differentiated into any desired electronic circuit or system. Cell Matrices are a desirable build target for nanoelectronics because:
  • Both the Cell and the Matrix are simple, elegant structures. A Matrix is a regular n-dimensional tiling of m-sided Cells, where each Cell is connected to its nearest neighbors. The full specification is openly available on the web. A Cell can contain as few as 48 bits of memory, 30 simple logic gates and interconnecting wires, and a Matrix can be arbitrarily small or large.
  • The hardware produced will work, even if there are imperfections within the Matrix such as holes, imperfect Cells, or misconnected or broken wires.
  • Once you build Cell Matrix hardware, it can be programmed as a replacement for any electronic component, circuit, chip -- for anything that can be translated to logic gates. This is a property of any suitably large reconfigurable device.
  • Cell Matrices give extra incentive for investing in nanotechnology because they provide something that computer systems engineers do not have now -- a way to design their own high performance parallel systems, quickly set them up, efficiently test and redesign them. Perfectly linear performance and sqrt(n) setup time was recently achieved on a Cell Matrix-based DNA sequence alignment.
References:
Wang B 2002 Implementation of a Dynamic Programming Algorithm for DNA Sequence Alignment on the Cell Matrix Architecture. M.S. Thesis Utah State University
Durbeck L and Macias N 2001 The Cell Matrix: an architecture for nanocomputing. Nanotechnology vol 12 pp 217-30 (Bristol, Philadelphia: Institute of Physics Publishing)
Durbeck L and Macias N 2000 Autonomously Self-Repairing Circuits. NASA Phase I SBIR
U.S. Patents number 5,886,537 ; 6,222,381 ; 6,297,667
Cell Matrix web site http://www.cellmatrix.com/