The Technology Collaborative
Round 15 Awards
1. "Wireless Mesh with Mobility"
The Pennsylvania State University
Penn State’s Department of Computer Science and Engineering will design and implement software to coordinate mobile Radio Frequency Identification (RFID) readers to efficiently gather and store data. The data will be stored in either a distributed or centralized cache, depending on the environment. Queries, for example, to determine inventory are routed to the cache instead of the RFID readers. The purpose is to greatly reduce search times and congestion in the RFID network because tags will not have to be read directly. The team’s work will include the design and implementation of:
- data gathering and query response protocols that coordinate the movement of the mobile readers to efficiently read data;
- centralized and distributed cooperative caching;
- mesh networking protocols to enable to mobile readers to coordinate movement and share data.
Commercialization will be led by Vocollect.
2. "IC Test Data Mining for Yield and Quality"
TestWorks
TestWorks is a newly-formed company that commercializes the patented fault modeling and patent-pending diagnosis technology developed at Carnegie Mellon. Existing approaches for understanding integrated circuit (IC) failures fall short of characterizing the systematic, yield-inhibiting defects inherent to modern ICs. To address this need, TestWorks proposes to integrate and extend prototype software developed at Carnegie Mellon into a comprehensive suite of tools that extract from test measurement data the types of defects that occur, their frequency of occurrence, and the IC design features that lead to their manifestation. Use of their software will allow chip makers and system integrators to efficiently maintain and improve both the yield and quality of current and future ICs.
3. "Spectrum Sensor™, Phase II"
NanoLambda
NanoLambda will develop an engineering sample to continue with the successful development of prototype Spectrum Sensor™ chip. Each material absorbs and reflects different wavelengths with a characteristic pattern: a spectral fingerprint. Optical sensing or diagnostic technology utilizing this spectral information of material has a significant advantage over others, contactless and non-intrusive. However the bulky size and expensive cost of conventional technologies or solutions have limited applications only to the professional uses in the labs. In contrast, NanoLambda’s novel nanotechnology-based Spectrum Sensor™ chip is only a few mm x mm in size and could enable multiple big volume markets in personal and everyday applications such as high resolution color sensing, mobile/wearable non-invasive health monitoring, and multiple gas detection. The objective of this proposed project is to develop a very low cost mass production process of the core component, the metallic nanowire arrays, by utilizing nanoimprint lithography (NIL) technique and suitable polymer materials to meet those volume market needs.
4. "Predictive Switchover for Hybrid Wireless Networks"
Accipiter Systems, Inc..
Accipiter Systems proposed research develops an algorithm which performs predictive switchover between multiple, diverse wireless links. Predictive switchover provides a more available wireless communications system by evaluating the performance of the current link in order to switch to an alternate prior to link failure. This research is in response to government agencies and commercial industries that require enhanced wireless communications to implement public safety systems, secure installations, and acquire status from field equipment. State-of-the-art wireless networking uses single link technologies that can be disrupted by environmental factors or can be obstructed by buildings, foliage, hills, or other structures. Agencies such as Homeland Security and FEMA require the ability to deploy wireless networks, with higher availability than is currently available, during crisis situations, such as terrorist attacks or natural disasters, in order to coordinate operations.
This TTC funding will allow Accipiter Systems (http://www.accipitersystems.com) to further develop its Wi-Hy product offerings. Wi-Hy is a next-generation wireless system that will eliminate network dropouts and provide a more reliable network for the military, emergency crews, surveillance of public infrastructure and for public safety personnel. The military, which widely uses robots to protect war fighters, will use the Wi-Hy wireless connection to safely and reliably perform surveillance of bunkers, tunnels, caves, and buildings. The new system will decrease the threat of network dropout while increasing the diversity of scenarios in which robots can be used. Additionally, the commercial markets which will benefit include wireless surveillance of power plants, dams, fuel depots and oil refineries. The funding will be used to develop the predictive switchover algorithm which improves system performance for real-time traffic types like voice and video.
5. "Application of Quantum Dot Based Emitters in Novel Organic Light
Emitting Diode Architectures"
Plextronics, Inc.
Plextronics, Inc. is a world leader in developing and delivering active layer technology for printed electronic devices - Organic Light Emitting Diode (OLED) displays and lighting, polymer solar cells and plastic circuitry. Active layer technology is the printed conductors and semiconductors that drive device performance. The company has established itself as a world leader in improving OLED performance through its Plexcore™ polymer active layer technology. The work done in this program will advance this technology to enable novel applications of OLEDs in displays and lighting.
Next generation technology in printed organic light emitting diodes (OLEDs), holds promise to deliver the next wave of inexpensive and efficient commercial products such as flat panel displays and white lighting. Utilization of OLEDs to replace incandescent and fluorescent lighting sources has the potential to reduce the total electrical energy consumption for lighting by 33% by 2027. Following completion of this program, Plexcore technology will solve key issues to enable this change.
6. "Advance Audio - Video Navigation and Tagging for Human-Computer Interfaces"
ContentVision
Digital video is rapidly becoming an important source for information, entertainment and a host of multimedia applications. With the size of these collections growing to thousands of hours, technology is needed to effectively present and navigate search results with the medium. Video is a rich source of data where at times visual and textual information occurs together (closed captions in broadcast video).
ContentVision provides infrastructure technology and software solutions for the automated analysis and retrieval of video content. The root of their current and continued success lies in their ability to generate quality metadata about the video it watches. The novelty of ContentVision’s approach is a system that combines visual and textual types of data which is more powerful than addressing only one of them. A key element of their technology platform is image-based, scene-change analysis. Their scene-change analysis produces a storyboard version of the video (series of pictures that depicts what is contained in the video) and provides a superior way to navigate video, especially when searching video in voluminous amounts. New scene-change technology will support detection for different sizes (4:3, 16:9) and for increased accuracy; it can examine specific regions in the video depending on the areas of interest. Furthermore abundant video exists that contains audio with language. Speech to text (STT), systems have a very difficult time identifying names and places correctly when encountered in audio. Names and places are very often key search terms for audio and video. For audio analysis, ContentVision proposes using the open source version of the Sphinx STT research from Carnegie Mellon on www.SourceForge.net. Their innovation will be to add the capability for phonetic audio indexing processing (not currently part of Sphinx) which research has shown is much better at recognizing names and places as well as enhance open Sphinx for commercial grade applications such as a real-time broadcast monitoring environment. They plan to implement as deliverables, dll’s and executables which will pass metadata information to files and provide solutions that will provide superior user interfaces.
7. "Control System for Dynamic Balancing"
Bossa Nova
Bossa Nova Concepts (Bossa Nova) is a Pittsburgh-based, robotic toy start up company, spun out of Carnegie Mellon's Robotic Institute. Bossa Nova proposes a system development program that will enable, for the first time, dynamic balancing while standing and running on a legged consumer product. The prototype will be a robotic toy with lifelike movements and a self-balancing motion that fascinates observers. The funding will allow Bossa Nova to develop the electronic board and control algorithms that enable biped balancing, walking, and running at a cost amenable to toy manufacturing.
Bossa Nova is entering a rapidly growing market. With an average annual growth of 51% over the past three years, robotic toys are one of the fastest growing sectors in the toy industry. NPD Research expects the global market to reach $1.5 billion by 2009. Bossa Nova is poised to capitalize on the strong growth as its products have competitive advantages over current robotic toys, offering superior motion and lifelike qualities. A recent market research study by Unite Productions indicates a clear preference for dynamic, lifelike behavior in robotic toys. Bossa Nova's advantages will be extended by leveraging the company's expertise in high-mobility robotics and utilizing their integrated product design process. To facilitate its commercialization plan, Bossa Nova is conducting due diligence with prospective manufacturers in Asia and has contracted the services of a toy industry consultant who has 35 years of experience in distributing and marketing toys to major retail channels.
8. "Development of an RBDS / ZigBee Emergency Alerting Radio System"
Sima Products
Is there a more compelling use of existing technology, than to warn innocent people of impending danger and help them through it?
Sima and its partners – Carnegie Mellon University and WDUQ FM - will test a new public warning system at CMU’s Oakland campus to disseminate emergency information by targeting receivers using a data sub-carrier on commercial FM bandwidth. Sima’s Strategic Mesh Alert Radio Transmission (SMART) project will allow a single entity, such as a university or chemical plant, to issue specific warnings to selected receivers during a crisis. CMU security can use SMART Radio to systematically evacuate campus by selecting messages and targeting different sets of receivers which automatically alert. The receivers will communicate within a local inter-radio mesh network to extend reception to FM cold spots typically found in university basements and labs. Upon completion of this project, Sima will have a superior solution for a market that is expanding rapidly.
SMART Radio is an installed system of receivers that receive data via RBDS broadcast and communicate via an inter-radio mesh network. The hardware specification for this project calls for using the latest in FM software digital receiver chipsets to receive and process the RBDS signals. The system administration software will allow access to a host server where CMU security personnel can select receiver locations and messages through easy-to-use menus. Students in CMU’s School of Design and Human Computer Interface departments will assist in developing form factors relevant to using the system on a university campus. For the CMU test system, SMART software will communicate with WDUQ’s FM transmission site, where data will be injected into the FM RBDS sub-carrier signal. Receivers continuously monitor RBDS data and alert when targeted. Receivers are both AC powered and equipped with long-life rechargeable batteries to minimize maintenance.
