The doubly constrained robust Capon beamformer (DCRCB), which employs a spherical uncertainty set of the steering vector together with the constant norm constraint, can provide robustness against arbitrary array imperfections. However, its performance can be greatly degraded when the uncertainty bound of the spherical set is not properly selected. In this paper, combining the DCRCB and the weight-vector-norm-constrained beamformer (WVNCB), we suggest a new robust adaptive beamforming method which allows us to overcome the performance degradation due to improper selection of the uncertainty bound. In WVNCB, its weight vector norm is limited not to be larger than a threshold. Both WVNCB and DCRCB belong to a class of diagonal loading methods. The diagonal loading range of WVNCB, which dose not consider negative loading, is extended to match that of DCRCB which can have a negative loading level as well as a positive one. In contrast to the conventional DCRCB with a fixed uncertainty bound, the bound in the proposed method varies such that the weight vector norm constraint is satisfied. Simulation results show that the proposed beamformer outperforms both DCRCB and WVNCB, being far less sensitive to the uncertainty bound than DCRCB.

In this paper, we present a construction method of non-binary low-density parity-check (LDPC) convolutional codes. Our construction method is an extension of Felstrom and Zigangirov construction [1] for non-binary LDPC convolutional codes. The rate-compatibility of the non-binary convolutional code is also discussed. The proposed rate-compatible code is designed from one single mother (2,4)-regular non-binary LDPC convolutional code of rate 1/2. Higher-rate codes are produced by puncturing the mother code and lower-rate codes are produced by multiplicatively repeating the mother code. Simulation results show that non-binary LDPC convolutional codes of rate 1/2 outperform state-of-the-art binary LDPC convolutional codes with comparable constraint bit length. Also the derived low-rate and high-rate non-binary LDPC convolutional codes exhibit good decoding performance without loss of large gap to the Shannon limits.

This paper presents a novel method for reconstructing 3D geometry of camera motion and human-face model from a video sequence. The approach combines the concepts of Powell's line minimization with gradient descent. We adapted the line minimization with bracketing used in Powell's minimization to our method. However, instead of bracketing and searching deep down a direction for the minimum point along that direction as done in their line minimization, we achieve minimization by bracketing and searching for the direction in the bracket which provides a lower energy than the previous iteration. Thus, we do not need a large memory as required by Powell's algorithm. The approach to moving in a better direction is similar to classical gradient descent except that the derivative calculation and a good starting point are not needed. The system's constraints are also used to control the bracketing direction. The reconstructed solution is further improved using the Levenberg Marquardt algorithm. No average face model or known-coordinate markers are needed. Feature points defining the human face are tracked using the active appearance model. Occluded points, even in the case of self occlusion, do not pose a problem. The reconstructed space is normalized where the origin can be arbitrarily placed. To use the obtained reconstruction, one can rescale the computed volume to a known scale and transform the coordinate system to any other desired coordinates. This is relatively easy since the 3D geometry of the facial points and the camera parameters of all frames are explicitly computed. Robustness to noise and lens distortion, and 3D accuracy are also demonstrated. All experiments were conducted with an off-the-shelf digital camera carried by a person walking without using any dolly to demonstrate the robustness and practicality of the method. Our method does not require a large memory or the use of any particular, expensive equipment.

Pedestrian detection from visual images, which is used for driver assistance or video surveillance, is a recent challenging problem. Co-occurrence histograms of oriented gradients (CoHOG) is a powerful feature descriptor for pedestrian detection and achieves the highest detection accuracy. However, its calculation cost is too large to calculate it in real-time on state-of-the-art processors. In this paper, to obtain optimal parallel implementation for an NVIDIA GPU, several kinds of parallelism of CoHOG-based detection are shown and evaluated suitability for implementation. The experimental result shows that the detection process can be performed at 16.5 fps in QVGA images on NVIDIA Tesla C1060 by optimized parallel implementation. By our evaluation, it is shown that the optimal strategy of parallel implementation for an NVIDIA GPU is different from that of FPGA. We discuss about the reason and show the advantages of each device. To show the scalability and portability of GPU implementation, the same object code is executed on other NVIDA GPUs. The experimental result shows that GTX570 can perform the CoHOG-based pedestiran detection 21.3 fps in QVGA images.

In multiuser multiple-input multiple-output (MU-MIMO) wireless downlink systems, block diagonalization (BD) is a technique, where the transmit precoding matrix of each user is designed such that its subspace lies in the null space of all the other remaining users, so that multiuser interference (MUI) is completely canceled. In low signal to noise power ratio (SNR) or low signal to interference plus noise power ratio (SINR) environments, regularized BD, that lets some MUI remain and maximizes the sum rate capacity of the BD MIMO channel, was also proposed. One of the problems of both the approaches is high complexity of computation due to a lot of singular value decomposition (SVD) processes. In this paper we propose new BD techniques utilizing QR decomposition (QRD) which can be practically achieved by Gram-Schmidt orthogonalization (GSO) with lower complexity compared to the conventional method employing SVD. We can show that the performance of the proposed approaches is close to the conventional approaches, while the proposed approaches have much lower complexity.

Recently, Haley and Grant introduced the concept of reversible codes – a class of binary linear codes that can reuse the decoder architecture as the encoder and encodable by the iterative message-passing algorithm based on the Jacobi method over F2. They also developed a procedure to construct parity check matrices of a class of reversible codes named type-I reversible codes by utilizing properties specific to circulant matrices. In this paper, we refine a mathematical framework for reversible codes based on circulant matrices through a ring theoretic approach. This approach enables us to clarify the necessary and sufficient condition on which type-I reversible codes exist. Moreover, a systematic procedure to construct all circulant matrices that constitute parity check matrices of type-I reversible codes is also presented.

A novel algorithm is presented for estimating the 3-D location (azimuth angle, elevation angle, and range) of multiple sources with a uniform circular array (UCA). Based on its centrosymmetric property, a UCA is divided into two subarrays. The steering vectors for these subarrays then yield a 2-D direction of arrival (DOA)-related rotational invariance property in the signal subspace, which enables 2-D DOA estimations using a generalized-ESPRIT algorithm. Based on the estimated 2-D DOAs, a range estimation can then be obtained for each source by defining the 1-D MUSIC spectrum. Despite its low computational complexity, the proposed algorithm can almost match the performance of the benchmark estimator 3-D MUSIC.

This study formulates and solves the wire planning problem with electro-migration and interference using an effective integer linear programming (ILP)-based approach. For circuits without obstacles, the proposed approach obtains a wire planning with the minimum wiring area. An effective approach for estimating the length of feasible routing wire is proposed to handle circuits with obstacles. In addition, the space reservation technique, which allocates the ring of the free silicon space around obstacles, is presented to improve interference among routing wires and on-obstacle wires. For circuits with obstacles, the proposed method minimizes total wiring area and reduces interference. Experimental results show that the integer linear-programming-based approach effectively and efficiently minimizes wiring area of routing wires.

Kudekar et al. recently proved that for transmission over the binary erasure channel (BEC), spatial coupling of LDPC codes increases the BP threshold of the coupled ensemble to the MAP threshold of the underlying LDPC codes. One major drawback of the capacity-achieving spatially-coupled LDPC codes is that one needs to increase the column and row weight of parity-check matrices of the underlying LDPC codes. It is proved, that Hsu-Anastasopoulos (HA) codes and MacKay-Neal (MN) codes achieve the capacity of memoryless binary-input symmetric-output channels under MAP decoding with bounded column and row weight of the parity-check matrices. The HA codes and the MN codes are dual codes each other. The aim of this paper is to present an empirical evidence that spatially-coupled MN (resp. HA) codes with bounded column and row weight achieve the capacity of the BEC. To this end, we introduce a spatial coupling scheme of MN (resp. HA) codes. By density evolution analysis, we will show that the resulting spatially-coupled MN (resp. HA) codes have the BP threshold close to the Shannon limit.

In 2010, Guo and Zhang proposed a group key agreement protocol based on the chaotic hash function. This letter points out that Guo-Zhang's protocol is still vulnerable to off-line password guessing attacks, stolen-verifier attacks and reflection attacks.

Power analysis can be used to attack many implementations of cryptosystems, e.g., RSA and ECC, and the doubling attack is a collision based power analysis performed on two chosen ciphertexts. In this paper, we introduced a modified doubling attack to threaten RSA and ECC implementations by exploiting only one chosen ciphertext of small order. To attack the RSA implementations we selected an input of order two while to attack the ECC implementations we exploited one chosen invalid point of small order on a cryptographically weak curve rather than on the original curve. We showed that several existing power analysis countermeasures for RSA and ECC implementations are still vulnerable to the proposed attack. To prevent the proposed attack, we suggested countermeasures for RSA as well as for ECC.

We propose a committee-based method of active learning for large vocabulary continuous speech recognition. Multiple recognizers are trained in this approach, and the recognition results obtained from these are used for selecting utterances. Those utterances whose recognition results differ the most among recognizers are selected and transcribed. Progressive alignment and voting entropy are used to measure the degree of disagreement among recognizers on the recognition result. Our method was evaluated by using 191-hour speech data in the Corpus of Spontaneous Japanese. It proved to be significantly better than random selection. It only required 63 h of data to achieve a word accuracy of 74%, while standard training (i.e., random selection) required 103 h of data. It also proved to be significantly better than conventional uncertainty sampling using word posterior probabilities.

Recently, Li and Xia proposed a physical-layer security design to guarantee a low probability of interception (LPI) for asynchronous cooperative systems without relying on upper-layer data encryption. The proposed scheme utilizes diagonal unitary codes to perform different encoding in the frequency domain over subcarriers within each OFDM block to randomize the transmitted signals. To build on their idea, in this letter, a subcarrier-reference (SR) transmission scheme is proposed with deliberate signal randomization to achieve LPI in multiuser MISO-OFDMA systems. For each user, one of the allocated subcarriers is chosen by the transmitter to send reference signals, and others are chosen to send the user's information symbols. By some deliberate signal randomization, the eavesdropper cannot detect the transmitted symbols, while the authorized users can operate the system successfully without knowledge of the channels by subcarrier-reference demodulation. Extensive simulations are conducted to demonstrate the scheme's effectiveness.

We propose frequency-domain optical code-division-multiplexing (CDM) employing quadrature-amplitude-modulation (QAM) using two of multi-level (M-ary) data generated based on electrical-domain spatial code spreading. Its spectral efficiency is enhanced compared to the conventional scheme with amplitude-shift-keying (ASK) using only one of M-ary data. Although it demands the recovery of amplitude and optical phase information, the practicality of the receiver is retained with self-homodyne detection using a phase-shift-keying (PSK) pilot light. Performance is theoretically evaluated and the optimal parameters are derived. Finally, the feasibility of the proposed technique is experimentally confirmed.

In this paper, we propose a novel feeding structure for a coaxial-excited compact waveguide filter, which is composed of planar resonators called frequency-selective surfaces (FSSs). In our proposed feeding structure, new FSSs located at the input and output ports are directly excited by the coaxial line. By using the FSSs, the transition from the TEM mode to the TE10 mode is realized by the resonance of the FSSs. Therefore, the backshort length from the coaxial probe to the shorted waveguide end can be made much shorter than one-quarter of the guided wavelength. Additionally, the coaxial-excited FSS provides one transmission zero at each stopband. As a design example, a three-stage bandpass filter with 4% bandwidth at the X band is demonstrated. The designed filter has a very compact size of one cavity and has high skirt selectivity with six transmission zeros. The effectiveness of the design is confirmed by the comparison of frequency characteristics obtained by the simulation and measurement.

We propose two types of active directional couplers to assure high TX cancellation: an asymmetric type and a symmetric type. For attaining low receiving through loss, coupling capacitors used in conventional couplers are replaced by amplifiers in the proposed active directional couplers. The asymmetric active directional coupler is composed of a small number of components and simple structure. The symmetric active directional coupler has wide-bandwidth TX cancellation. Measurement results show that receiving through loss of -5.3 dB and the TX cancellation of -67.6 dB are obtained in the asymmetric active directional coupler, and receiving through loss of -6.7 dB and the TX cancellation of -66.4 dB are obtained in the symmetric active directional coupler. Compared to the asymmetric active directional coupler, the symmetric active directional coupler has advantage of wider bandwidth of 1.25 MHz to reduce TX leakage of less than -55 dB. Both the proposed active directional couplers achieve high TX cancellation, and the symmetric active directional coupler can be applied in a UHF RFID system with 10-m communication range.

On the huge-scale array antenna for SSPS (space solar power systems), the problem of faulty elements and effect of mutual coupling between array elements should be considered in practice. In this paper, the effect of faulty elements as well as mutual coupling on the performance of the huge-scale array antenna are analyzed by using the proposed IEM/LAC. The result shows that effect of faulty elements and mutual coupling on the actual gain of the huge-scale array antenna are significant.

Microwave/millimeter-wave phase and amplitude characteristics of the optically controlled phased array antenna with a different SMF for each antenna feed were measured. Suitable phases for the beam steering can be realized by the adjustment of the LD wavelength independently with multiple SMFs. In addition to the phase, amplitude of each antenna feed can be controlled stably using LD current without phase variation. Furthermore, effectiveness of the calibration method of the phased array using multiple SMFs by LD wavelength adjustment is experimentally verified. Excellent microwave/millimeter-wave phase characteristics using 2- and 3-element optically controlled phased array feed were experimentally demonstrated with calibration of the phases. Phase characteristics of the array using multiple SMFs were also compared with that using a single SMF experimentally.

Due to an increase in multimedia content and the acceleration of digital convergence, demand for next-generation IPTV service is rapidly growing. IPTV service seamlessly provides both real-time broadcasting and content sharing services on diverse terminals through complex networks. In this paper, a secure and scalable content sharing framework is proposed for next-generation IPTV service. The proposed framework has an advantage over conventional content protection techniques in producing scalable content with transcodable, adjustable, and perceptual security features. Moreover, it ensures end-to-end security over the entire service range based on a single security mechanism. The suitability of the proposed approach is demonstrated experimentally using a practical service scenario with real-world environments. The experiments show that the proposed approach can provide several different levels of content security, from a perceptual level to an almost unintelligible level, while keeping the additional time overhead low. Consequently, it is expected that use of this security technology alone can have a practical contribution in creating new business opportunities for IPTV services.

In this letter, we propose an effective method to retrieve images from a 2D shape image database using discrete shock graphs combined with an adaptive selection algorithm. Experimental results show that our method is more accurate and fast than conventional approaches and reduces computational complexity.