A novel bimodal method for face recognition under low-level lighting conditions is proposed. It fuses an enhanced gray level image and an illumination-invariant geometric image at the feature-level. To further improve the recognition performance under large variations in attributions such as poses and expressions, discriminant features are extracted from source images using the wavelet transform-based method. Features are adaptively fused to reconstruct the final face sample. Then FLD is used to generate a supervised discriminant space for the classification task. Experiments show that the bimodal method outperforms conventional methods under complex conditions.

This letter deals with an iterative decoding algorithm (IDA) for product codes. In the IDA, a soft-input and output iterative bounded-distance and encoding-based decoding algorithm is used for the component codes. Simulation results over an AWGN channel with BPSK modulation is presented and show the effectiveness of the IDA.

A new acceleration scheme that decreases the number of required iterations in relaxation methodology is proposed. The proposed scheme uses dynamic error prediction of an improved approximation to the solution during an iterative computation. The proposed scheme's application to circuit simulations required an average of 67.3% fewer iterations compared to un-accelerated relaxation methods.

In low bit-rate video transmission, the payload of a single packet can often contain a whole coded frame due to the high compression ratio in both spatial and temporal domains. Thus, the loss of a single packet can lead to the loss of a whole video frame. In this paper, we propose a novel error concealment algorithm that can effectively reconstruct the lost frame and protect the quality of video streams from the degradation caused by propagation errors. The proposed algorithm employs a bilateral motion estimation scheme where the weighted sum of the received motion vectors (MVs) in the neighboring frames is utilized to construct the MV field for the concealed frame. Unlike the conventional algorithms, the proposed scheme does not produce any overlapped pixel and hole region in the reconstructed frame. The proposed algorithm can be applied not only to the case of single frame loss but also adaptively extended to the case of multiframe loss. Experimental results show that the proposed algorithm outperforms other conventional techniques in terms of both peak signal-to-noise ratio (PSNR) performance and subjective visual quality.

The heterodyne laser interferometer acts as an ultra-precise measurement apparatus in semiconductor manufacture. However the periodical nonlinearity property caused from frequency cross-talk is an obstacle to improve the high measurement accuracy in nanometer scale. In order to minimize the nonlinearity error of the heterodyne interferometer, we propose a frequency cross-talk compensation algorithm using an artificial intelligence method. The feedforward neural network trained by back-propagation compensates the nonlinearity error and regulates to minimize the difference with the reference signal. With some experimental results, the improved accuracy is proved through comparison with the position value from a capacitive displacement sensor.

We present a very simple probabilistic, passive attack against the protocols HB and HB+. Our attack presents some interesting features: it requires less captured transcripts of protocol executions when compared to previous results; It makes possible to trade the amount of required transcripts for computational complexity; the value of noise used in the protocols HB and HB+ need not be known.

A self-stabilizing protocol is a protocol that achieves its intended behavior regardless of the initial configuration (i.e., global state). Thus, a self-stabilizing protocol is adaptive to any number and any type of topology changes of networks: after the last topology change occurs, the protocol starts to converge to its intended behavior. This advantage makes self-stabilizing protocols extremely attractive for designing highly dependable distributed systems on dynamic networks. While conventional self-stabilizing protocols require that the networks remain static during convergence to the intended behaviors, some recent works undertook the challenge of realizing self-stabilization in dynamic networks with frequent topology changes. This paper introduces some of the challenges as a new direction of research in self-stabilization.

This paper focuses on optical integration technology and its application in optical microsensors used in biomedical fields. The integration is based on the hybrid integration approach, achieving high performance, small size and weight, and lower cost. First, we describe the key technologies used in hybrid integration, namely passive alignment technology, low temperature bonding technology, and packaging technology for realizing advanced microsensors. Then, we describe an integrated laser Doppler flowmeter that can monitor blood flow in human skin.

We present a novel model, named Category Constraint-Latent Dirichlet Allocation (CC-LDA), to learn and recognize natural scene category. Previous work had to resort to additional classifier after obtaining image topic representation. Our model puts the category information in topic inference, so every category is represented in a different topics simplex and topic size, which is consistent with human cognitive habit. The significant feature in our model is that it can do discrimination without combined additional classifier, during the same time of getting topic representation. We investigate the classification performance with variable scene category tasks. The experiments have demonstrated that our learning model can get better performance with less training data.

A Feedback Error Learning (FEL) scheme was found to be applicable to joint angle control by Functional Electrical Stimulation (FES) in our previous study. However, the FEL-FES controller had a problem in learning of the inverse dynamics model (IDM) in some cases. In this paper, methods of applying the FEL to FES control were examined in controlling 1-DOF movement of the wrist joint stimulating 2 muscles through computer simulation under several control conditions with several subject models. The problems in applying FEL to FES controller were suggested to be in restricting stimulation intensity to positive values between the minimum and the maximum intensities and in the case of very small output values of the IDM. Learning of the IDM was greatly improved by considering the IDM output range with setting the minimum ANN output value in calculating ANN connection weight change.

This letter proposes an adaptive threshold control algorithm for a low complex noncoherent IR-UWB receiver using a 1-bit ADC. To estimate and control a threshold level in the noncoherent IR-UWB receiver, it uses binary output of the 1-bit ADC instead of energy level of received signals using a high resolution ADC, which reduces hardware complexity of the receiver. Theoretical performance evaluation and computer simulation demonstrates that the performance of the proposed algorithm is similar to that of theoretically optimum one.

In order to reduce the peak-to-average power ratio (PAPR) of orthogonal frequency division multiplexing (OFDM) signals over fading channels, the selected mapping (SLM) scheme and channel estimation technique can be used. Because of the side information of SLM, however, the data rate decreases in the conventional transmission method. In this letter, to overcome this impairment, we propose a technique which transmits side information by using pilot tones. Numerical analysis and computer simulations show that this method can maintain the PAPR reduction ability of SLM while improving error performance.

We consider the problem of secret key agreement in Gaussian Maurer's Model. In Gaussian Maurer's model, legitimate receivers, Alice and Bob, and a wire-tapper, Eve, receive signals randomly generated by a satellite through three independent memoryless Gaussian channels respectively. Then Alice and Bob generate a common secret key from their received signals. In this model, we propose a protocol for generating a common secret key by using the result of soft-decision of Alice and Bob's received signals. Then, we calculate a lower bound on the secret key rate in our proposed protocol. As a result of comparison with the protocol that only uses hard-decision, we found that the higher rate is obtained by using our protocol.

In this paper, the impacts of using multiple transmit antennas under doubly correlated MIMO channels on CDMA uplink code acquisition is studied. The performance of a MIMO code acquisition system is analyzed by considering spatial fading correlations, which depend on antenna spacing and azimuth spread at both MS and BS. The detection performance and mean acquisition time in the presence of spatially correlated MIMO channel are presented on a frequency selective fading channel and compared with the cases of spatial fading decorrelation via numerical evaluation. It is observed that the acquisition performance relies on the degree of spatial fading correlations. In addition, it is surprisingly seen that a MIMO code acquisition system provides worse performance than SIMO.

Software model checking is typically applied to components of large systems. The assumption generation is the problem of finding the least restrictive environment in which the components satisfy a given safety property. There is an algorithm to compute the environment for properties given as a regular language. In this paper, we propose a general scheme for computing the assumption even for non-regular properties, and show the uniqueness of the least restrictive assumption for any class of languages. In general, dealing with non-regular languages may fall into undecidability of problems. We also show a method to compute assumptions based on visibly pushdown automata and their finite-state abstractions.

In this letter, we present an exact analytic expression for the maximum signal-to-interference ratio (SIR) for receivers communicating with multiple transmitting nodes over a general time-varying channel, where one of the nodes is chosen as a desired signal source based on the instantaneous channel condition and the other nodes act as interference sources. As an illustrative example, the maximum SIR distribution of a mobile receiver surrounded by three base stations (BS) is determined in a closed-form formula for Rayleigh fading channels, and its accuracy is confirmed using simulation results.

In multiple antenna systems that use spatial multiplexing to raise transmission rates, it is preferable to use maximum likelihood (ML) detection to exploit the full receive diversity and minimize the error probability. In this paper, we present two tree based approximate ML detectors that use new two ordering criteria in conjunction with efficient search strategies. Unlike conventional tree detectors, the new detectors closely approximate the error performance of the exact ML detector while achieving a dramatic reduction in complexity. Moreover, they ensure a fixed detection delay and high level of parallelization in the tree search.

Role-based access control (RBAC) is widely used as an access control mechanism in various computer systems. Since an organization's lines of authority influence the authorized privileges of jobs, roles also form a hierarchical structure. A role graph is a model that represents role hierarchies and is suitable for the runtime phase of RBAC deployment. Since a role graph cannot take various forms for given roles and cannot handle abstraction of roles well, however, it is not suitable for the design phase of RBAC deployment. Hence, an extended role graph, which can take a more flexible form than that of a role graph, is proposed. The extended role graph improves diversity and clarifies abstraction of roles, making it suitable for the design phase. An equivalent transformation algorithm (ETA), for transforming an extended role graph into an equivalent role graph, is also proposed. Using the ETA, system administrators can deploy efficiently RBAC by using an extended role graph in the design phase and a standard role graph in the runtime phase.

A space-efficient approximation algorithm for the grammar-based compression problem, which requests for a given string to find a smallest context-free grammar deriving the string, is presented. For the input length n and an optimum CFG size g, the algorithm consumes only O(g log g) space and O(n log*n) time to achieve O((log*n)log n) approximation ratio to the optimum compression, where log*n is the maximum number of logarithms satisfying log log log n > 1. This ratio is thus regarded to almost O(log n), which is the currently best approximation ratio. While g depends on the string, it is known that g=Ω(log n) and for strings from k-letter alphabet [12].

The spanning tree problem is to find a tree that connects all the vertices of G. This problem has many applications, such as electric power systems, computer network design and circuit analysis. Klein and Stein demonstrated that a spanning tree can be found in O(log n) time with O(n+m) processors on the CRCW PRAM. In general, it is known that more efficient parallel algorithms can be developed by restricting classes of graphs. Circular permutation graphs properly contain the set of permutation graphs as a subclass and are first introduced by Rotem and Urrutia. They provided O(n2.376) time recognition algorithm. Circular permutation graphs and their models find several applications in VLSI layout. In this paper, we propose an optimal parallel algorithm for constructing a spanning tree on circular permutation graphs. It runs in O(log n) time with O(n/log n) processors on the EREW PRAM.