by DAN CALLOWAY
Published 25 January 2010 @ 15:54 UTC
WEAVERVILLE, NC - Three topics of interest to me at present that require additional research in the realm of IT technical foundations are presented in this article. Over the next eight weeks, I will be conducting research into one of these three areas of IT innovation that I wish to pursue further.
Currently, I am torn between two of the topics. My interest lies in the area of RFID mainly because of my affiliation with the IoTC, headquartered in Amsterdam, The Netherlands, and its founder in Council, Rob van Kranenburg, who has been instrumental in the development of the DIFR networks there. However, another area that peaks my interest very much is that of silicon-optics because of its potential to extend the life expentancy of silicon-based transisters and chip development, which is being threatened by the laws of physics as more and more chips are pushed onto existing chip architecture.
After reading the research selections provided here, comment and let me know which topic you would like to see researched further. I will be posting my entire research paper in roughly 10 weeks on my website and creating an article on my blog pointing to that research paper. Keep watching!
Silicon-Optics is a relatively new technological innovation that brings both silicon-based technology and laser optics together on the chip. Two reasons for replacing silicon-based technology; that is, in the manufacture of silicon-based transistors and chip construction in the IT industry today, are the physical problems that silicon presents in overall power consumption and heat issues at the chip level, especially as more transistors are brought in closer proximity to one another when added to existing chip architecture. Silicon-optics is seen to have the potential to enhance computing power, reduce joule heat within the chip, increase data transfer rate, and potentially extend the life of silicon-based technology and its use in transistor and chip manufacturing. Bringing laser optics onto the chip alleviates the restrictions of electrical capacitance and resistance associated with copper wiring in printed circuit cards and chip construction that contribute to the power loss and increase in joule heat within the chip. In addition, light beams used in optical transmissions can be split into multiple communications channels that can be multiplexed onto a single link, thereby offering very high data capacities.
RFID networking technology and its incorporation into real-world objects allow them to become smart objects, giving devices the ability to communicate in a pervasive and salient fashion with other devices via a ubiquitous network we are beginning to refer to as The Internet of Things.
Although radio-frequency technology itself isn’t necessarily a new concept since it was first envisioned by Harry Stockman in papers he wrote back in 1948, and a patent for the first true RFID device: a passive radio transponder with memory, was issued to Mario Cardullo in 1973, what is relatively new is the refinement in the development of RFID micro-chip technology and its incorporation into objects or devices that have been used to improve supply-chain management, IT asset management, retail sales, and inventory control through enhanced barcoding technology, which has seen its increase in popularity thanks to such organizations as Wal-Mart and the Department of Defense beginning in the 1980s. Since this time, RFID chips have found their way into such things as smart homes, smart toasters, smart meters (electrical and water), mobile phones, toll roads, public transportation systems, airport baggage handling systems, the aerospace industry, and animals. The potential use of RFID technology for surveillance purposes and possibly its implantation into human beings for tracking purposes is something that is being researched today and may already be in use. An organization called Pachube, pronounced Patch-Bay, headquartered in the UK, is actively using RFID technology that allows one to tag and share real-time sensor data over the Internet from objects, devices, buildings, and environments both physical and virtual.
Software implementation of neural networks and the development of silicon technology to learn and relearn to perform particular functions. Although the modern computing architecture developed under the von Neumann architecture design concept, which relies on silicon-based transistor and chip technology, may be facing its extinction within the next decade, the idea of replacing silicon-based technology with alternatives such as molecular-, biological-, or quantum-computing technologies and architecture is not recommended since these alternatives are still in their infancy and much more research is needed before they become a viable replacement for silicon and conventional computing architectures.
Using the potential applications of software implementation of artificial neural networks as a biological approach (found in nature) to solve complex computational problems is a means of complementing current silicon-based technology and extending the usefulness of silicon in the design and manufacture of both transistor and chip manufacture. The advantages of utilizing silicon-based technology in conjunction with the software implementation of artificial neural networks discussed here outweigh the disadvantages of attempting to move to alternative technologies that would replace silicon, which require many more years of research and refinement before they can be fully implemented.
Some applications that lend themselves to the artificial neural network approach in solving complex problems can be found in the areas of sales forecasting, industrial process control, data validation, risk management, and target marketing. Another area where the artificial neural network is being used today is in the medical field where research is being conducted in modeling parts of the human body to diagnose diseases using CAT scans, electrocardiograms, and ultrascans. The Institute of Neuromorphic Engineering is currently researching the use of artificial neural networks in the development of a VLSI circuit design for a trainable adaptive filter for audio processing that feeds output to an artificial cochlear, and for the development of robust robotic motion in a high-degree-of-freedom system known as the Wormbot project.
The Defense Sciences Office’s [Bio:Info:Micro] Program, in collaboration with other DARPA offices, is currently conducting research in the use of artificial neural networks in the fields of biology, microsystems technology, and information technology to develop tools that model the functional capabilities of biological systems and to study biological systems extending from single cells to the mammalian brain. Some of the most recent accomplishments include: (1) the development of a cognitive prosthetic that decodes motor signals; (2) the development of the suspended microchannel resident biosensor yielding extremely high sensitivity; and (3) the demonstration of DNA moving in channels under 100 nm in width resulting in uncoiled DNA, which has lead to a greater quantitative understanding of the nature of DNA within those channels.
Recent Comments