| |  | +List of Projects:+ |
| | * [*Google's Android Mobile Platform*|#googleAndroidMobilePlatform] |
| | * [*Open Source Hardware Platform for Android*|#openSourceHWPlatform] |
| | * [*Harnessing the Exponential Function*|#harnessingExponentialFunc] |
| | * [*Visualization and Validation Techniques for UML-Based Top-Down Design*|#visualizationValidationTech] |
| | * [*Adaptive Control Algorithms for Time-Series Analysis*|#adaptiveControlAlgo] |
| | * [*Improved Customer Experience in Mobile Phone Use*|#improvedCustomerExperience] |
| | * [*Automation Tools for Small Business: Business Models to Benefit the Community while Seeking Fair Profits*|#automationToolsSmallBusiness] |
| | * [*MTC (Mobile Technology Consortium)*|#mobileTechConsortium] |
| | * [*Highly Scalable Multiplier (HSM)*|#highlyScalableMult] |
| | * [*One Pass to Production (OPP)*|#onePassProduction] |
| | \\ |
| | |
| | ---- |
 | | |
| | [*MTC (Mobile Technology Consortium)*|Mobile Technology Consortium] |
| | |
| | | {anchor:googleAndroidMobilePlatform} |
| | *Google's Android Mobile Platform* |
| | This is the new open source platform from Google that has created much excitement and several product releases from most leading mobile communications companies and service providers, such as Motorola, Samsung, T-Mobile, Verizon, and others. Dr. Shankar has received a SBA (small business administration) grant to train high school students in application development on this platform. The high school teams will involve students with passion and interest in arts/creativity, science/technology, and business/management. Thus, it goes far beyond the usual paradigm of attracting students who are already interested in science and technology. This should help broaden the appeal of technology to all high school students, and help the field in turn to attract creative and enterprising students to the fold. This involves two other faculty members, Drs. Agarwal and Hamza-Lup, in addition to Dr. Shankar. Mr. and Mrs Fonoage are two PhD students working with this team. C. Norona and V. Gallego are two undergraduate students working on this project. A graduate course on Android software components, and an undergraduate course on application development using Android platform ("Software-Hardware CoDesign") are being offered in spring '10 to students at FAU. |
| | |
| | ---- |
| | {anchor:openSourceHWPlatform} |
| | *Open Source Hardware Platform for Android* |
| | Android operating system from Google is an excellent application development environment and operating system that has the potential to revolutionize the application development for mobile systems. However, the mobile systems are still powered off of low performance hardware processors, primarily because of the desire to conserve power consumption and thus enhance the battery life. We expect that Droid from Motorola will indeed show the advantage of an optimized design that can yield long battery power. It is our intention in this project to develop an open source hardware platform that will allow application and system developers to optimize their code and / or hardware to yield better QOS (quality of service) metrics, such as performance, latency, power dissipation, size, user friendliness, etc. This collaboration involves Dr. Shankar and Mr. Jaime Borras, who was formerly a senior VP and CTO at Motorola's iDEN division in FL. An NSF proposal is expected to be submitted in January '10 on this topic. |
| | |
| | ---- |
| | {anchor:harnessingExponentialFunc} |
| | *Harnessing the Exponential Function* |
| | Dr. Shankar recently submitted a proposal to NSF on this topic. The project proposes to evolve a system/domain dependent method to map complex natural and engineered systems to modern distributed processing systems, with the goal to take better advantage of the computational speed, while not compromising on reliability and robustness metrics. Exponential growth in engineered system complexity is a continuing inevitable trend. This is both a challenge and an opportunity. Dr. Shankar has been successful in three different domains in exploiting this as an opportunity and in evolving solutions that are radically better. The first one led to patents on early and noninvasive detection of atherosclerosis; the second one has led to patents on a highly scalable multiplier; and the third one was used in radically enhancing engineering design productivity of mobile systems. There are no patents for the third one. However, Motorola has acknowledged that if they were to incorporate our flow fully, their product development time would be reduced from the current 18 months to about 3 months. The recently submitted NSF proposal goes one step beyond the third domain solution. It takes a holistic approach and is based on a study of natural systems (for their 'near decomposability') and advanced organizations (for their high dependability). These systems are highly complex and yet continue to perform well in a reliable, robust, and dependable manner. A provisional patent has been filed by FAU with USPTO on this. |
| | |
| | ---- |
| | {anchor:visualizationValidationTech} |
| | *Visualization and Validation Techniques for UML-Based Top-Down Design* |
| | UML (unified modeling language) is well suited for top-down design of complex systems. We used it recently to develop documentation for ABET accreditation for our computer engineering curriculum. The results were so impressive that the ABET accreditation committee gave our computer engineering program a six year accreditation, the highest possible, without any reservations. This may be compared with 3 and 4 year accreditations received by other programs within our college. Dr. Shankar has also used UML for top-down software design in his course on software-hardware codesign. There are several issues that need to be addressed from the designer/student perspective: a typical engineer wishes to get started with the real design/coding as soon as possible. It is hard for them to do a sufficient amount of due-diligence before getting started with coding, as would be necessary for this UML based analysis. Unfortunately, if one does not do enough due diligence, the result could be major integration and field errors at a later time. On the other hand, the engineer is under constant pressure by his/her manager to produce and produce fast. While UML is a good top-down design tool, it also is a collection of disconnected diagrams and do not result in executable code that could be put to use quickly. It is also difficult to visualize the design across these multiple diagrams and to validate the correctness of these multiple diagrams. This initiative seeks to find ways to improve the visualization and validation of UML diagrams so engineers (and students) have better intuitive feel and will be able to communicate better with others. This is a collaborative project with Prof. Fran McAfee, Arts and Letters College, FAU, and Drs. Agarwal and Hamza-Lup within the center. |
| | |
| | ---- |
| | {anchor:adaptiveControlAlgo} |
| | *Adaptive Control Algorithms for Time-Series Analysis* |
| | This is a project led by Dr. Pajunen, and is in collaboration with Dr. Shankar. Dr. Pajunen has 10\+ patents on adaptive control algorithms. Real world nonlinear systems are not amenable to reductionist methods of abstraction and simplified input to output transformations. Dr. Pajunen proposes to use a modified version of a non-parametric Volterra model to capture real-world data, under various conditions of normal and faulty behavior of a system, and use it to predict behavior and provide diagnostics information in real-time at a later time. This provides yet another way to manage system complexity. An NSF proposal is being prepared now. |
| | |
| | ---- |
| | {anchor:improvedCustomerExperience} |
| | *Improved Customer Experience in Mobile Phone Use* |
| | This is a project led by Dr. Aalo, and is in collaboration with Dr. Shankar. Dr. Aalo has developed algorithms that will facilitate horizontal and vertical phone-call hand-offs in a more efficient manner. This project will build realistic models with SystemC, a concurrent language derived from C++, in a hierarchical manner. We will then use these models to find ways to trade-off various QOS (quality of service) metrics, so a given user can decide on the most appropriate set of services that he/she desires to sign up for. This customer-centric approach will help the service providers retain most of their customers. An NSF proposal is being prepared now. |
| | |
| | ---- |
| | {anchor:automationToolsSmallBusiness} |
| | *Automation Tools for Small Business: Business Models to Benefit the Community while Seeking Fair Profits* |
| | We propose to develop IT (information technology) tools to bridge the communication gap between the professionals and the community in a given domain. Health 2.0 is such an effort that has evolved in the health care domain. The past 2 years saw a sharp rise and fall in Health 2.0 oriented businesses. A few have survived, by focusing on a specific sub-community. Some of these use advertisements to raise revenue while charging the individual little, while others charge the individual much to stay free of advertisements. Other domains, such as climate change, energy, education, insurance, etc., have communities of interest which also are loaded down by the enormity of information and an inability to process the information objectively and succinctly, to decide how it affects them and how they can protect themselves or benefit from these developments. This is a proposal submitted for funding by FAU. It involves research experts in the areas of energy, climate change, and mobile applications, since they are the targeted areas for development. Prof. F. McAfee, Arts and Letters, and Dr. C. Mishra, Office Depot Chair, College of Business, are partnering with Dr. Shankar on this project. |
| | |
| | ---- |
| | {anchor:mobileTechConsortium} |
| | [*MTC (Mobile Technology Consortium)*\|http://www.csi.fau.edu/display/MTC/Home] |
| | |
| | |
| | ---- |
| | {anchor:highlyScalableMult} |
| | [*Highly Scalable Multiplier (HSM)*|Highly Scalable Multiplier] |
| | Highly Scalable Multiplier uses a folding technique to allow one to trade off speed, power dissipation, and area as needed in the implementation of multipliers. |
| | |
| | |
| | ---- |
| | |
| | | {anchor:onePassProduction} |
| | [*One Pass to Production (OPP)*|One Pass to Production] |
| | This is a project funded by Motorola during 2003-present ($1.2 M so far) on developing an integrated methodology to radically enhance engineering design productivity. The goal is to reduce the design cycle from 24 months (the cycle time in 2003) to 24 hours. We believe that our current methodology after 5 years of research can lead to a one month design cycle. Motorola's internal evaluation puts it more conservatively at 3 months. We are exploring many new innovative ideas to further reduce the design cycle time. |
| | |
| | | This project was funded by Motorola during 2003 to 2008 for a total of about $1.1 M, based on the vision of Jaime Borras, VP and CTO, iDEN, Motorola, during that period. His vision was to integrate mobile phone development process, so the engineering development cycle that typically takes 18 months could be completed in 1 day. This implies an engineering productivity improvement of 500+. A typical mobile phone requires the interaction (and integration of the work) of various types of engineers (RF, analog, DSP, computer, software, test, and manufacturing), managers, customers (service providers), and marketing personnel. Each group has a different vocabulary which manifests itself in terms of different standards, programming languages, engineering methodologies, and software and hardware tools. Some of the productivity impact arises from communication bottlenecks and mishaps. We have been able to address these issues; this has led to a substantial reduction in the product development time, to 3 months (Motorola acknowledges this improvement). Several projects within Motorola, with and without our involvement, have validated our claims. Do note, however, that substantially more improvement is possible. We are at present addressing the issue of automation of requirements capture. This will use the concepts of domain specific language and semantic web, two recent industry innovations that have been adapted by many trail blazing companies. |
| | |
| | ---- |
