We propose a new method to introduce the Minimum Description Length (MDL) criterion to the automatic generation of non-uniform, context-dependent HMM topologies. Phonetic decision tree clustering is widely used, based on the Maximum Likelihood (ML) criterion, and only creates contextual variations. However, the ML criterion needs to predetermine control parameters, such as the total number of states, empirically for use as stop criteria. Information criteria have been applied to solve this problem for decision tree clustering. However, decision tree clustering cannot create topologies with various state lengths automatically. Therefore, we propose a method that applies the MDL criterion as split and stop criteria to the Successive State Splitting (SSS) algorithm as a means of generating contextual and temporal variations. This proposed method, the MDL-SSS algorithm, can automatically create adequate topologies without such predetermined parameters. Experimental results for travel arrangement dialogs and lecture speech show that the MDL-SSS can automatically stop splitting and obtain more appropriate HMM topologies than the original one.

A single solid-state magnetic sensor can be used to measure a current by sensing the field near the conductor in a non-contact way. In order to improve the accuracy of the measuring system, magnetic sensor arrays have been introduced in the current measurement around the conductor. An analytical algorithm based on Discrete Fourier Transform (DFT) is presented in this paper, which can separate the effects of the field generated by the current under measurement from the interference fields. A general mathematical model of the interference analysis is set up, which can be used for both DC and AC current measurement and has no restriction on the shape and number of the current conductors. Numerical simulations associated with preliminary experimental results confirm the validity of the approach.

Multi-Input Multi-Output (MIMO) systems, which utilize multiple antennas at both the receiver and transmitter, promise very high data rates in a rich scattering environment. It was proven in literature that with optimal power allocation, MIMO eigenmode transmission system (EMTS) is optimal because MIMO capacity is maximized. However, the performance of MIMO EMTS is very sensitive to the accuracy of channel state information and thus it is of practical importance to analyze its performance when channel state information is corrupted under realistic system and propagation conditions. In this paper, we lower bound the mutual information of MIMO EMTS with imperfect channel estimation and delayed quantized feedback in a spatially correlated continuous fading channel. Our results showed that this lower bound is tight and can serve as a comprehensive guide to the actual performance of MIMO EMTS under practical operating conditions.

Inside a connector an interface with low insertion force and contact resistance is required, utilizing low cost materials such as copper alloys surrounded by tin coating. Relating to the application, the operating parameters have a wide range of values of currents, forces and materials. In this paper, we present a new experimental method based on non-intrusive probing of the deflection of the spring terminal with a laser technique. The main feature is that the reflection of the Laser beam onto the spring allows the determination of the contact force of the lamella-spring inside the female part. The technique requires the following insertion parameters during the insertion stroke: contact deflection δ, which allows contact force Fc, insertion force Fi and contact resistance Rc. It was found that the insertion force has a maximum value which decreases to the stable value, and depends on the size and the material of the pin. However contact resistance decreases sharply when first inserting, and tends to stable values on completing the insertion process, which is less sensitive to the pin diameter. Furthermore the final value which is important for the connector characterization is related and discussed. Finally, discrepancies were observed between the experimental and calculated data with simple numerical models. More complex models are in progress, which should improve the convergence of the theoretical approach to experimental results and proceed to the optimization of the connector parameters.

A new indirect approach for designing low-order linear-phase IIR filters is presented in this paper. Given an FIR filter, we utilize a new Krylov subspace projection method, called the rational Arnoldi method with adaptive orders, to synthesize an approximated IIR filter with small orders. The synthesized IIR filter can truly reflect essential dynamical features of the original FIR filter and indeed satisfies the design specifications. Also, from simulation results, it can be observed that the linear-phase property in the passband is stilled retained. This indirect approach is accomplished using the state-space realization of FIR filters, multi-point Pade approximations, the Arnoldi algorithm, and an intelligent scheme to select expansion points in the frequency domain. Such methods are quite efficient in terms of computational complexity. Fundamental developments of the proposed method will be discussed in details. Numerical results will demonstrate the accuracy and the efficiency of this two-step indirect method.

A new algorithm, termed reduced-order Root-MUSIC, for high resolution direction finding is proposed. It requires finding all the roots of a polynomial with an order equaling twice the number of propagating signals. Some Monte Carlo simulations are used to test the effectiveness of the proposed algorithm.

It is novel to apply magnetic field in the indoor air purification by collecting pre-charged suspended particles. Based on experiments and analysis of relative data, the effects of some influential factors (the number of discharge electrode, the polarity and magnitude of discharge voltage, the direction and magnitude of magnetic field, the initial velocity of charged particles, the distance between particle collecting plates) on the efficiency of air purification are discussed. The results indicate that the purification efficiency is improved by applying the proper magnetic field, -6 kV direct current voltage is an optimal voltage and there are optimal magnitudes of the distance between collecting plate and the initial velocity of particles in the purification process.

Design points and the results seen in the development of a dynamically reconfigurable logic LSI, PCA-2, are described. PCA-2 enables the realization of flexible parallel processing based on the autonomous reconfiguration of logic circuits. To realize this feature, we introduce an asynchronous circuit design and a homogeneous cell array structure. PCA-2 represents an advance on the earlier LSI, PCA-1. Cutting edge CMOS technology is used to realize the structural merits of PCA hardware. Compared to PCA-1, PCA-2 offers 16 times greater integration level for programmable logic. Due to miniaturization and design refinement, PCA-2 provides a 6-fold increase in the circuit frequency of the configuration controller and a 3-fold increase in the operating frequency of the programmable logic. The results gained confirm the effects of refinement and the suitability of our architecture for device miniaturization.

In this paper, we propose a Time Interleave (TI)-Ultra Wideband (UWB)-Impulse Radio (IR) system where Nf monocycle pulses per information symbol are transmitted not continuously but discontinuously through the time interleaver to get more time diversity. Moreover, we compare the diversity gains of the UWB-IR systems on block fading channels with the fading block size q: the UWB-IR system with a single transmit antenna, the Space Time (ST)-UWB-IR system with Nt transmit antennas, and the TI-UWB-IR system with a single transmit antenna. Simulation results show that when the fading block size q = 1, the diversity gains of all the systems are same. When q is larger than or equal to 2 (q 2) and when both the number of transmit antennas Nt of the ST-UWB-IR system and the time interleaver size NTI of the TI-UWB-IR system are larger than or equal to q (Nt, NTI q), we show the following; both the ST-UWB-IR system and the TI-UWB-IR system achieve the same diversity gain that is constant regardless of Nt and NTI. On the other hand, when q 2 and when Nt and NTI < q, we show the following; when Nt = NTI, the ST-UWB-IR system and the TI-UWB-IR system get the same diversity gain; when Nt > NTI, the diversity gain of the ST-UWB-IR system is larger than that of the TI-UWB-IR system; and vice when Nt > NTI.

To provide multimedia services, a server system can use a cluster of servers to serve many users concurrently. In the literature, several assignment algorithms have been proposed to assign the media objects to these servers. The popularities of the media objects may change with time, and the service provider may add new objects or delete obsolete objects. In these cases, it is necessary to execute an assignment algorithm to find a new assignment. To realize the new assignment from the current one, it is necessary to relocate some objects. In this paper, we formulate and solve this relocation problem, and our objective is to minimize the relocation time for minimal service interruption. We propose a two-stage algorithm to solve this relocation problem. In the first stage, we introduce the concept of server relabelling, and apply the Hungarian method to find the largest common assignment between the new and current assignments. In the second stage, we propose a chain relocation schedule by which every object is directly relocated from its original server to the new server without deadlock. We use computer simulation to demonstrate the effectiveness of the optimal relocation method.

We attempt to achieve corporative behavior of autonomous decentralized agents constructed via Q-Learning, which is a type of reinforcement learning. As such, in the present paper, we examine the piano mover's problem including a find-path problem. We propose a multi-agent architecture that has an external agent and internal agents. Internal agents are homogenous and can communicate with each other. The movement of the external agent depends on the composition of the actions of the internal agents. By learning how to move through the internal agents, avoidance of obstacles by the object is expected. We simulate the proposed method in a two-dimensional continuous world. Results obtained in the present investigation reveal the effectiveness of the proposed method.

In this paper, we will propose a robust watermarking system for digital audio sound to protect the copyright of publication and claim of ownership. The proposed watermarking scheme uses the frequency extent between 1 Hz and 20 Hz, which cannot be heard by the unaided human ear, to embed the watermark. Thus, the original audio quality will not be influenced by the watermark. Currently, the techniques of Perceptual Audio Coder contain MPEG-1, -2, -2.5, MPEG-2 AAC, MPEG-4 AAC and Window Media Audio. From experimental results, the proposed watermarking system can resist attacks of previous audio coders and low bit-rate compression. The watermark is extracted with 100% correction after previous encoder attacks. Furthermore, to authenticate the audio signal, the system can quickly extract the watermark without the knowledge of original audio signals.

In this letter, adaptive multi-stage parallel interference cancellation (PIC) receiver is considered for multi-rate DS-CDMA system. In each stage of the adaptive multi-stage PIC receiver, multiple access interference (MAI) estimates are obtained by the sub-bit estimates from the previous stage and the adaptive weights for the sub-bit estimates. The adaptive weights are obtained by minimizing the mean squared error between the received signal and its estimate through normalized least mean square (LMS) algorithm. It is shown that the adaptive multi-stage PIC receiver achieves smaller BER than the matched filter receiver, multi-stage PIC receiver, and multi-stage partial PIC receiver for the multi-rate DS-CDMA system in a Rayleigh fading channel.

The enlargement of the left lobe of the liver and the shrinkage of the right lobe are helpful signs at MR imaging in diagnosis of cirrhosis of the liver. To investigate whether the volume ratio of left-to-whole (LTW) is effective to differentiate cirrhosis from a normal liver, we developed an automatic algorithm for three-dimensional (3D) segmentation and volume calculation of the liver region in multi-detector row CT scans and MR imaging. From one manually selected slice that contains a large liver area, two edge operators are applied to obtain the initial liver area, from which the mean gray value is calculated as threshold value in order to eliminate the connected organs or tissues. The final contour is re-confirmed by using thresholding technique. The liver region in the next slice is generated by referring to the result from the last slice. After continuous procedure of this segmentation on each slice, the 3D liver is reconstructed from all the extracted slices and the surface image can be displayed from different view points by using the volume rendering technique. The liver is then separated into the left and the right lobe by drawing an inter-segmental plane manually, and the volume in each part is calculated slice by slice. The degree of cirrhosis can be defined as the ratio of volume in these two lobes. Four cases including normal and cirrhotic liver with MR and CT slices are used for 3D segmentation and visualization. The volume ratio of LTW was relatively higher in cirrhosis than in the normal cases in both MR and CT cases. The average error rate on liver segmentation was within 5.6% after employing in 30 MR cases. These results demonstrate that the performance in our 3D segmentation was satisfied and the LTW ratio may be effective to differentiate cirrhosis.

A large body of research in the measurement of software complexity at code level has been conducted, but little effort has been made to measure the architectural-level complexity of a software system. In this paper, we propose some architectural-level metrics which are appropriate for evaluating the architectural attributes of a software system. The main feature of our approach is to assess the architectural-level complexity of a software system by analyzing its formal architectural specification, and therefore the process of metric computation can be automated completely.

This paper analyzes the timing of eyeblink during visual identification of katakana characters on a display, which were presented under the constraint of a restricted visual field (R.V.F.). Blinks frequently occurred when the subject slowly brought the R.V.F. near a feature point (e.g., terminal point, crossing point).

A simple millimeter-wave quasi-maximal-ratio-combin-ing antenna diversity system based on the millimeter-wave self-heterodyne transmission technique is described. The millimeter-wave self-heterodyne transmission technique is useful for developing millimeter-wave systems with enhanced characteristics in regard to system miniaturization, development and fabrication cost, and the frequency stability of the signal transmission. We also show that applying this technique with an antenna diversity receiver configuration can easily solve a problem peculiar to millimeter-wave systems--the fact that the transmission link always requires a line-of-sight path--without requiring hardware designed with millimeter-scale precision. In this paper, we theoretically analyze the operating principle of a combining antenna diversity system based on the millimeter-wave self-heterodyne transmission technique. We further prove that we can obtain a diversity gain in accordance with that of a maximal-ratio combining diversity system without resorting to any complicated control of the received signal envelope and phase. Our experiments using the simplest two-branch diversity structure have validated the operating principle derived in our theoretical analysis. Our results show that a received CNR improvement of 3 dB is obtained as a diversity gain. We also demonstrate that circuit precision corresponding to the wavelength of the intermediate frequency, rather than to the millimeter wavelength, is sufficient to obtain the diversity effect when we control the signal phase or delay in combining the received signals.

Computer simulation of cellular process is one of the most important applications in bioinformatics. Since such simulators need huge computational resources, many biologists must use expensive PC/WS clusters. ReCSiP is an FPGA-based, reconfigurable accelerator which aims to realize economical high-performance simulation environment on desktop computers. It can exploit fine-grain parallelism in the target applications by small hardware modules in the FPGA which work in parallel manner. As the first step to implement a simulator of cellular process on ReCSiP, a solver to perform a basic simulation of metabolism was implemented. The throughput of the solver was about 29 times faster than the software on Intel's PentiumIII operating at 1.13 GHz.

In a high-rate indoor wireless personal communication system, the delay spread due to multi-path propagation results in intersymbol interference which can significantly increase the transmission bit error rate (BER). The technique most commonly used for combating the intersymbol interference and frequency-selective fading found in communications channels is the adaptive equalization. In this paper, we propose a novel neural detector based on self-organizing map (SOM) to improve the system performance of the receiver. In the proposed scheme, the SOM is used as an adaptive detector of equalizer, which updates the decision levels to follow the received faded signal. To adapt the proposed scheme to the time-varying channel, we use the Euclidean distance, which will be updated automatically according to the received faded signal, as an adaptive radius to define the neighborhood of the winning neuron of the SOM algorithm. Simulations on a 16 QAM system show that the receiver using the proposed neural detector has a significantly better BER performance than the traditional receiver.

This letter describes consecutive zero and one bits detection circuits designed for a 1.25 Gbit/s burst mode laser driver realized in a SiGe 0.35 µm BiCMOS technology with 3.3 V power supply. The architecture is based on a frequency divider and a delay line counting per four consecutive zero or one bits. The detector was designed with high-speed split-output stage flip-flops modified to have a reset input. Experimental results validate the design of the detector.