The award will run from October 19th, 2009 to April 30th, 2011.

Research Abstract - The team of North Carolina State University (NCSU) and Teledyne Scientific & Imaging (TSI) will develop tunable materials in accordance with the Defense Microelectronic Activity (DMEA) High Performance Tunable Materials Program (HPTM Task 09-9C3). The materials development effort will be founded on combinatorial synthesis methods ideally suited to solving issues of defect chemistry which regulate the relationships between the non-linear dielectric response and dielectric loss. Furthermore, a powerful combination of microwave theory and electromagnetic simulations will be used to identify the best methods for incorporating ferroelectric capacitors into microwave circuits. Finally, this collective learning will be translated into a set of prototype microwave circuits of military and commercial interest. These will demonstrate the potential for improved wireless systems through integration of functional advanced ferroelectric thin films.

The award will run from September 16th, 2009 to August 15th, 2012.

Research Abstract - The objective of the proposed work is to use electrospinning technology to integrate dissimilar materials (lithium alloy and carbon) into novel composite nanofiber anodes, which simultaneously have high energy density, reduced cost, and improved abuse tolerance. The nanofiber structure also allows the anodes to withstand repeated cycles of expansion and contraction. These composite nanofibers are electrospun into nonwoven fabrics with thickness of 50 ?m or more, and then directly used as anodes in a lithium-ion battery. This will eliminate the presence of non-active materials (e.g., conducting carbon black and polymer binder) and result in high energy and power densities. The nonwoven anode structure also provides a large electrode-electrolyte interface and, hence, high rate capacity and good low-temperature performance capability.

One important emerging threat are hardware attacks, which exploit the fact that data can be read or modified directly in the system's memory using devices that dump or scan memory chips. Data transferred along system buses is similarly vulnerable to hardware attacks. These attacks may be more difficult to perform than software-based attacks, but they are also very powerful. A physical attack can bypass all software security protection in the system, allowing attackers to read memory locations that store cryptographic keys and other sensitive information that may be used in software protection schemes. Widely available and inexpensive mod-chips that bypass Digital Rights Management in game systems demonstrate that physical attacks are very realistic threats.

We propose: (1) To conduct detailed investigation into secure booting and configuration mechanisms for secure processors, (2) To explore how secure processors can support system features such as virtualization, virtual memory, inter-process communication, secure I/O communication, and achieve all those with low performance and storage overheads, and (3) To investigate how secure processor technology can be supported in a variety of computer platforms such as single processor systems, mobile systems, and multiprocessor systems with various interconnect topologies.

The award will run from September 15th, 2009 to August 31st, 2012.

Research Abstract - A computer program consists of many low-level instructions that are executed by a microprocessor. The key to executing a program faster is executing more instructions in parallel. Branch instructions hinder this process. A branch instruction is a decision, whose outcome determines the next instruction to be executed. Thus, the branch must be executed before subsequent instructions can be executed. A microprocessor attempts to circumvent this constraint by predicting the outcome of the branch, enabling instructions from the predicted target to be executed speculatively and without delay.

Branch predictors proposed over the years are clever variations of the same basic approach and the accuracy of this approach only scales to a point. A breakthrough in branch predictor design would be transformational. It would not only have immediate performance scaling benefits for microprocessors today, but would also free designers to reinvent how microprocessors execute programs.

This project provides insight into why conventional branch predictors are limited. A new direction in branch predictor design is revealed by this understanding. Two interrelated problems are exposed: 1) conventional predictors often fail to distinguish dynamic branches for which specialized predictions are required, especially memory-dependent branches, and 2) explicitly specializing predictions for these dynamic branches does not fix the problem alone, because stores to their dependent memory addresses change their future outcomes anyway. We propose two unprecedented principles for branch predictor design: first, explicitly identifying dynamic branches in order to provide them with specialized predictions and, second, actively updating their predictions when stores occur to their dependent memory addresses. The goal of the proposed research is to apply these two principles to design predictors that achieve leaps in branch prediction accuracy, halving or more than halving the number of mispredictions with respect to the best known predictor.

The award will run from September 14th, 2009 to September 13th, 2010.

Research Abstract - The PI and team will investigate polymer Polarization Gratings (PGs) optimized for reflective Liquid Crystal (LC) microdisplay projector systems.  The target is to employ optical techniques and materials improvements to develop PG elements that produce enhanced extinction ratio and brightness of a PG-based projector using a reflective LC microdisplay, and to develop PG processing techniques and tools that enable commercially viable fabrication.  As part of the effort, the NCSU team will coordinate and interact with the sponsor and with various microdisplay, projector, thin-film manufacturing, and optical companies, who are engaged by the sponsor, in order to facilitate access to advice, materials, components, fabrication tools, and independent verification supporting the research goals.

The award will run from September 1st, 2009 to August 31st, 2012.

Research Abstract - The proposed work will improve memory systems for high-performance real-time embedded systems. We will develop methods implemented in hardware and software to create fast, efficient memory systems with easily predicted performance.  This is critical to designers of systems with real-time requirements, as memory access time affects task execution time. A task's worst-case execution time (WCET) is the fundamental metric which is used in equations to determine how to make task scheduling decisions. For most non-trivial hardware and software, computing the exact WCET is impossible. This leads designers to estimate a safe WCET bound which is guaranteed by design to never be smaller than the actual WCET. The closer the WCET bound is to the WCET, the more efficient the system can be, as there is less time dedicated to a margin of safety. Unfortunately, it is difficult to analyze the timing impact of caches and virtual memory systems on a task's WCET bound, much less WCET, resulting in overly conservative WCET bound estimates which may make the use of cache and virtual memory impractical. The goal of this proposal is to investigate methods to improve cache and virtual memory performance in a way that is easy to predict.

The award will run from September 1st, 2009 to August 31st, 2012.

Research Abstract - We propose a new paradigm in antenna design and development that can lead to very low-loss radiating/receiving devices that are mechanically stretchable, flexible, and "self-healing". These properties offer a wide spectrum of possibilities in reconfigurable antennas for wireless and tactical sensor applications, with spectral response and geometry that can be modulated by strain, motion, pressure, and electric potential.

The award will run from July 29th, 2009 to December 31st, 2010.

Research Abstract - NCSU will develop algorithms, sensor and sensor processing circuits for self-healing analog and radio frequency (RF) integrated circuits.

Increasingly, complex analog and RF circuits require in-situ calibration in order to function correctly in the presence of manufacturing and temporal variations.  Digital correction is used to "correct" the operation of the transistor level analog circuit.  Currently, the implementation structures for self-calibration are ad-hoc, with each sub-circuit often having a unique hand-crafted approach.  In this project, we are determining new general algorithmic approaches to support the implementation of self-calibrating circuits.  These techniques are intended to be widely employable, and have low implementation overhead.  In addition, we are working on general approaches to designing the calibration sub-circuits that will improve their efficiency and implementation cost.

The award will run from August 1st, 2009 to July 31st, 2012.

Research Abstract - This proposal describes a collaborative cybersystem research for the next three years. Its main objective is to defend Internet services against malicious attacks through innovative secure mechanisms applied jointly at network and physical layers.  The research agenda includes robust on-line network anomaly detection, agile packet filtering, and hardware design innovation for cyber-security.  It will deliver effective, low-cost, and easy-to-upgrade solutions to secure the next-generation Internet.

The award will run from June 16th, 2009 to June 15th, 2010.

Research Abstract - This proposal proposes to build an outdoor wireless mesh testbed comprised of a large number of low-cost experimental fabricated nodes and a small number of commercially available nodes.

The testbed will be built in two stages: in the first stage, nodes placed on pushcarts will be temporarily placed outdoors for trials and tests; in the second phase, permanent antenna placements will be installed on equipment poles over a large area of the Centennial Campus of North Carolina State University.  The testbed will leverage experience of, as well as enable the research of NCSU researchers participating in the Secure Open Systems Institute (SOSI), currently engaged in DoD, NSF, and other projects.

Current and envisaged research activities of SOSI researchers address secure and redundant routing, energy-efficient routing, topology control, localization, cross-layer optimization, security and performance of SIP and VoIP, secure virtualization of network and compute resources, social networking. The proposed testbed will provide realistic large-scale outdoor wireless network environments for evaluating and validating the ideas, protocols and systems conceived from these activities.

The data and experience gained from operating and managing a real network environment will also provide practical insights for students and researchers on the operation of large-scale heterogeneous mesh networks that help identify new security and performance problems and develop their practical solutions.

The award will run from June 7th, 2009 to June 6th, 2010.

Research Abstract - Wireless communications systems are widely modeled and measured. The successful model by is used by many groups and forms the basis for understanding propagation effects in a standard wireless system. It only allows statistical description of a channel, however, so is not adequate for testing algorithms whose performance is strongly affected by the precise local environment. In particular, testing these algorithms requires the creation of scenarios that are typical and others that are challenging.

An example of such an algorithm is long-range prediction, which has been shown to enable adaptive modulation to achieve significant gains on the wireless channel, and is becoming widely studied. The variation of the channel parameters with position plays a large role in determining the achievable performance of the algorithm. We have developed a physical model to test long-range prediction. It provides the necessary insights for predicting challenging or typical scenarios, and creates simulated channels for testing long-range prediction that include physically realistic parameter variation in space. The physical model compares well with measurements for narrow band channels in a suburban environment, as judged by the performance of the long-range prediction algorithm. This is in contrast to the prediction of a Jake's model simulated channel, which does not include physically meaningful parameter variations (it is stationary).

As we modify the model to other situations, such as prediction at a frequency other than the one sampled, prediction near complex scattering objects, and peer-to-peer systems that utilize sectored antennas, we have moved well beyond comparison of the model to measured channels. The measurements needed to verify the model in these cases are too involved and need too much interaction with model building to be realistically carried out by a remote group. We therefore propose here to purchase equipment and develop the measurement capabilities to insure that the models are reasonable and to develop insights to further enhance the model for these and future projects.

The award will run from May 19th, 2009 to November 19th, 2012.

Research Abstract - This research program proposes to explore the feasibility of engineering thermal radiation for application to thermal energy harvesting.  The approach is to utilize the high energy density stored in the evanescent field of surface excitations present on a thermal source composed of a polar semiconductor, by transforming it into spectrally and/or spatially selective radiation for ready extraction.

At nanoscale distances, near-field thermal excitations of polar semiconductors occupy narrow bands at THz frequencies. These excitations, in the form of surface waves, establish a quasi-coherent, evanescent field with high energy density.  Properly designed surface microstructures can convert the power available in evanescent modes to propagating modes with high efficiency.

The proposed effort is directed towards advancements in a theoretical model that can be used to both (1) investigate and develop an understanding of the fundamental physical mechanisms associated with the phenomenon, and (2) serve as an aid in designing experiments that can be used to verify the phenomenon, as well as serve as a guide to optimizing device structures.

Various test structures will be fabricated and tested in order to verify the theoretical predictions and demonstrate the feasibility of engineered thermal emission. A number of rectification schemes will also be investigated for harvesting the enhanced thermal radiation in the THz frequency and eventually converting it to the dc energy.

Terahertz radiation, encompassing wavelengths between 30 mm and 1 mm in the electromagnetic spectrum, was not fully utilized until continuous-wave and pulsed sources became available in the  last few decades. Since then, a wide gamut of applications opened for investigation from long wavelength radio astronomy into deep space, communications systems to terahertz imaging. For example, in biological systems, the major advantage comes from the fast absorption of terahertz radiation by water molecules, increasing the contrast between organic and inorganic materials turning it a useful tool for imaging, food screening and even security applications like screening of baggage and explosives in entry ports.

For remote and sensing applications, adverse atmospheric attenuation at terahertz frequencies requires higher level of the output power as well as broadband operation not present in state-of-the-art continuous-wave sources.

This project will extend the parameters of wide band gap materials and processes learned from Phase 1 to design and fabricate a new laser induced semiconductor switch as a terahertz radiation source.

The award will run from December 22, 2008 to December 21, 2010.

Prof. James Tuck and Prof. Yan Solihin have been awarded a two-year grant from the National Science Foundation to study the use of helper computing to improve security and reliability in multicore processors.

As software complexity increases and threats from security attacks grow, a new low-overhead approach for improving software reliability and security is urgently needed. In helper computing, relatively autonomous "helper" threads or processes execute extra code on behalf of the application on separate processor cores or thread contexts. In the past, the use of helper threads was constrained to prefetching and branch prediction.  Only recently has helper computing been exploited for performing metafunctions, such as memory management and bug detection.

The award will run from January 15th, 2009 to December 31st, 2010.

Research Abstract - In this proposed effort with Valencell, Inc., NC State University (NCSU) will be responsible for thin film deposition, device fabrication and optoelectronic characterization of both.  NCSU will investigate and optimize thin film deposition parameters such as film composition, temperature and deposition environment to obtain bright phosphors with spectrally narrow luminescence at specific wavelengths.  This includes both crystalline materials such as gallium oxide and aluminum oxide, as well as spin-coated materials such as doped spin-on-glass.  Multi-wavelength light emitting devices will be fabricated at NCSU using readily available photolithographic techniques to produce both coplanar and vertically stacked phosphor arrays on sapphire, ruby and other substrates.  Fabrication process steps will be refined to optimize the efficiency and reliability of the resultant devices including phosphor geometry.  Compatibility with unpackaged UV LEDs for monolithic devices will also be investigated.  Devices and thin films will be characterized at NCSU, using available visible and near-infrared spectrometers and detectors for wavelength and power optimization.

The project aims to apply GPU-based computing to existing VLSI CAD (computer-aided design) software programs.  The short term activities involve the manual parallelization of existing algorithms so that they can be executed on GPUs.  Design experience will be accumulated to determine algorithmic characteristics that cannot be easily migrated to GPUs.  Software solutions will then be proposed to assist the conversion of programs with the above characteristics and improve the efficiency of the resulting programs.  Finally, conversion tools will be created to implement semi-automatic parallelization of algorithms. The ultimate goal is to substantially lower the difficulty of sequential to parallel program transformation.