This paper investigates the problem of constructing a logical multicasting tree which dispatches data to multiple destinations according to their bandwidth requirements. An optimization problem is formulated to minimize the maximum delay between a sender and multiple receivers. An algorithm of finding the optimum branching locations is presented. Performance analysis from the closed queueing network theory is given to evaluate a multicasting tree network based on this proposed algorithm.

The design and engineering of new network intelligence platforms to accommodate the ever-changing and growing demands of customers, presents rich market opportunities and challenges tempered by concerns arising from the problematic experiences of similar system and network developments. As the telecommunications industry evolves, customers are increasingly coming to expect the perception of instantaneous access to service providers together with transparency to network failures. System performance dictates that response times need to be minimised, sufficient redundant capacity installed in case of failure and controls embedded within the design to manage the exceptional situations (such as media stimulated events) that continually threaten network integrity. Network design based on a 'top-down,' 'end-to-end' methodology plays a fundamental role in delivering solutions that meet customers' performance needs. It is necessary to consider service scenario mixes, service demand, physical network topology, signalling message flows, the mapping of functional entities to physical components, and routing as part of the network design process to ensure that performance requirements are met. The use of 'what-if' design tools is particularly relevant as part of this process. A challenging task faces the System Designer with the often conflicting goals of good performance and provision of service flexibility.

We propose a new network architecture nemed Holonic Network for personalized multimedia communications, which is characterized by distributed cooperative networking based on autonomous management and all-optical transport networks. We than propose autonomous routing method. Moreover, an information searching method and a route generation method with network maps, which are essential for this network, are proposed. Lastly, we evaluate the proposed network performance by theoretical analysis and system emulation.

A novel CDMA multi-user detection scheme, orthogonal decision-feedback detector (ODFD), is proposed for a synchronous CDMA system in this paper. It is robust for its near-far resistance and high multi-user detection efficiency with a performance similar to that of decorrelating decision-feedback detector (DDFD) but with a reduced complexity. The ODFD scheme employs a match-filter bank that matches a set of ortho-normal sequences. The ortho-normal sequences are generated by the Gram-Schmidt orthogonalisation procedure based on the spreading codes. The ODFD algorithm involves only with the ortho-normal coefficient-matrix which requires no frequent recalculations even when system parameters change. Successive decision-feedback detection is carried out immediately at the output of the ODFD match-filter bank without matrix inversion operations, resulting in a much simplified structure.

In this paper we discuss the limiting behavior of the search direction of the steepest descent method in minimizing the Rayleigh quotient. This minimization problem is equivalent to finding the smallest eigenvalue of a matrix. It is shown that the search direction asymptotically alternates between two directions represented by linear combinations of two eigenvectors of the matrix. This is similar to the phenomenon in minimizing the quadratic form. We also show that these eigenvectors correspond to the largest and second-smallest eigenvalues, unlike in the case of the quadratic form.

A new learning algorithm is proposed to enhance fault tolerance ability of the feedforward neural networks. The algorithm focuses on the links (weights) that may cause errors at the output when they are open faults. The relevances of the synaptic weights to the output error (i.e. the sensitivity of the output error to the weight fault) are estimated in each training cycle of the standard backpropagation using the Taylor expansion of the output around fault-free weights. Then the weight giving the maximum relevance is decreased. The approach taken by the algorithm described in this paper is to prevent the weights from having large relevances. The simulation results indicate that the network trained with the proposed algorithm do have significantly better fault tolerance than the network trained with the standard backpropagation algorithm. The simulation results show that the fault tolerance and the generalization abilities are improved.

A Reed-Solomon coded Type-I Hybrid ARQ scheme based on a Selective-Repeat (SR) ARQ with multicopy retransmission is proposed for mobile/personal satellite communication systems of a transmitter and a receiver both with the finite buffer. The performance of the proposed scheme on fading channels is analyzed. The basic idea of the strategy is the use of two modes; the SR mode and the multicopy mode. In the latter mode, erroneous blocks stored in the transmitter buffer are alternatively retransmitted multiple times when ν consecutive retransmissions in the SR mode are received in error. Numerical and simulation results for ν1 show that the proposed scheme presents better performance than the conventional SR+ST scheme 2 of the 2N block buffer by Miller and Lin.

In this letter, we propose a blind adaptation method for the decision feedback equalizer (DFE). In the proposed scheme, a DFE is divided into two parts: a front-end linear equalizer (LE), and a prediction error filter (PEF) followed by a feedback part. The coefficients of the filters in each part are updated using constant modulus algorithm and decision feedback prediction algorithm, respectively. The front-end LE removes intersymbol interference ISI, and the PEF with feedback part whitens the noise to reduce noise power enhanced by the LE. Pre-processing by the LE enables the DFE to equalize nonminimum phase channels. Simulation results show that the proposed scheme provides reliable convergence, and the resulting symbol error rate is much less than that of the conventional LE and very close to that of the DFE using a training sequence.

It has been known that an N-vertex binary tree can be embedded into the path and grid with dilation O(N/logN) and O((N/logN)), respectively. This paper shows that an N-vertex binary tree with proper pathwidth at most k can be embedded into the path grid with dilation O(N/N1/k) and O((N/N1/2k)), respectively.

We have proposed a digital beamforming (DBF) self-beam-steering array antenna which features maximal ratio combining enabling it to efficiently use the received power or to rapidly track the desired signal. The DBF self-beam-steering array antenna utilizes digital signal processing with an active array antenna configuration. ASIC implementation of the digital signal processor is inevitable for DBF antenna application in practical mobile communications environments. In this paper, we present a scheme for implementing a digital signal processor in ASICs using ten FPGAs (Field Programmable Gate Arrays) for the DBF self-beam-steering array antenna. Results of some experiments obtained in a large radio anechoic chamber are shown to confirm a basic function of the system.

The distributed program reliability (DPR) is a useful measure for reliability evaluation of distributed systems. In previous methods, a two-mode failure model (working or failed) is assumed for each computing node. However, this assumption is not realistic because data transfer may be possible by way of a computing node even when this node can neither execute programs nor handle its data files. In this paper, we define a new three-mode failure model for representing such a degraded operational state of computing nodes, and present a simple and efficient analysis method based on graph theory. In order to represent the degraded operational state, a given graph expressing a distributed system is augmented by adding new edges and vertices. By traversing this augmented graph, the reliability measure can be computed. Examples show the clear difference between the results of our proposed method and those of the previous ones.

A complex fuzzy adaptive decision feedback equalizer based on the RLS algorithm is proposed. The proposed equalizer not only improves the performance but also reduces the computational complexity compared with the conventional complex fuzzy adaptive equalizers under the assumption of perfect knowledge of the linear and nonlinear channels.

This paper describes a millimeter-wave (MMW) transmission system over fiber-optic links applicable for high-speed mobile communications. The system design is presented considering both the MMW radio link and fiber-optic link. To prove the capability of the MMW fiber-optic link, an experimental system has been constructed. The results of in-door transmission experiments showed that this system could be capable of transmitting 118 Mbps digital signals with a BER of less than 10-6. The developed system is easily applicable to a wireless access system which can connect subscribers with a broadband optical fiber network.

In this paper, a new spectral subtraction technique with two microphone inputs is proposed. In conventional spectral subtraction using a single microphone, the averaged noise spectrum is subtracted from the observed short-time input spectrum. This results in reduction of mean value of noise spectrum only, the component varying around the mean value remaining intact. In the method proposed in this paper, the short-time noise spectrum excluding the speech component is estimated by introducing the blocking matrix used in the Griffiths-Jim-type adaptive beamformer with two microphone inputs, combined with the spectral compensation technique. By subtracting the estimated short-time noise spectrum from the input spectrum, not only the mean value of the noise spectrum but also the component varying around the mean value can be reduced. This method can be interpreted as a partial construction of the adaptive beamformer where only the amplitude of the short-time noise spectrum is estimated, while the adaptive beamformer is equivalent to the estimator of the complex short-time noise spectrum. By limiting the estimation to the amplitude spectrum, the proposed system achieves better performance than the adaptive beamformer in the case when the number of sound sources exceeds the number of microphones.

Recently, various efficient algorithms for solving combinatorial optimization problems using BDD-based set manipulation techniques have been developed. Minato proposed O-suppressed BDDs (ZBDDs) which is suitable for set manipulation, and it is utilized for various search problems. In terms of practical limits of space, however, there are still many search problems which are solved much better by using conventional branch-and-bound techniques than by using BDDs or ZBDDs, while the ability of conventional branch-and-bound approaches is limited by computation time. In this paper, an extension of APPLY operation, named APPRUNE (APply + PRUNE) operation, is proposed, which performs APPLY operation (ZBDD construction) and pruning simultaneously in order to reduce the required space for intermediate ZBDDs. As a prototype, a specific algorithm of APPRUNE operation is shown by assuming that the given condition for pruning is a threshold function, although it is expected that APPRUNE operation will be more effective if more sophisticated condition are considered. To reduce size of ZBDDs in intermediate steps, this paper also pay attention to the number of cared variables. As an application, an exact-minimization algorithm for generalized Reed-Muller expressions (GRMs) is implemented. From experimental results, it is shown that time and memory usage improved 8.8 and 3.4 times, respectively, in the best case using APPRUNE operation. Results on generating GRMs of exact-minimum number of not only product terms but also literals is also shown.

In this paper, we introduce a parallel algorithm for parsing context-free languages. Our algorithm can handle arbitrary context-free grammars since it is based on Earley's algorithm. Our algorithm can operate on any loosely coupled multiprocessor which can provide a topology of a one-way ring. Our algorithm uses p processors to parse an input string of length n where 1 p n. It is shown that our algorithm requires O(n3/p) time. The algorithm uses a simple job allocation strategy. However, it achieves high load balancing and uses the processors efficiently.

This paper describes how to effectively use discriminative training based on Minimum Classification Error (MCE) criterion for a small amount of data in order to attain the highest level of recognition performance. This method is a combination of MCE training and Vector-Field-Smoothed Bayesian learning called MAP/VFS, which combines maximum a posteriori (MAP) estimation with Vector Field Smoothing (VFS). In the proposed method, MAP/VFS can significantly enhance MCE training in the robustness of acoustic modeling. In model training, MCE training is performed using the MAP/VFS-trained model as an initial model. The same data are used in both trainings. For speaker adaptation using several dozen training words, the proposed method has been experimentally proven to be very effective. For 50-word training data, recognition errors are drastically reduced by 47% compared with 16.5% when using only MCE. This high rate, in which 39% is due to MAP, an additional 4% is due to VFS, and a further improvement of 4% is due to MCE, can be attained by enhancing MCE training capability by MAP/VFS.

The Time-Slicing paradigm is a newly developed method for the training of neural networks for speech recognition. The neural net is trained to spot the syllables in a continuous stream of speech. It generates a transcription of the utterance, be it a word, a phrase, etc. Combined with a simple error recovery method the desired units (words or phrases) can be retrieved. This paradigm uses a recurrent neural network trained in a modular fashion with natural connectionist glue. It processes the input signal sequentially regardless of the input's length and immediately extracts the syllables spotted in the speech stream. As an example, this character string is then compared to a set of possible words, picking out the five closest candidates. In this paper we describe the time-slicing paradigm and the training of the recurrent neural network together with details about the training samples. It also introduces the concept of natural connectionist glue and the recurrent neural network's architecture used for this purpose. Additionally we explain the errors found in the output and the process to reduce them and recover the correct words. The recognition rates of the network and the recovery rates for the words are also shown. The presented examples and recognition rates demonstrate the potential of the time-slicing method for continuous speech recognition.

Very-low-voltage 1.2-V mixed-signal CMOS technology is a device/circuit solution aimed at ultra-low-power portable systems such as digital cellular terminals and PDAs. We have developed an experimental 1.2-V mixed analog and digital LSI circuit/device technology. This technology is based on a new transistor structure that has a 0.3-µm gate length and a low Vth of 0.4 V, and that suppresses the short-channel effect. In this paper, we will mainly discuss low-voltage analog circuit design that uses this technology. We show that low Vth is essential not only to digital circuits, but also to 1.2-V analog amplifier, A/D converter and analog switch designs. To achieve high-conversion rate A/D converters, a pipeline architecture is used for low-voltage operation. To increase the attainable gain-bandwidth of the operational amplifier of the converter, a feedforward phase-compensated three-stage amplifier is proposed. The addition of a feedforward capacitor allows a high frequency signal to pass directly to the second stage, which optimizes use of the second stage bandwidth. Pole-zero canceling is used to achieve a fast settling of the amplifier. Although gain precision is degraded by the positive feedback through the feedforward capacitor, this can be offset by increasing the equivalent second-stage gain with an inner feedforward compensated amplifier. The gain-bandwidth of the proposed double feedforward amplifier is two to three times wider than with the conventional Miller compensation. With these techniques, we used 1.2-V mixed-signal CMOS technology to create a basic logic gate with a 400-ps delay and 0.4-µW/MHz power, and a 9-bit 2-Msample/s pipeline A/D converter with power dissipation of only 4 mW.

A generalized Reed-Muller expression (GRM) is obtained by negating some of the literals in a positive polarity Reed-Muller expression (PPRM). There are at most 2(n2)^(n-1) different GRMs for an n-variable function. A minimum GRM is one with the fewest products. This paper presents certain properties and an exact minimization algorithm for GRMs. The minimization algorithm uses binary decision diagrams. Up to five variables, all the representative functions of NP-equivalence classes were generated and minimized. Tables compare the number of products necessary to represent four-and five-variable functions for four classes of expressions: PPRMs, FPRMs, GRMs and SOPs. GRMs require, on the average, fewer products than sum-of-products expressions (SOPs), and have easily testable realizations.