Super resolution (SR) reconstruction is the process of fusing a sequence of low-resolution images into one high-resolution image. Many researchers have introduced various SR reconstruction methods. However, these traditional methods are limited in the extent to which they allow recovery of high-frequency information. Moreover, due to the self-similarity of face images, most of the facial SR algorithms are machine learning based. In this paper, we introduce a facial SR algorithm that combines learning-based and regularized SR image reconstruction algorithms. Our conception involves two main ideas. First, we employ separated frequency components to reconstruct high-resolution images. In addition, we separate the region of the training face image. These approaches can help to recover high-frequency information. In our experiments, we demonstrate the effectiveness of these ideas.

In this paper, we study the problem of joint resource allocation in the two-way relay system, where a pair of multi-antenna users wish to exchange information via multi-antenna amplify-and-forward relay under orthogonal frequency-division multiplexing (OFDM) modulation. We formulate a sum-rate maximization problem subject to a limited power constraint for each user and relay. Our resource allocation strategy aims at finding the best pairing scheme and optimal power allocation over subchannels in frequency and space domains. This turns out to be a mixed integer programming problem. We then derive an asymptotically optimal solution though the Lagrange dual decomposition approach. Finally, simulation results are provided to demonstrate the performance gain of the proposed algorithms.

Recently, probabilistic topic models have been applied to various types of data, including text, and their effectiveness has been demonstrated. Latent Dirichlet allocation (LDA) is a well known topic model. Variational Bayesian inference or collapsed Gibbs sampling is often used to estimate parameters in LDA; however, these inference methods incur high computational cost for large-scale data. Therefore, highly efficient technology is needed for this purpose. We use parallel computation technology for efficient collapsed Gibbs sampling inference for LDA. We assume a symmetric multiprocessing (SMP) cluster, which has been widely used in recent years. In prior work on parallel inference for LDA, either MPI or OpenMP has often been used alone. For an SMP cluster, however, it is more suitable to adopt hybrid parallelization that uses message passing for communication between SMP nodes and loop directives for parallelization within each SMP node. We developed an MPI/OpenMP hybrid parallel inference method for LDA, and evaluated the performance of the inference under various settings of an SMP cluster. We further investigated the approximation that controls the inter-node communications, and found out that it achieved noticeable increase in inference speed while maintaining inference accuracy.

In this paper, several spiral inductors with various ground clearance structures and turns were investigated to achieve noise suppression up to the fourth harmonic (3.2 GHz) regime of DDR3-1600. Their performances were characterized in terms of their capability to effectively suppress simultaneous switching noise (SSN) in the frequency region of interest. For a wider noise suppression bandwidth, a spiral inductor with large ground clearance, which provides a high self resonance frequency (SRF) as well as high inductances, was implemented. The proposed spiral inductor exhibited good noise suppression characteristics in the frequency domain and achieved 50% voltage fluctuation reduction in the time domain, compared to the identical 4-turn spiral without pattern ground structure.

In this paper, we consider the XPath satisfiability problem under restricted DTDs called “duplicate free”. For an XPath expression q and a DTD D, q is satisfiable under D if there exists an XML document t such that t is valid against D and that the answer of q on t is nonempty. Evaluating an unsatisfiable XPath expression is meaningless, since such an expression can always be replaced by an empty set without evaluating it. However, it is shown that the XPath satisfiability problem is intractable for a large number of XPath fragments. In this paper, we consider simple XPath fragments under two restrictions: (i) only a label can be specified as a node test and (ii) operators such as qualifier ([]) and path union (∪) are not allowed. We first show that, for some small XPath fragments under the above restrictions, the satisfiability problem is NP-complete under DTDs without any restriction. Then we show that there exist XPath fragments, containing the above small fragments, for which the satisfiability problem is in PTIME under duplicate-free DTDs.

A method of calculating the exact top event probability of a fault tree with dynamic gates and repeated basic events is proposed. The top event probability of such a dynamic fault tree is obtained by converting the tree into an equivalent Markov model. However, the Markov-based method is not realistic for a complex system model because the number of states that should be considered in the Markov analysis increases explosively as the number of basic events in the model increases. To overcome this shortcoming, we propose an alternative method in this paper. It is a hybrid of a Bayesian network (BN) and an algebraic technique. First, modularization is applied to a dynamic fault tree. The detected modules are classified into two types: one satisfies the parental Markov condition and the other does not. The module without the parental Markov condition is replaced with an equivalent single event. The occurrence probability of this event is obtained as the sum of disjoint sequence probabilities. After the contraction of modules without parent Markov condition, the BN algorithm is applied to the dynamic fault tree. The conditional probability tables for dynamic gates are presented. The BN is a standard one and has hierarchical and modular features. Numerical example shows that our method works well for complex systems.

This paper presents a method for learning an overcomplete, nonnegative dictionary and for obtaining the corresponding coefficients so that a group of nonnegative signals can be sparsely represented by them. This is accomplished by posing the learning as a problem of nonnegative matrix factorization (NMF) with maximization of the incoherence of the dictionary and of the sparsity of coefficients. By incorporating a dictionary-incoherence penalty and a sparsity penalty in the NMF formulation and then adopting a hierarchically alternating optimization strategy, we show that the problem can be cast as two sequential optimal problems of quadratic functions. Each optimal problem can be solved explicitly so that the whole problem can be efficiently solved, which leads to the proposed algorithm, i.e., sparse hierarchical alternating least squares (SHALS). The SHALS algorithm is structured by iteratively solving the two optimal problems, corresponding to the learning process of the dictionary and to the estimating process of the coefficients for reconstructing the signals. Numerical experiments demonstrate that the new algorithm performs better than the nonnegative K-SVD (NN-KSVD) algorithm and several other famous algorithms, and its computational cost is remarkably lower than the compared algorithms.

A use of ultra wideband (UWB) technology within spacecrafts has been proposed with a view to partially replacing wired interface buses with wireless connections. Adoption of wireless technologies within the spacecrafts could contribute to reduction in cable weight (and launching cost as a result), reduction in the cost of manufacture, more flexibility in layout of spacecraft subsystems, and reliable connections at rotary, moving, and sliding joints. However, multipath propagation in semi-closed conductive enclosures, such as spacecrafts, restricts the link performance. In this paper, UWB and narrowband propagation were measured in a UWB frequency band (from 3.1 to 10.6 GHz, the full-band UWB approved in the United States) within a small spacecrafts and a shield box of the same size. While narrowband propagation resulted in considerable spatial variations in propagation gain due to interferences caused by multipath environments, UWB yielded none. This implies that the UWB systems have an advantage over narrowband from a viewpoint of reducing fading margins. Throughputs exceeding 80 Mb/s were obtained by means of commercially-available UWB devices in the spacecraft. Path gains and throughputs were also measured for various antenna settings and polarizations. Polarization configurations were found to produce almost no effect on average power delay profiles and substantially small effects on the throughputs. Significantly long delay spreads and thus limited link performance are caused by a conductive enclosure (the shield box) without apertures on the surfaces. Even in such an environment, it was found that delay spreads can be suppressed by partially paneling a radio absorber on the inner surfaces. More than 96 Mb/s throughputs were attained when the absorber panel covered typically 4% of the total inner surface area.

This paper presents a method of designing transmission line couplers (TLCs) and a mixer-based receiver for dicode partial response communications. The channel design method results in the optimum TLC design. The receiver with mixers and DC balancing circuits reduces the threshold control circuits and digital circuits to decode dicode partial response signals. Our techniques enable low inter-symbol interference (ISI) dicode partial response communications without three level decision circuits and complex threshold control circuits. The techniques were evaluated in a simulation with an EM solver and a transistor level simulation. The circuit was designed in the 90-nm CMOS process. The simulation results show 12-Gb/s operation and 52mW power consumption at 1.2V.

Recursive least squares-based online nonnegative matrix factorization (RLS-ONMF), an NMF algorithm based on the RLS method, was developed to solve the NMF problem online. However, this method suffers from a partial-data problem. In this study, the partial-data problem is resolved by developing an improved online NMF algorithm using RLS and a sparsity constraint. The proposed method, RLS-based online sparse NMF (RLS-OSNMF), consists of two steps; an estimation step that optimizes the Euclidean NMF cost function, and a shaping step that satisfies the sparsity constraint. The proposed algorithm was evaluated with recorded speech and music data and with the RWC music database. The results show that the proposed algorithm performs better than conventional RLS-ONMF, especially during the adaptation process.

This paper clarifies the effects of metal wires placed around a Multiple-Input-Multiple-Output (MIMO) array with the goal of improving the channel capacity in near-field MIMO systems. Tests are performed on dual-dipole arrays with metal wires placed parallel to the dipoles. If the antenna elements have an appropriate half-power beamwidth (HPBW), there is a clear improvement in the channel capacity of the dual-dipole array. The metal wires are used to increase the multipath richness and the locations of the wires significantly impact the channel capacity. A significant increase in the channel capacity is observed even if only one metal wire is placed in the proper location. We verified the generality of applying a metal wire to improve the channel capacity and that the improvement in the channel capacity is approximately proportional to the number of metal wires.

The trench-gate type high-voltage (HV) MOSFET is one of the variants of HV-MOSFET, typically with its utility segments lying on a larger power consumption domain, compared to planar HV-MOSFETs. In this work, the HiSIM_HV compact model, originally intended for planar LDMOSFETs, was adequately extended to accommodate trench-gate type HV-MOSFETs. The model formulation focuses on a closed-form description of the current path in the highly resistive drift region, specific to the trench-gate HV-MOSFETs. It is verified that the developed compact expression can capture the conductivity in the drift region, which varies with voltage bias and device technology such as trench width. The notable enhancement of current drivability can be accounted for by the electrostatic control exerted by the trench gate within the framework of this model.

Turbo equalization is an iterative equalization and decoding technique that can achieve impressive performance gains for communication systems. In this letter, we investigate the turbo equalization method for the decoding of the Davey-MacKay (DM) construction over the IDS-AWGN channels, which indicates a cascaded insertion, deletion, substitution (IDS) channel and an additive white Gaussian noise (AWGN) channel. The inner decoder for the DM construction can be seen as an maximum a-posteriori (MAP) detector. It receives the beliefs generated by the outer LDPC decoder when turbo equalization is used. Two decoding schemes with different kinds of inner decoders, namely hard-input inner decoder and soft-input inner decoder, are investigated. Simulation results show that significant performance gains are obtained for both decoders with respect to the insertion/deletion probability at different SNR values.

The nondeterminism of message-passing communication brings challenges to program debugging, testing and fault-tolerance. This paper proposes a novel deterministic message-passing implementation (DMPI) for parallel programs in the distributed environment. DMPI is compatible with the standard MPI in user interface, and it guarantees the reproducibility of message with high performance. The basic idea of DMPI is to use logical time to solve message races and control asynchronous transmissions, and thus we could eliminate the nondeterministic behaviors of the existing message-passing mechanism. We apply a buffering strategy to alleviate the performance slowdown caused by mismatch of logical time and physical time. To avoid deadlocks introduced by deterministic mechanisms, we also integrate DMPI with a lightweight deadlock checker to dynamically detect and solve these deadlocks. We have implemented DMPI and evaluated it using NPB benchmarks. The results show that DMPI could guarantee determinism with incurring modest runtime overhead (14% on average).

A highly parallel turbo decoder for 3GPP LTE/LTE-Advanced systems is presented. It consists of 32 radix-4 soft-in/soft-out (SISO) decoders. Each SISO decoder is based on the proposed full-parallel sliding window (SW) schedule. Implemented in a 0.13 µm CMOS technology, the proposed design occupies 12.96 mm2 and achieves 1.5 Gb/s while decoding size-6144 blocks with 5.5 iterations. Compared with conventional SW schedule, the throughput is improved by 30–76% with 19.2% area overhead and negligible energy overhead.

A novel tunneling field-effect transistor (TFET) featuring the sigma-shape embedded SiGe sources and recessed channel is proposed. The gate facing the source effectively focuses the E-field at the tip of the source and eliminates the gradual turn-on issue of planar TFETs. The fabrication scheme modified from the state-of-the-art 45 nm/32 nm CMOS technology flows provides a unique benefit in the co-integrability and the control of ID-VGS characteristics. The feasibility is verified with TCAD process simulation of the device with 14 nm of the gate dimension. The device simulation shows 5-order change in the drain current with a gate bias change less than 300 mV.

The average Hamming correlation is an important performance indicator of frequency-hopping sequences (FHSs). In this letter, the average partial Hamming correlation (APHC) properties of FHSs are discussed. Firstly, the theoretical bound on the average partial Hamming correlation of FHSs is established. It works for any correlation window with length 1≤ω≤υ, where υ is the sequence period, and generalizes the bound developed by Peng et al which is valid only when ω=υ. A sufficient and necessary condition for a set of FHSs having optimal APHC for any correlation window is then given. Finally, sets of FHSs with optimal APHC are presented.

The performance of a mobile database management system (DBMS) in which most queries are made up of random data accesses if the NAND flash memory is used as storage media of the DBMS is degraded. The reason for this is that the performance of NAND flash memory is good for writing sequentially but poor when writing randomly. Thus, a new storage structure and querying policies are needed in mobile DBMS when flash memory is used as the storage media. In this letter, we propose a new policy of database page management to enhance the frequent random update performance, and then evaluate the performance experimentally.

In forthcoming sensor networks, a multitude of sensor nodes deployed over a large geographical area for monitoring traffic, climate, etc. are expected to become an inevitable infrastructure. Clustering algorithms play an important role in aggregating a large volume of data that are produced continuously by the huge number of sensor nodes. In such networks, equal-sized multi-hop clusters which include an equal number of nodes are useful for efficiency and resiliency. In addition, scalability is important in such large-scale networks. In this paper, we mathematically design a decentralized equal-sized clustering algorithm using a partial differential equation based on the Fourier transform technique, and then design its protocol by discretizing the equation. We evaluated through simulations the equality of cluster sizes and the resiliency against packet loss and node failure in two-dimensional perturbed grid topologies.

High-resolution time of arrival (TOA) estimation techniques have great promise for the high range resolution required in recently developed radar systems. A widely known super-resolution TOA estimation algorithm for such applications, the multiple-signal classification (MUSIC) in the frequency domain, has been proposed, which exploits an orthogonal relationship between signal and noise eigenvectors obtained by the correlation matrix of the observed transfer function. However, this method suffers severely from a degraded resolution when a number of highly correlated interference signals are mixed in the same range gate. As a solution for this problem, this paper proposes a novel TOA estimation algorithm by introducing a maximum likelihood independent component analysis (MLICA) approach, in which multiple complex sinusoidal signals are efficiently separated by the likelihood criteria determined by the probability density function (PDF) of a complex sinusoid. This MLICA schemes can decompose highly correlated interference signals, and the proposed method then incorporates the MLICA into the MUSIC method, to enhance the range resolution in richly interfered situations. The results from numerical simulations and experimental investigation demonstrate that our proposed pre-processing method can enhance TOA estimation resolution compared with that obtained by the original MUSIC, particularly for lower signal-to-noise ratios.