Recently, a graph labeling technique based on prime numbers has been suggested for reducing the costly transitive closure computations in RDF query languages. The suggested prime number graph labeling provides the benefit of fast query processing by a simple divisibility test of labels. However, it has an inherent problem that originates with the nature of prime numbers. Since each prime number must be used exclusively, labels can become significantly large. Therefore, in this paper, we introduce a novel optimization technique to effectively reduce the problem of label overflow. The suggested idea is based on graph decomposition. When label overflow occurs, the full graph is divided into several sub-graphs, and nodes in each sub-graph are separately labeled. Through experiments, we also analyze the effectiveness of the graph decomposition optimization, which is evaluated by the number of divisions.

We present a full description of a polarization-independent athermal differential quadrature phase shift keying (DQPSK) demodulator that employs silica-based planar lightwave circuit (PLC) technology. Silica-based PLC DQPSK demodulator has good characteristics including low polarization dependence, mass producibility, etc. However delay line interferometer (DLI) of demodulator had the large temperature dependence of its optical characteristics, so it required large power consumption to stabilize the chip temperature by the thermo-electric cooler (TEC). We previously made a quick report about an athermal DLI to reduce a power consumption by removing the TEC. In this paper, we focus on the details of the design and the fabrication method we used to achieve the athermal characteristics, and we describe the thermal stability of the signal demodulation and the reliability of our demodulator. We described two athermalization methods; the athermalization of the transmission spectrum and the athermalization of the polarization property. These methods were successfully demonstrated with keeping a high extinction ratio and a small footprint by introducing a novel interwoven DLI configuration. This configuration can also limit the degradation of the polarization dependent phase shift (PDf) to less than 1/10 that with the conventional configuration when the phase shifters on the waveguide are driven. We used our demodulator and examined its demodulation performance for a 43 Gbit/s DQPSK signal. We also verified its long-term reliability and thermal stability against the rapid temperature change. As a result, we confirmed that our athermal demodulator performed sufficiently well for use in DQPSK systems.

In this paper, a resonator based on composite right/left handed transmission line concept is discussed. This resonator excites --1st order resonance mode. We start with half-wave resonators consisting of two unit cells of a composite right/left handed transmission line. From the simulated field distributions, the center of these half-wave resonators can be short-circuited to obtain a quarter-wave resonator in the --1st mode. Susceptance slope parameters are calculated for the resonator. Then this resonator is applied for a 2-pole filter made by LTCC, which can be designed with standard filter design theory owing to the slope parameter. The dimension of the experimental filter implemented by LTCC is 2.5 mm by 1.35 mm by 0.52 mm. The insertion loss is 1.80 dB at the 2.4 GHz band. Good agreement between measured and computed results is obtained.

In recent years elephant flows are increasing by expansion of peer-to-peer (P2P) applications on the Internet. As a result, bandwidth is occupied by specific users triggering unfair resource allocation. The main packet-scheduling mechanism currently employed is first-in first-out (FIFO) where the available bandwidth of short flows is limited by elephant flows. Least attained service (LAS), which decides transfer priority of packets by the total amount of transferred data in all flows, was proposed to solve this problem. However, routers with LAS limit flows with large amount of transferred data even if they are low-rate. Therefore, it is necessary to improve the quality of low-rate flows with long holding times such as voice over Internet protocol (VoIP) applications. This paper proposes rate-based priority control (RBPC), which calculates the flow rate and control the priority by using it. Our proposed method can transfer short flows and low-rate flows in advance. Moreover, its fair performance is shown through simulations.

Information on false paths in a circuit is useful for design and testing. The use of this information may contribute not only to reducing circuit area, the time required for logic synthesis, test generation and test application of the circuit, but also to alleviating over-testing. Since identification of the false paths at gate level is hard, several methods using high-level design information have been proposed. These methods are effective only if the correspondence between paths at register transfer level (RTL) and at gate level can be established. Until now, giving restriction on logic synthesis is the only way to establish the correspondence. However, it is not practical for industrial designs. In this paper, we propose a method for mapping RTL false paths to their corresponding gate level paths without such a specific logic synthesis; it guarantees that the corresponding gate level paths are false. Experimental results show that our path mapping method can establish the correspondences of RTL false paths and many gate level false paths.

This paper treats a band-broadening design technique of a dual-band branch-line coupler with matching networks composed of an impedance step and a short-circuited stub based on the equivalent admittance approach. By replacing each right-handed transmission line (RH-TL) with a composite right/left-handed transmission line (CRLH-TL), very flat couplings over a relative bandwidth of about 10% can be obtained at two arbitrary operating frequencies in comparison with previous CRLH-TLs branch-line couplers. Furthermore, by adding periodical open-circuited stubs into RH-TLs of the designed CRLH-TLs branch-line coupler with matching networks, the entire size of the coupler can be reduced to about 50%. Verification of these band-broadening and size-reduction design techniques can be also shown by an electromagnetic simulation and experiment.

Orthogonal frequency division multiplexing (OFDM) signals have high peak-to-average power ratio (PAPR) and cause large nonlinear distortions in power amplifiers (PAs). Memory effects in PAs also become no longer ignorable for the wide bandwidth of OFDM signals. Digital baseband predistorter is a highly efficient technique to compensate the nonlinear distortions. But it usually has many parameters and takes long time to converge. This paper presents a novel predistorter design using a set of orthogonal polynomials to increase the convergence speed and the compensation quality. Because OFDM signals are approximately complex Gaussian distributed, the complex Hermite polynomials which have a closed-form expression can be used as a set of orthogonal polynomials for OFDM signals. A differential envelope model is adopted in the predistorter design to compensate nonlinear PAs with memory effects. This model is superior to other predistorter models in parameter number to calculate. We inspect the proposed predistorter performance by using an OFDM signal referred to the IEEE 802.11a WLAN standard. Simulation results show that the proposed predistorter is efficient in compensating memory PAs. It is also demonstrated that the proposal acquires a faster convergence speed and a better compensation effect than conventional predistorters.

A method to realize the fine frequency-tuning steps using tiny capacitors instead of Metal-Insulator-Metal (MIM) capacitors is proposed for a digitally controlled oscillator (DCO). The tiny capacitors are realized by the coplanar transmission lines which are arranged unsymmetrical in a 6 metal layers (M6) foundry of 0.18 µm CMOS technology. These transmission line based capacitors are designed by using electro-magnetic field simulator, and co-designed by using SPICE simulator. Finally, these capacitors are employed to design 15 bit DCO and fabricated the proposed DCO in 0.18 µm CMOS technology, and tested. The measured phase noise of DCO was -118.3 dBc/Hz (@1 MHz offset frequency), and the oscillating frequency tuned from 4.86 GHz to 5.36 GHz in the minimum frequency-tuning step of 18 kHz.

The use of image contrast enhancement has become increasingly essential due to the need to better show the visual information contained within the image for all vision-based systems. This has lead to motivation for the design of a powerful and accurate automatic contrast enhancement for a digital image. Histogram equalization is the most commonly used contrast enhancement method. However, the conventional histogram equalization methods usually result in excessive contrast enhancement, which causes the unnatural look and visual artifacts of the processed image. In this paper, we propose a novel histogram equalization method using the automatic histogram separation along with the piecewise transformed function. The contrast enhancement results of the proposed method were not only analyzed through qualitative visual inspection and for quantitative accuracy, but are also compared to the results of other state-of-the-art methods.

Virtual Private Network (VPN) is a cost effective method to provide integrated multimedia services. Usually heterogeneous multimedia data can be categorized into different types according to the required Quality of Service (QoS). Therefore, VPN should support the prioritization among different services. In order to support multiple types of services with different QoS requirements, efficient bandwidth management algorithms are important issues. In this paper, I employ the Kalai-Smorodinsky Bargaining Solution (KSBS) for the development of an adaptive bandwidth adjustment algorithm. In addition, to effectively manage the bandwidth in VPNs, the proposed control paradigm is realized in a dynamic online approach, which is practical for real network operations. The simulations show that the proposed scheme can significantly improve the system performances.

The present paper focuses on the application of the base station cooperation (BSC) technique in fractional frequency reuse (FFR) networks. Fractional frequency reuse is considered to be a promising scheme for avoiding the inter-cell interference problem in OFDMA cellular systems, such as WiMAX, in which the edge mobile stations (MSs) of adjacent cells use different subchannels for separate transmission. However, the problem of FFR is that the cell edge spectral efficiency (SE) is much lower than that of the cell center. The BSC technique, in which adjacent BSs perform cooperative transmission for one cell edge MS with the same channel, may improve the cell edge SE. However, since more BSs transmit signals for one cell edge MS, the use of BSC can also increase the inter-cell interference, which might degrade the network performance. In this paper, with a focus on this tradeoff, we propose an adaptive BSC scheme in which BSC is only performed for the cell edge MSs that can achieve a significant capacity increase with only a slight increase in inter-cell interference. Moreover, a channel reallocation scheme is proposed in order to further improve the performance of the adaptive BSC scheme. The simulation results reveal that, compared to the conventional FFR scheme, the proposed schemes are effective for improving the performance of FFR networks.

The construction of EM absorber and frequency selective shielding has been investigated by using two dimensional metal fiber array (MFA) composites. The MFA composite shows a resonant type frequency dispersion in the complex relative permittivity spectra (εr = εr' - jεr") having a negative εr' region. The frequency characteristics of the conventional ferrite-rubber EM absorber can be improved by combining with the negative permittivity property of the MFA composite. A frequency selective shielding can be achieved by the evanescent EM wave propagation in the layered MFA composite structure.

Error-propagation is an important issue and should be carefully coped with in the decision-feedback equalizers (DFE). Ignoring the impact of error-propagation often leads to impractical laboratory results. In this paper, we investigate two novel layered space-frequency equalizers (LSFE) for single-carrier multiple-input multiple-output (MIMO) systems, where the recently proposed frequency-domain equalizer with time domain noise-predictor (FDE-NP) is adopted at each stage of the LSFE. We first derive the partially-connected LSFE with noise predictor (PC-LSFE-NP) which has exactly the same mean square error (MSE) as the conventional LSFE under the assumption of perfect feedback. However, if error-propagation is considered, the proposed PC-LSFE-NP can achieve better performance than the conventional LSFE due to the more reliable feedback output by the decoders. To reduce the interference from the not yet detected layers in the feedback section, we then introduce the fully-connected LSFE with noise predictor (FC-LSFE-NP), in which all layers are implicitly equalized within each stage and their decisions fed back internally. The powerful feedback filter of FC-LSFE-NP brings significant performance superiority over the conventional LSFE and PC-LSFE-NP with either perfect or imperfect feedback. Moreover, we propose a simple soft-demapper for the equalizers to avoid information loss during decoding, and thus, further improve the performance. Finally, we compare the performance of (PC/FC)-LSFE-NP with the existing schemes by computer simulations.

In this letter, a derivative constraint minimum output energy (MOE) receiver is proposed the offers enhanced robustness against carrier frequency offset (CFO). A theoretical analysis of the output signal-to-interference-plus-noise ratio (SINR) is presented to confirm its efficacy. Numerical results demonstrate that the proposed receiver basically offers the same performance as an optimal receiver with no CFO present.

At some point in a digital communications receiver, the received analog signal must be sampled. Good performance requires that these samples be taken at the right times. The process of synchronizing the sampler with the received analog waveform is known as timing recovery. Conventional timing recovery techniques perform well only when operating at high signal-to-noise ratio (SNR). Nonetheless, iterative error-control codes allow reliable communication at very low SNR, where conventional techniques fail. This paper provides a detailed review on the timing recovery strategies based on per-survivor processing (PSP) that are capable of working at low SNR. We also investigate their performance in magnetic recording systems because magnetic recording is a primary method of storage for a variety of applications, including desktop, mobile, and server systems. Results indicate that the timing recovery strategies based on PSP perform better than the conventional ones and are thus worth being employed in magnetic recording systems.

In this letter, a new approximation of log-likelihood ratio (LLR) for soft input channel decoding is proposed. Conventional simplified LLR using log-sum approximation can degrade the performance of bit interleaved coded modulation (BICM) systems employing hybrid automatic repeat request (HARQ) at low SNR. The proposed LLR performs as well as the exact LLR, and at the same time, requires only a small number of elementary operations.

This paper presents a fast method capable of accelerating the Smith-Waterman algorithm for biological database search on a cluster of graphics processing units (GPUs). Our method is implemented using compute unified device architecture (CUDA), which is available on the nVIDIA GPU. As compared with previous methods, our method has four major contributions. (1) The method efficiently uses on-chip shared memory to reduce the data amount being transferred between off-chip video memory and processing elements in the GPU. (2) It also reduces the number of data fetches by applying a data reuse technique to query and database sequences. (3) A pipelined method is also implemented to overlap GPU execution with database access. (4) Finally, a master/worker paradigm is employed to accelerate hundreds of database searches on a cluster system. In experiments, the peak performance on a GeForce GTX 280 card reaches 8.32 giga cell updates per second (GCUPS). We also find that our method reduces the amount of data fetches to 1/140, achieving approximately three times higher performance than a previous CUDA-based method. Our 32-node cluster version is approximately 28 times faster than a single GPU version. Furthermore, the effective performance reaches 75.6 giga instructions per second (GIPS) using 32 GeForce 8800 GTX cards.

Feedback shift registers are basic building blocks for many cryptographic primitives. Due to the insecurities of Linear Feedback Shift Register (LFSR) based systems, the use of Nonlinear Feedback Shift Registers (NFSRs) became more popular. In this work, we study the feedback functions of NFSRs with period 2n. First, we provide two new necessary conditions for feedback functions to be maximum length. Then, we consider NFSRs with k-monomial feedback functions and focus on two extreme cases where k=4 and k=2n-1. We study construction methods for these special cases.

This paper presents a wide range DCO with fine discrete tuning steps using a ΣΔ modulation technique for UWB frequency synthesizer. The proposed discrete tuning scheme provides a low effective frequency resolution without any degradation of the phase noise performance. With its three step discrete tunings, the DCO simultaneously has a wide tuning range and fine tuning steps. The frequency synthesizer was implemented using 0.13 µm CMOS technology. The tuning range of the DCO is 5.8-6.8 GHz with an effective frequency resolution of 3.9 kHz. It achieves a measured phase noise of -108 dBc/Hz at 1 MHz offset and a tuning range of 16.8% with the power consumption of 5.9 mW. The figure of merit with the tuning range is -181.5 dBc/Hz.

Broadband wireless access networks are promising technology for providing better end user services. For such networks, designing a scheduling algorithm that fairly allocates the available bandwidth to the end users and maximizes the overall network throughput is a challenging task. In this paper, we develop a centralized fair scheduling algorithm for IEEE 802.16 mesh networks that exploits the spatio-temporal bandwidth reuse to further enhance the network throughput. The proposed mechanism reduces the length of a transmission round by increasing the number of non-contending links that can be scheduled simultaneously. We also propose a greedy algorithm that runs in polynomial time. Performance of the proposed algorithms is evaluated by extensive simulations. Results show that our algorithms achieve higher throughput than that of the existing ones and reduce the computational complexity.