A multilinear constraint on dichromatic planes for illumination estimation.

A new multilinear constraint on the color of the scene illuminant based on the dichromatic reflection model is proposed. The formulation avoids the problem, common to previous dichromatic methods, of having to first identify pixels corresponding to the same surface material. Once pixels from two or more materials have been identified, their corresponding dichromatic planes can be intersected to yield the illuminant color. However, it is not always easy to determine which pixels from an arbitrary region of an image belong to which dichromatic plane. The image region may cover an area of the scene encompassing several different materials and, hence, pixels from several different dichromatic planes. The new multilinear constraint accounts for this multiplicity of materials and provides a mechanism for choosing the most plausible illuminant from a finite set of candidate illuminants. The performance of this new method is tested on a database of real images.     

利用结构化ASIC实现信号处理应用

FPGA所具有的设计灵活性和大吞吐量特性使其成为传统数字信号处理(DSP)器件可靠的芯片解决方案,例如无线基站、医学成像和图像记录等高性能DSP应用.在很多情况下,FPGA和高密度ASIC、DSP一起布置在同一块电路板上.通常由ASIC和FPGA分担的硬件功能现在主要由FPGA来实现,这是因为FPGA能够为DSP提供具有成本效益的方案,广泛应用于各种领域.     

Accelerating Machine-Learning Algorithms on FPGAs using Pattern-Based Decomposition

Machine-learning algorithms are employed in a wide variety of applications to extract useful information from data sets, and many are known to suffer from super-linear increases in computational time with increasing data size and number of signals being processed (data dimension). Certain principal machine-learning algorithms are commonly found embedded in larger detection, estimation, or classification operations. Three such principal algorithms are the Parzen window-based, non-parametric estimation of Probability Density Functions (PDFs), K-means clustering and correlation. Because they form an integral part of numerous machine-learning applications, fast and efficient execution of these algorithms is extremely desirable. FPGA-based reconfigurable computing (RC) has been successfully used to accelerate computationally intensive problems in a wide variety of scientific domains to achieve speedup over traditional software implementations. However, this potential benefit is quite often not fully realized because creating efficient FPGA designs is generally carried out in a laborious, case-specific manner requiring a great amount of redundant time and effort. In this paper, an approach using pattern-based decomposition for algorithm acceleration on FPGAs is proposed that offers significant increases in productivity via design reusability. Using this approach, we design, analyze, and implement a multi-dimensional PDF estimation algorithm using Gaussian kernels on FPGAs. First, the algorithm’s amenability to a hardware paradigm and expected speedups are predicted. After implementation, actual speedup and performance metrics are compared to the predictions, showing speedup on the order of 20× over a 3.2 GHz processor. Multi-core architectures are developed to further improve performance by scaling the design. Portability of the hardware design across multiple FPGA platforms is also analyzed. After implementing the PDF algorithm, the value of pattern-based decomposition to support reuse is demonstrated by rapid development of the K-means and correlation algorithms.     

不同的思维——苹果公司的Mac Mini电脑

苹果电脑公司把大量的计算能力都塞进了小巧玲珑的Mac Mini电脑中。苹果电脑公司经过仔细设计的Mac Mini是一个性能优越的、低噪声的、经济的计算机,但大小只有6．5英寸×6．5英寸×2英寸、重量不足3磅。想更深入地了解下面介绍的这些模块的特性,请访问www．edn．com/051216pe。感谢Other World Computing公司(www．otherowrldcomputing．com)提供的本篇报道所使用的高分辨率图片。     

高清晰光盘格式陷入“低水平”竞争

蓝色激光（Blu-ray，简称蓝光，以下同）光盘和HD DVD均开始进入量产，但格式大战的赢家还不明朗。红色激光：简称红光，以下同）格式仍是某些应用的可靠替代方案，但用户版权限制和显示器与视觉的限制使格式的未来令人疑惑。     

Queue Lengths and Waiting Times for Multiserver Queues with Abandonment and Retrials

We consider a Markovian multiserver queueing model with time dependent parameters where waiting customers may abandon and subsequently retry. We provide simple fluid and diffusion approximations to estimate the mean, variance, and density for both the queue length and virtual waiting time processes arising in this model. These approximations, which are generated by numerically integrating only 7 ordinary differential equations, are justified by limit theorems where the arrival rate and number of servers grow large. We compare our approximations to simulations, and they perform extremely well.     

The determination of protein,fat and moisture in bread by near infrared reflectance spectroscopy

Near infrared (n.i.r.) reflectance spectroscopy has been employed for the determination of protein, fat and moisture in sliced white bread. N.i.r. reflectance at six wavelengths was measured using circular samples from each of six alternate slices taken from one half of each of 30 loaves of different composition. The six readings for each loaf at each wavelength were averaged and used to produce calibrations which, on prediction of the compositions of a further 30 loaves sampled in the same way, gave rise to standard deviations of differences between n.i.r. and standard procedures of 0.20% for protein, 0.18% for fat and 0.51% for moisture. Calibrations derived from the other halves of the loaves, which had been air-dried and ground to a powder, resulted in similar standard deviation of differences for protein and fat.     

InAlSb/InAs/AlGaSb Quantum Well Heterostructures for High-Electron-Mobility Transistors

Heterostructures for InAs-channel high-electron-mobility transistors (HEMTs) were investigated. Reactive AlSb buffer and barrier layers were replaced by more stable Al_0.7Ga_0.3Sb and In_0.2Al_0.8Sb alloys. The distance between the gate and the channel was reduced to 7–13 nm to allow good aspect ratios for very short gate lengths. In addition, n⁺-InAs caps were successfully deposited on the In_0.2Al_0.8Sb upper barrier allowing for low sheet resistance with relatively low sheet carrier density in the channel. These advances are expected to result in InAs-channel HEMTs with enhanced microwave performance and better reliability.     

Packet Error Rate Distribution Between Random Bluetooth Piconet Pairs

The packet error rate between two piconets depends on the temporal alignment of their packets and the spectral alignment of the intervals from which the frequencies in their hop sequence are chosen. The relationship between two randomly paired piconets is one of over 828 billion possible relationships. We define these relationships and derive an expression for determining the packet error rate for a specific pair of piconets using single-slot packets. We derive the probability mass function for the packet error rate and extend it to provide the possible packet error rates for an arbitrary number of neighboring piconets. We also derive a probability mass function for the goodput of a piconet with a neighboring piconet. The probability mass functions for the packet error rate is bimodal, meaning the expected value of the goodput or packet error rate is not a good choice for piconet performance analysis. Brian S. Peterson is Chief of the Advanced MASINT Research and Requirements Branch at the National Air and Space Intelligence Center, Wright-Patterson AFB, Ohio. He received the B.S.E.E degree in 1991 from the United States Air Force Academy, an M.S. degree in Systems Engineering in 1995 from, and an M.S.E.E. degree from Florida State University in 1998. He received his Ph.D. degree in Electrical Engineering in 2005 from the Air Force Institute of Technology. Dr. Peterson's research interests include computer communication protocols and wireless networking. Dr. Peterson is a member of the IEEE. Rusty O. Baldwin is an Associate Professor of Computer Engineering in the Department of Electrical and Computer Engineering at the Air Force Institute of Technology, Wright-Patterson AFB, Ohio. He received the B.S.E.E degree (with honors) in 1987 from the New Mexico State University and the M.S. degree in Computer Engineering in 1992 from AFIT. He received his Ph.D. degree in Electrical Engineering in 1999 from Virginia Polytechnic Institute and State University. Dr. Baldwin's research interests include computer communication protocols, information warfare, and wireless networking. Dr. Baldwin is a Senior member of the IEEE. Richard A. Raines is an Associate Professor of Electrical Engineering in the Department of Electrical and Computer Engineering at the Air Force Institute of Technology (AFIT), Wright-Patterson AFB, Ohio. He received the B.S.E.E degree (with honors) in 1985 from the Florida State University and the M.S. degree in Computer Engineering in 1987 from AFIT. He received his Ph.D. degree in Electrical Engineering in 1994 from Virginia Polytechnic Institute and State University. Dr. Raines' research interests include computer communication protocols, information security, and wireless networking. Dr. Raines is a Senior member of the IEEE.