The ability to estimate a target location is essential in many applications of wireless sensor networks. Received signal strength indicator (RSSI)-based maximum likelihood (ML) method in a wireless sensor network usually requires a pre-determined statistical model on the variation of RSSI in a sensing area and uses it as an ML function when estimating the location of a target in the sensing area. However, when estimating the location of a target, due to several reasons, we often measure the RSSIs which do not follow the statistical model, in other words, which are outlier on the statistical model. As the result, the effect of the outlier RSSI data worsens the estimation accuracy. If the wireless sensor network has a lot of sensor nodes, we can improve the estimation accuracy intentionally rejecting such outlier RSSIs. In this paper, we propose a simple outlier RSSI data rejection algorithm for an ML location estimation. The proposed algorithm iteratively eliminates the anchor nodes which measure outlier RSSIs. As compared with the location estimation methods with previously proposed outlier RSSI data rejection algorithms, our proposed method performs better with much less computational complexity.

We propose a selective detection scheme based on pulse repetition considering both BER (Bit Error Rate) performance and complexity in coherent UWB (Ultra Wide Band) systems. To take system complexity into account, the proposed scheme transmits the UWB signals by pulse repetition at the transmitter, like conventional PRC (Pulse Repetition Coding). However, to effectively improve BER performance of the system, the proposed scheme performs selective detection by estimating the SNR (Signal-to-Noise Ratio) of the received pulse-repeated signal at the UWB receiver.

Conveying information about who, what, when and where is a primary purpose of some genres of documents, typically news articles. Statistical models that capture dependencies between named entities and topics can play an important role in handling such information. Although some relationships between who and where should be mentioned in such a document, no statistical topic models explicitly address the textual interactions between a who-entity and a where-entity. This paper presents a statistical model that directly captures the dependencies between an arbitrary number of word types, such as who-entities, where-entities and topics, mentioned in each document. We show that this multitype topic model performs better at making predictions on entity networks, in which each vertex represents an entity and each edge weight represents how a pair of entities at the incident vertices is closely related, through our experiments on predictions of who-entities and links between them. We also demonstrate the scale-free property in the weighted networks of entities extracted from written mentions.

Various types of data from environment are used to provide user-adaptive services. Among them, data of a user's past moving paths are useful to predict a moving user's next location and provide related services. This paper proposes a method to predict a moving user's location through analyzing his/her past moving paths. This method analyzes the user's moving path by using three elements of distance, time and direction of moving paths and Dynamic Time Warping (DTW), and selects the representative path, which is the one most similar to the current moving path in the past paths. The selected path can be used to provide service like space and time estimation.

Ultra wideband (UWB) radar cross sections (RCSs) of several targets have been measured using various combinations of transmitting and receiving linear polarizations (V-V, H-H, and +45- -45) with a view to obtaining information on the design of vehicular short-range radars. This paper reports the UWB RCSs (σLR and σLL) of a typical passenger automobile using two circular polarization combinations (L and R denote left and right circular polarizations). The wideband measurements were carried out with use of a vector network analyzer by sweeping the frequency from 24.5 to 28.8 GHz in a radio anechoic chamber. The UWB RCSs were derived by integrating the received power in the frequency domain. Similar to the linear polarization results, fluctuations of the RCSs were smaller in the UWB than in narrowband for both L-R and L-L, because the ultra-wide bandwidth cancels out RCS plunges caused by narrowband interference among reflected waves from various facets of the target. The median of (σLR- σLL) was 2 dB, while the median of (σHH - σ+45 - -45) or (σVV-σ+45- -45) was 6 dB. This is because the body of the automobile comprises a number of smaller scattering objects yielding σ LL, either similar to the corner reflectors or asymmetrical to the radar boresight. Frequency-domain responses showed a number of notches caused by the interference between numerous reflecting waves having power levels of a similar order and different round-trip path lengths. Some representative reflective parts of the automobile were identified through analyses of time-domain responses.

In this letter we propose a robust detection algorithm for audio watermarking for copyright protection. The watermark is embedded in the time domain of an audio signal by the normally used spread spectrum technique. The scheme of detection is an improvement of the conventional correlation detector. A high-pass filter is applied along with the linear prediction error filter for whitening the audio signal and an adaptive threshold is chosen for decision comparing. Experimental results show that our detection algorithm outperforms the conventional one not only because it improves the robustness to normal attacks but also because it can provide the robustness to time-invariant pitch-scale modification.

This paper represents a novel second harmonic power combining oscillator using mutually synchronized Gunn diodes embedded on slot line resonators. A both-sided MIC technology is adopted in the oscillator. The oscillator consists of Gunn diodes, slot line resonators and microstrip lines. By embedding Gunn diodes on the slot line resonators, the harmonic RF signal can be generated very easily. The microstrip lines are used for the power combining output circuit. This oscillator has advantages such as easy circuit design, simple circuit configuration and miniaturization of the circuit size. The second harmonic oscillator is designed and fabricated in K-Band. The output power is +5.75 dBm at the design frequency of 19.0 GHz (2f0) with the phase noise of -111.7 dBc/Hz at the offset frequency of 1 MHz. Excellent suppression of the undesired fundamental frequency signal (f0) of -39 dBc is achieved.

In March, 2007, IEEE802.15.4a was standardized as a low-rate and low-power UWB system for sensor networks. In general, detection of the IEEE802.15.4a signal is considered to be difficult because of its low transmitting power density and low duty cycle. However, if detecting of the IEEE802.15.4a signal is available, it is possible to avoid interference issues both among the IEEE802.15.4a systems and between the 15.4a and other UWB systems. This letter proposes a simple detection method using non-coherent detectors. The possibility of detecting of the IEEE802.15.4a signal by proposal detection method was examined. By conducting experiments with an emulated 15.4a RF signal, the signal detection probability was examined, and 15.4a signal from the range of about 11 meters in the radius could be detected. From this observation, the CSMA/CA method with detecting the signal in 15.4a system may be applied for alternative access method for 15.4a systems.

A system which employs a stylus as an input device is suitable for creative activities like writing and painting. However, such a system does not always provide the user with a GUI that is easy to operate using the stylus. In addition, system usability is diminished because the stylus is not always integrated into the system in a way that takes into consideration the features of a pen. The purpose of our research is to improve the usability of a system which uses a stylus as an input device. We propose shortcut actions, which are interaction techniques for operation with a stylus that are controlled through a user's hand motions made in the air. We developed the Context Sensitive Stylus as a device to implement the shortcut actions. The Context Sensitive Stylus consists of an accelerometer and a conventional stylus. We also developed application programs to which we applied the shortcut actions; e.g., a drawing tool, a scroll supporting tool, and so on. Results from our evaluation of the shortcut actions indicate that users can concentrate better on their work when using the shortcut actions than when using conventional menu operations.

Recently, multi-car elevator (MCE) system has captured attention as an effective mean for the improvement of transportation capability in a high-rise building. The MCE has two or more cars in one shaft. In this paper, we propose an algorithm for group controllers of MCE system, and show the effectiveness of our algorithm through computer simulation.

In this paper, we propose a new algorithm to detect the presence of narrow band interference signals on the band of an Ultra Wide-Band (UWB) system when the UWB spectrum overlaps the bands of other narrow band wireless services. In our proposed algorithm for an UWB Multi-Band Orthogonal Frequency Division Multiplexing (MB-OFDM) system, an appropriate model is selected from the assumed interference models based on the Akaike Information Criterion (AIC) which is an explicit theoretic criterion and a measure of fit of the model. The proposed algorithm does not need a priori information on the interference signals except that we can reduce a computational complexity to implement the algorithm if we have knowledge of the bands of the interference signals. Furthermore, we introduce the Expectation Maximization (EM) algorithm to our algorithm in order to estimate the transmitted signals and the interference signals simultaneously. The proposed algorithm may not require the pilot symbols in the assumed UWB system to detect the presence of other systems. By computer simulations, we show that the proposed algorithm validly detects the presence of interference signals on the UWB band.

In this paper, we propose a method for selecting n-mode singular vectors in higher-order singular value decomposition. We select the minimum number of n-mode singular vectors, when the upper bound of a least-squares cost function is thresholded. The reduced n-ranks of all modes of a given tensor are determined automatically and the tensor is represented with the minimum number of dimensions. We apply the selection method to simultaneous low rank approximation of matrices. Experimental results show the effectiveness of the n-mode singular vector selection method.

Compact and planar active integrated antenna arrays with a high power multi-stage amplifier were developed with effective heat sink mechanism. By attaching an aluminum plate to the backside of the creased amplifier circuit board, effective cooling can be achieved. The nonlinear behavior of the amplifier agrees well with the simulation based on the Angelov model. The high power amplifier circuit consisted of the three-stage amplifier and operated with an output power of 4 W per each element at 5.8 GHz. The 32-element active integrated antenna array stably operated with the output power of 120 W under the effective heat sink design. With a weight of 4 kg, the weight-to-output power ratio and the volume-to-output power ratio of the antenna array are 33.3 g/W and 54.5 cm3/W, respectively. Wireless power transmission was also successfully demonstrated.

This paper introduces a theory for fast optimization of the circuit topology and parameters in sinusoidal oscillators. The theory starts from a system model composed of standard active and passive elements. We then include even the output load in the circuit, so that there is no longer any interaction with the outside of the system through the port. This model is thus called no-input-no-output (NINO) oscillator. The circuit is cut at an arbitrary branch, and is characterized in terms of the scalar impedance from the cut point. This is called active impedance because it is a function of not only the stimulating frequency but also the active device gain. The oscillation frequency and necessary device gain are estimated by solving impedance-domain Barkhausen equilibrium equations. This estimation works for the adjustment of circuit elements to meet the specified oscillation frequency. The estimation of necessary device gain enables us to maximize the oscillation amplitude, thanks to the inherent negative-slope nonlinearity of active devices. The active impedance is also used to derive the oscillation Q (quality) factor, which serves as a key criterion for sideband noise minimization i.e. frequency spectrum purification. As an alternative measure to active impedance, we also introduce branch admittance matrix determinant. This has the same numerical effect as the scalar impedance but can be used to formulate oscillator characteristics in a more elegant fashion, and provides a lucent picture of the physical behavior of each element in the circuit. Based on the proposed theory, we provide the tabled formulas of oscillation frequency, necessary device gain, active Q factor for a variety of typical Colpitts, Hartley, and cross-coupled twin-FET (field-effect transistor) oscillators.

In order to timely response to a user query at run-time, next generation Semantic Web applications demand a robust mechanism to dynamically select one or more existing ontologies available on the Web and combine them automatically if needed. Although existing ontology retrieval systems return a lengthy list of resultant ontologies, they cannot identify which ones can completely meet the query requirements nor determine a minimum set of resultant ontologies that can jointly satisfy the requirements if no single ontology is available to satisfy them. Therefore, this paper presents an ontology retrieval system, namely combiSQORE, which can return single or combinative ontologies that completely satisfy a submitted query when the available ontology database is adequate to answer such query. In addition, the proposed system ranks the returned results based on their semantic similarities to the given query and their modification (integration) costs. The experimental results show that combiSQORE system yields practical combinative ontologies and useful rankings.

To understand the structural and functional properties of large-scale complex networks, it is crucial to efficiently extract a set of cohesive subnetworks as communities. There have been proposed several such community extraction methods in the literature, including the classical k-core decomposition method and, more recently, the k-clique based community extraction method. The k-core method, although computationally efficient, is often not powerful enough for uncovering a detailed community structure and it produces only coarse-grained and loosely connected communities. The k-clique method, on the other hand, can extract fine-grained and tightly connected communities but requires a substantial amount of computational load for large-scale complex networks. In this paper, we present a new notion of a subnetwork called k-dense, and propose an efficient algorithm for extracting k-dense communities. We applied our method to the three different types of networks assembled from real data, namely, from blog trackbacks, word associations and Wikipedia references, and demonstrated that the k-dense method could extract communities almost as efficiently as the k-core method, while the qualities of the extracted communities are comparable to those obtained by the k-clique method.

We report effects of washing rubbed polyimide film on the near surface. Especially we focused dependence of solvent. Rubbed polyimide films have been used as liquid crystal alignment films in Liquid crystal displays (LCDs), and in actual LCD panel fabrication washing on film surfaces after rubbing is essential process to remove dust and pollution. We investigated the effects of washing by grazing incidence X-ray diffraction (GIXD) measurements. In GIXD, the X-ray penetration into the polymer was changed from 8 nm (suface sensitive) to 4 nm (bulk sensitive) by variation of the X-ray incidence angle. It was found that crystallization near the surface induced by soaking was considerably dependent on solvent. However, in-plane distribution of the surface polymer chains of polyimide film was not found to be dependent on the solvents.

An indirectly reactive sputtering coater has been developed to deposit various high quality metallic and metal oxide films at high deposition rate. In this letter, several kinds of filters such as antireflection (AR) coating, IR-cut filter, and Rugate filter were deposited for the benchmark test of implemental capabilities. Our coater was established to be a powerful tool for both discrete multilayer and Rugate filters due to high stability and reproducibility of the refractive index and the deposition rate.

Transmission of compressed video over error prone channels may result in packet losses or errors, which can significantly degrade the image quality. Therefore an error concealment scheme is applied at the video receiver side to mask the damaged video. Considering there are 3 types of MBs (Macro Blocks) in natural video frame, i.e., Textural MB, Edged MB, and Smooth MB, this paper proposes an adaptive spatial error concealment which can choose 3 different methods for these 3 different MBs. For criteria of choosing appropriate method, 2 factors are taken into consideration. Firstly, standard deviation of our proposed edge statistical model is exploited. Secondly, some new features of latest video compression standard H.264/AVC, i.e., intra prediction mode is also considered for criterion formulation. Compared with previous works, which are only based on deterministic measurement, proposed method achieves the best image recovery. Subjective and objective image quality evaluations in experiments confirmed this.

This paper describes a method for recognizing romanized Japanese words in learner English. They become noise and problematic in a variety of systems and tools for language learning and teaching including text analysis, spell checking, and grammatical error detection because they are Japanese words and thus mostly unknown to such systems and tools. A problem one encounters when recognizing romanized Japanese words in learner English is that the spelling rules of romanized Japanese words are often violated. To address this problem, the described method uses a clustering algorithm reinforced by a small set of rules. Experiments show that it achieves an F-measure of 0.879 and outperforms other methods. They also show that it only requires the target text and an English word list of reasonable size.