Maher Rizkalla

A constructive algorithm for stabilization of nonlinear neutral time-delayed systems occurring in bioengineering

A criterion is given for the stabilization of the trivial solutions solution of a nonlinear functional differential equation of neutral type occurring in bioengineering under the assumptions of continuity and boundedness of the coefficients. The criterion is a condition on the roots of a certain ‘quasi-polynomial’, i.e. a polynomial in a variable and exponential of that variable. The derivation is based on a constructive algorithm which generates a stabilizing control. The method is illustrated by the construction of stabilizing control for a smooth per suit eye movement system. It is shown that for the synthesis of required control function it is unnecessary in a number of cases to solve the corresponding Riccati matrix equation; rather, it is sufficient to find the roots of a certain ‘quasi-polynomial’, i.e. a polynomial in a variable and exponential of that variable. The model of the smooth pursuit eye movement system can be mathematically described by a set of nonlinear, neutral, functional differenti...

Modified Motion Vector Searches for H.264/AVC

Motion estimation is an important and essential part of video compression. In the encoding process motion estimation is the most computationally intensive part of the encoder. Different search patterns can be used to find the similar blocks between frames. The similarity between the frames is transmitted in the form of motion vectors. In this paper the JM 9.0 software is modified to implement different motion vector searches. This paper shows the effect of the different search patterns on the time taken for motion estimation, the total time taken for the encoding process and bitrate

Electromagnetic and Thermal Simulations of Human Neurons for SAR Applications

The impact of the electromagnetic waves (EM) on human neurons (HN) has been under investigation for decades, in efforts to understand the impact of cell phones (radiation) on human health, or radiation absorption by HN for medical diagnosis and treatment. Research issues including the wave frequency, power intensity, reflections and scattering, and penetration depths are of important considerations to be incorporated into the research study. In this study, computer simulation for the EM exposure to HN was studied for the purpose of determining the upper limits of the electric and magnetic field intensities, power consumption, reflections and transmissions, and the change in temperature resulting from the power absorption by human neurons. Both high frequency structural simulators (HFSS) from ANSYS software, and COMSOL multi-physics were used for the simulation of the EM transmissions and reflections, and the temperature profile within the cells, respectively. For the temperature profile estimation, the study considers an electrical source of 0.5 watt input power, 64 MHz. The EM simulation was looking into the uniformity of the fields within the sample cells. The size of the waveguide was set to be appropriate for a small animal model to be conducted in the future. The incident power was fully transmitted throughout the waveguide, and less than 1% reflections were observed from the simulation. The minimum reflected power near the sample under investigation was found to be with negligible reflected field strengths. The temperature profile resulting from the COMSOL simulation was found to be near 0.25 m°K, indicating no change in temperature on the neuro cells under the EM exposure. The paper details the simulation results for the EM response determined by HFSS, and temperature profile simulated by COMSOL.

All-zero block detection in VC-1

VC-1 is a video codec specification that has been standardized by the Society of Motion Picture and Television Engineers (SMPTE) and implemented by Microsoft as Windows Media Video (WMV) 9. In this paper, zero-block detection is used to reduce the computation in the VC-1 encoder for very low bit rate video, like videophone applications. The zero-block detection depends on comparing the sum of the absolute difference (SAD) of a residual block to a value related to the quantization parameter (QP) during the process of the motion search for the enter-frame encoding. We introduced two options. One is used to save additional time and the other one is used to preserve the quality. One of the VC-1 innovations is how it deals with an entire block of zero coefficients. The transform information is not sent, and the inverse transform is a zero block. Preserving the quality, results showed that an adaptive technique can reduce the encoding and decoding times, up to %3.5 in encoding and %10.5 in decoding times, and the code length, up to %8 reduction. This comes with negligible reduction in the quality represented by the peak signal to noise ratio (PSNR), less than 0.2 db. The measurements based on the SSIM, structural similarity shows improvement in the perceptual quality of the decoded sequences.

Motion Vector Search modified to reduce encoding time in H.264/AVC

Video compression is an essential requirement for transmission and storage of data. As technology advances, the type, size and bitrate requirements of a system become more complex. Motion estimation is an important part of video compression. In this paper the JM 9.0 software is modified to implement different motion estimation techniques. These techniques reduce the time taken for motion estimation and for the total encoding process while not significantly affecting the bitrate

VOLUMETRIC COMPARISON OF AUTOMATICALLY SEGMENTED HIPPOCAMPAL SUBFIELDS FROM 4-MIN ACCELERATED VERSUS 8-MIN T2-WEIGHTED 3T MRI SCANS

AUTOMATICALLY SEGMENTED HIPPOCAMPAL SUBFIELDS FROM 4-MINACCELERATEDVERSUS 8-MIN T2-WEIGHTED 3T MRI SCANS Shan Cong, Shannon L. Risacher, John D. West, Yu-Chien Wu, Liana G. Apostolova, Eileen F. Tallman, Maher Rizkalla, Paul Salama, Andrew J. Saykin, Li Shen, Purdue School of Engineering and Technology, IUPUI, Indianapolis, IN, USA; Indiana University School of Medicine, Indianapolis, IN, USA; Indiana Alzheimer Disease Center, Indianapolis, IN, USA; 4 University of California Los Angeles, Los Angeles, CA, USA; 5 Indiana University Network Science, Bloomington, IN, USA. Contact e-mail: asaykin@iupui.edu

Interfacing nanoparticles to CMOS quad instrumentation amplifiers for gas sensing devices

In this study, we demonstrated a novel approach of a low noise performance gas sensory system, which is capable of minimizing the cross talk between four instrumentation amplifiers. These amplifiers will receive gas sensing information from four different nanoparticle assemblies using monolayer Graphene sheets. The cross talk was minimized by using guard rings surrounding the amplifiers. Guard rings serve as dummy collectors for the minority carriers crossing the parasitic BJT devices within the quad instrumentation amplifier System on chip (SOC). This study aims determination of the gas sensing signature array (GSSA), including the type of gas, its concentration, and dynamic performance (rise and fall times). The developed devices were based on the detection of the change in the graphene electrical characteristics when exposed to various gases. The paper presents the first phase of a long-term project, detailing hardware and circuit simulation software design, and its practical implementation, leading to the optimum gas sensing system resulting from the GSSA data. The practical model was based on monolayer graphene films deposited using CVD process on top of SiO2 layers of silicon substrate wafers. Gold electrodes were deposited with sputtering process. The device was then tested with CO2 gas with 100% and 5% concentrations to estimate its sensitivity. As high as 48% change in resistance was detected from 100% CO2 concentration, and near 4% change in response to 5% CO2 concentration, a level of change that can be easily monitored by the quad CMOS instrumentation chip. A gain of 83dB for the instrumentation amplifier was calculated from the circuit simulation. The Quad instrumentation amplifiers were developed using MOSIS service and showed as low attenuation as 90dB.

Fast Block-Size Selection Using Residual Texture for VC-1

The SMPTE Standard 421M, commonly known as VC-1, is a slate-of-the-art video compression format that provides highly competitive video quality, from very low through very high bit rates, at a reasonable computational complexity. VC-1 has many differences, one of which is adaptive block size transforms, from previous standards such as MPEG-2 and H.264/AVC. In this paper, we propose an algorithm that is aimed to speed up the choice of the appropriate block size in inter-frame coding. Instead of using the data from the image, we look at data after the residual texture has been calculated. The residual texture of each block is small and uniform, allowing for simplified calculations and faster block choices. Motion estimation is the first step in the process; this is followed by partitioning the residual data. Then we calculate the residual texture, and make a choice of block size using the constraints set forth. The algorithm speeds up block size selection without harming the compression ratio.

Advanced Motion Estimation Techniques for VC-1

The SMPTE standard 421 M, commonly known as VC-1, is a state-of-the-art video compression format that provides highly competitive video quality, from very low through very high bit rates, at a reasonable computational complexity. VC-1 has many differences previous standards such as MPEG-2 and H.264/AVC. In this paper, we implement four different motion estimation techniques. These techniques are well known in the video compression world, but have a different effect on each video codec. The effect in VC-1 is shown as an increase in speed and a gain, on average, in the signal-to-noise ratio.

Modeling and Simulation of Carbon Nanotube Interconnect Network

The electrical properties of metallic carbon nanotubes (CNT) can rival, or even exceed, the best metals known. It is a potential candidate for future on-chip interconnect, whose performance will be dominant in the next generation integrated circuits. In this paper, a study on the modeling and simulation techniques for the CNT interconnect network is carried out. The frequency-independent models of CNT interconnects in terms of resistance, inductance and capacitance are summarized. A novel frequencydependent circuit model is proposed for CNT for various high-frequency applications. Preliminary analysis shows a good match between numerical simulations and the compact model. The proposed modeling and simulation techniques for CNT interconnect network are expected to play an important role in the future CNT nanotechnology applications.