This paper proposes a hybrid multiband (MB) ultra wideband (UWB) system with direct sequence (DS) spreading. The theoretical error analysis for the DS-MB-UWB multiple access system with Rake receiver in the presence of multipath and narrowband interference is developed. The developed theoretical framework models the multiple access interference (MAI), multipath interference (MI) and narrowband interference for the designed UWB system. It is shown that the system error performance corresponding to the combining effects of these interference can be accurately modeled and calculated. Monte Carlo simulation results are provided to validate the accuracy of the model. Additionally, it is found that narrowband interference can be mitigated effectively in the multiband UWB system by suppressing the particular UWB sub-band co-existing with the interfering narrowband signal. A typical improvement of 5 dB can be achieved with 75% sub-band power suppression. On the other hand, suppression of UWB sub-band is also found to decrease frequency diversity, thus facilitating the increase of MAI. In this paper, the developed model is utilized to determine the parameters that optimize the UWB system performance by minimizing the effective interference.

We present a novel model named Integrated Latent Topic Model (ILTM), to learn and recognize natural scene category. Unlike previous work, which considered the discrepancy and common property separately among all categories, Our approach combines universal topics from all categories with specific topics from each category. As a result, the model is implemented to produce a few but specific topics and more generic topics among categories, and each category is represented in a different topics simplex, which correlates well with human scene understanding. We investigate the classification performance with variable scene category tasks. The experiments have shown our model outperforms latent-space methods with less training data.

This paper describes an evolution and standardization trends of the wireless channel modeling activities towards IMT-Advanced. After a background survey on various channel modeling approaches is introduced, two well-known multiple-input-multiple-output (MIMO) channel models for cellular systems, namely, the 3GPP/3GPP2 Spatial Channel Model (SCM) and the IMT-Advanced MIMO Channel Model (IMT-Adv MCM) are compared, and their main similarities are pointed out. The performance of MIMO systems is greatly influenced by the spatial-temporal correlation properties of the underlying MIMO channels. Here, we investigate the spatial-temporal correlation characteristics of the 3GPP/3GPP2 SCM and the IMT-Adv MCM in term of their spatial multiplexing and spatial diversity gains. The main goals of this paper are to summarize the current state of the art, as well as to point out the gaps in the wireless channel modeling works, and thus hopefully to stimulate research in these areas.

This paper describes a novel algorithm for approximate nearest neighbor searching. For solving this problem especially in high dimensional spaces, one of the best-known algorithm is Locality-Sensitive Hashing (LSH). This paper presents a variant of the LSH algorithm that outperforms previously proposed methods when the dataset consists of vectors normalized to unit length, which is often the case in pattern recognition. The LSH scheme is based on a family of hash functions that preserves the locality of points. This paper points out that for our special case problem we can design efficient hash functions that map a point on the hypersphere into the closest vertex of the randomly rotated regular polytope. The computational analysis confirmed that the proposed method could improve the exponent ρ, the main indicator of the performance of the LSH algorithm. The practical experiments also supported the efficiency of our algorithm both in time and in space.

We investigate numerically the applicability of photonic crystal fiber (PCF) with a uniform air hole structure as a wide-band transmission medium. We show that accumulated dispersion over the PCF can be reduced effectively by optimizing the index profile of dispersion compensating fiber (DCF). We also show that a bandwidth of more than 300 nm will be available for 40 Gbit/s NRZ transmission by using the PCF as a transmission medium instead of conventional 1.3 µm zero-dispersion single-mode fiber (SMF).

In modern VLSI physical design, huge integration scale necessitates hierarchical design and IP reuse to cope with design complexity. Besides, interconnect delay becomes dominant to overall circuit performance. These critical factors require some modules to be placed along designated boundaries to effectively facilitate hierarchical design and interconnection optimization related problems. In this paper, boundary constraints of general floorplan are solved smoothly based on the novel representation Single-Sequence (SS). Necessary and sufficient conditions of rooms along specified boundaries of a floorplan are proposed and proved. By assigning constrained modules to proper boundary rooms, our proposed algorithm always guarantees a feasible SS code with appropriate boundary constraints in each perturbation. Time complexity of the proposed algorithm is O(n). Experimental results on MCNC benchmarks show effectiveness and efficiency of the proposed method.

One of the most important challenges in speaker recognition is intersession variability (ISV), primarily cross-channel effects. Recent NIST speaker recognition evaluations (SRE) include a multilingual scenario with training conversations involving multilingual speakers collected in a number of other languages, leading to further performance decline. One important reason for this is that more and more researchers are using phonetic clustering to introduce high level information to improve speaker recognition. But such language dependent methods do not work well in multilingual conditions. In this paper, we study both language and channel mismatch using a support vector machine (SVM) speaker recognition system. Maximum likelihood linear regression (MLLR) transforms adapting a universal background model (UBM) are adopted as features. We first introduce a novel language independent statistical binary-decision tree to reduce multi-language effects, and compare this data-driven approach with a traditional knowledge based one. We also construct a framework for channel compensation using feature-domain latent factor analysis (LFA) and MLLR supervector kernel-based nuisance attribute projection (NAP) in the model-domain. Results on the NIST SRE 2006 1conv4w-1conv4w/mic corpus show significant improvement. We also compare our compensated MLLR-SVM system with state-of-the-art cepstral Gaussian mixture and SVM systems, and combine them for a further improvement.

Transform each coordinate of the realizations of several random variables (RVs) by the distribution function of the corresponding RV and partition the range space into a uniform grid. The expected number of occupied grid-boxes will be greatest when these RVs are independent. Based on this fact, we propose a novel measure of independence named grid occupancy (GO). We also address the problem of how to make optimum selection of the parameters in GO, i.e., the number of observations and the number of quantization levels. In addition, we apply GO to independent component analysis (ICA). The GO based ICA algorithm can separate signals with arbitrary continuous distributions and favors digital hardware implementation.

Shamir's (k,n)-threshold secret sharing scheme (threshold scheme) has two problems: a heavy computational cost is required to make shares and recover the secret, and a large storage capacity is needed to retain all the shares. As a solution to the heavy computational cost problem, several fast threshold schemes have been proposed. On the other hand, threshold ramp secret sharing schemes (ramp scheme) have been proposed in order to reduce each bit-size of shares in Shamir's scheme. However, there is no fast ramp scheme which has both low computational cost and low storage requirements. This paper proposes a new (k,L,n)-threshold ramp secret sharing scheme which uses just EXCLUSIVE-OR(XOR) operations to make shares and recover the secret at a low computational cost. Moreover, by proving that the fast (k,n)-threshold scheme in conjunction with a method to reduce the number of random numbers is an ideal secret sharing scheme, we show that our fast ramp scheme is able to reduce each bit-size of shares by allowing some degradation of security similar to the existing ramp schemes based on Shamir's threshold scheme.

In multiuser MIMO-BC (Multiple-Input Multiple-Output Broadcasting) systems, user selection is important to achieve multiuser diversity. The optimal user selection algorithm is to try all the combinations of users to find the user group that can achieve the multiuser diversity. Unfortunately, the high calculation cost of the optimal algorithm prevents its implementation. Thus, instead of the optimal algorithm, some suboptimal user selection algorithms were proposed based on semiorthogonality of user channel vectors. The purpose of this paper is to achieve multiuser diversity with a small amount of calculation. For this purpose, we propose a user selection algorithm that can improve the orthogonality of a selected user group. We also apply a channel prediction technique to a MIMO-BC system to get more accurate channel information at the transmitter. Simulation results show that the channel prediction can improve the accuracy of channel information for user selections, and the proposed user selection algorithm achieves higher sum rate capacity than the SUS (Semiorthogonal User Selection) algorithm. Also we discuss the setting of the algorithm threshold. As the result of a discussion on the calculation complexity, which uses the number of complex multiplications as the parameter, the proposed algorithm is shown to have a calculation complexity almost equal to that of the SUS algorithm, and they are much lower than that of the optimal user selection algorithm.

In this letter, we present an automatic approach of objective singing performance evaluation for untrained singers by relating acoustic measurements to perceptual ratings of singing voice quality. Several acoustic parameters and their combination features are investigated to find objective correspondences of the perceptual evaluation criteria. Experimental results show relative strong correlation between perceptual ratings and the combined features and the reliability of the proposed evaluation system is tested to be comparable to human judges.

A group signature scheme introduced by Chaum and Van Heyst allows a group member to sign messages anonymously on behalf of the group. However, in the case of a dispute, the identity of a signer of a group signature can be revealed only by a privileged entity, called a group manager. The group signature scheme has mainly been studied from the viewpoint of computational security setting so far. The main contribution of this paper is to study group signature schemes in unconditional security. More specifically, we newly introduce strong security notions of unconditionally secure group signatures (USGS for short) based on the idea of those of computationally secure group signatures proposed by Bellare, Micciancio and Warinschi. We also provide a generic method to construct USGS that is provably secure in our security definition. More precisely, we construct USGS by combining an encryption scheme with a signature, and show that the constructed scheme is unconditionally secure if the encryption and the signature used in the construction are unconditionally secure. Finally, we provide an instantiation of the one-time secure group signature scheme based on the generic construction.

In OFDM systems, the estimation/correction of carrier frequency offset (CFO) is crucial to maintain orthogonality among subcarriers. However, the CFO estimation suffers from DC offset (DCO) generated in low-cost direct-conversion receivers (DCRs). More seriously, in practice, DCO is time-varying due to the automatic gain control. In this paper, a novel CFO estimator in the presence of time-varying DCO is proposed. It is shown the residual DCO after high-pass filtering varies in a linear fashion. Based on this observation and the periodicity of the training sequence, we derive a CFO estimator independent of DCO. Also, the residual DCO can be estimated, using the obtained CFO. The validity of the proposed estimation method is demonstrated by simulations.

In this paper, a new approach to channel order selection of single-input multiple-output (SIMO), finite impulse response (FIR) channels is proposed for blind channel estimation. The approach utilizes cross spectral density (CSD) of the channel outputs, and minimizes the distance between two CSD's, one calculated non-parametrically from the observed output data, and the other calculated from the blindly estimated channel parameters. The CSD criterion is numerically tested on randomly generated SIMO-FIR channels, and shown to be very effective compared to existing channel order selection methods especially under low SNR settings. Blind estimates of the channels with the selected channel order also show superiority of the CSD criterion.

Single-carrier ultra-wideband (SC-UWB) is weak due to the problem of serious inter-symbol interference (ISI), which is generated in dense multipath with a long root-mean-square (RMS) delay spread. The selective RAKE (SRAKE) based RAKE-decision feedback equalizer (RAKE-DFE) receiver is usually employed to combat ISI in practical SC-UWB systems. Considering the system complexity, however, the number of RAKE fingers is usually small. In this case, conventional RAKE-DFE receivers can hardly collect enough energy to achieve a good performance. In this paper, the optimum SRAKE based RAKE-DFE receiver was proposed as a solution. Theoretical analysis and simulations are presented. Results and conclusions show that the proposed SRAKE scheme is optimum to collect energy of multipath. Moreover, the proposed RAKE-DFE receiver outperforms conventional RAKE-DFE receiver by about 1 dB, but the complexity for them both is almost the same.

In this paper, we present a forecasting method for the view of Mt. Fuji as an application of Earth observation data (EOD) obtained by satellites. We defined the Mt. Fuji viewing index (FVI) that characterises how well the mountain looks on a given day, based on photo data produced by a fixed-point observation. A long-term predictor of FVI, ranging from 0 to 30 days, was constructed through support vector machine regression on climate and earth observation data. It was found that the aerosol mass concentration (AMC) improves prediction performance, and such performance is particularly significant in the long-term range.

In this paper, we propose a new feature selection algorithm for multi-class classification. The proposed algorithm is based on Gaussian mixture models (GMMs) of the features, and it uses the distance between the two least separable classes as a metric for feature selection. The proposed system was tested with a support vector machine (SVM) for multi-class classification of music. Results show that the proposed feature selection scheme is superior to conventional schemes.

Using the transient current switch circuit in parallel with the energizing switching contacts for timely control of breaking operation, the increase of contact voltage is suppressed at the last stage of the breaking of electric contacts. Breaking contact voltage Vc and current Ic of electromagnetic relays with Ag contacting electrodes were measured with 12.5-50 V and 0.1-20 A for two hinge springs (Spring constants; 2 N/mm and 0.2 N/mm). The current-decreasing process was clearly measured at the melting voltage Um. After Vc=Um, the breaking time of contact current did not depend on mechanical motion controlled by the two hinge springs and energizing power-supply voltage, but depended on the contact current. The residue of melt electrode was observed optically as a white fusion spot, with radius depending on the energizing current.

This letter proposes a type of mutually cooperative relaying (MCR) scheme based on a spatially coordinate-interleaved orthogonal design (SCID), in which two cooperative users are spatially multiplexed without bandwidth expansion. It provides not only diversity gain (with order of two) as in the existing MCR scheme, but also additional coding gain. Our simulation results demonstrate that the proposed SCID scheme is useful for improving the uplink performance as long as one user can find another active user as a close neighbor that is simultaneously communicating with the same destination, e.g., a base station in the cellular network.

In this paper, a method for adaptive reconstruction of missing textures based on kernel canonical correlation analysis (CCA) with a new clustering scheme is presented. The proposed method estimates the correlation between two areas, which respectively correspond to a missing area and its neighboring area, from known parts within the target image and realizes reconstruction of the missing texture. In order to obtain this correlation, the kernel CCA is applied to each cluster containing the same kind of textures, and the optimal result is selected for the target missing area. Specifically, a new approach monitoring errors caused in the above kernel CCA-based reconstruction process enables selection of the optimal result. This approach provides a solution to the problem in traditional methods of not being able to perform adaptive reconstruction of the target textures due to missing intensities. Consequently, all of the missing textures are successfully estimated by the optimal cluster's correlation, which provides accurate reconstruction of the same kinds of textures. In addition, the proposed method can obtain the correlation more accurately than our previous works, and more successful reconstruction performance can be expected. Experimental results show impressive improvement of the proposed reconstruction technique over previously reported reconstruction techniques.