Three multiplicative algorithms for the floating-point divide operation are compared: the Newton-Raphson method, Goldschmidt's algorithm, and a naive method that simply calculates a form of the Taylor series expansion of a reciprocal. The series also provides a theoretical basis for Goldschmidt's algorithm. It is well known that, of the Newton-Raphson method and Goldschmidt's algorithm, the former is the more accurate while the latter is the faster on a pipelined unit. However, little is reported about the naive method. In this report, we analyze the speed and accuracy of each method and present the results of numerical tests, which we conducted to confirm the validity of the accuracy analysis. Basically, the comparison are made in the context of software implementation (e. g. , a macro library) and compliance with the IEEE Standard 754 rounding is not considered. It is shown that the naive method is useful in a realistic setting where the number of iterations is small and the method is implemented on a pipelined floating-point unit with a multiply-accumulate configuration. In such a situation, the naive method gives a more accurate result with a slightly lower latency, as compared with Goldschmidt's algorithm, and is much faster than but slightly inferior in accuracy to the Newton-Raphson method.

A method of visualization of multimodal images by one monochromatic image is presented on the basis of the projection pursuit approach of the inverse process of the anisotropic diffusion which is a method of image restoration enhancing contrasts at edges. The extension of the projection from a linear one to nonlinear sigmoidal functions enhances the contrast further. The deterministic annealing technique is also incorporated into the optimization process for improving the contrast enhancement ability of the projection. An application of this method to a pair of MRI images of brains reveals its promising performance of superior visualization of tissues.

We have proposed an algorithm to apply perfectly matched layer (PML) absorbing boundary condition to the noncubic cell time-domain method. The extended method has a merit of flexibility in truncating the computational domain by the use of a curvilinear PML. In this paper we apply a circular PML for computing the scattered fields of a dielectric cylinder or cylindrical shell of arbitrary cross section shape. Numerical results are presented to demonstrate the accuracy of this method.

This paper describes a method to represent m output functions using shared multi-terminal binary decision diagrams (SMTBDDs). The SMTBDD(k) consists of multi-terminal binary decision diagrams (MTBDDs), where each MTBDD represents k output functions. An SMTBDD(k) is the generalization of shared binary decision diagrams (SBDDs) and MTBDDs: for k=1, it is an SBDD, and for k=m, it is an MTBDD. The size of a BDD is the total number of nodes. The features of SMTBDD(k)s are: 1) they are often smaller than SBDDs or MTBDDs; and 2) they evaluate k outputs simultaneously. We also propose an algorithm for grouping output functions to reduce the size of SMTBDD(k)s. Experimental results show the compactness of SMTBDD(k)s. An SMTBDDmin denotes the smaller SMTBDD which is either an SMTBDD(2) or an SMTBDD(3) with fewer nodes. The average relative sizes for SBDDs, MTBDDs, and SMTBDDs are 1. 00, 152. 73, and 0. 80, respectively.

A thin-film waveguide amplifier based on Er-doped Garnet crystals is proposed and transient amplification characteristics, studied numerically using time-dependent rate equations and mode evolution equations, are presented. The potential of the amplifier for integration with active devices operating at the present communication wavelength of 1. 53 µm band is revealed. Pump wavelengths in the visible and near infrared lead to excited state absorption, and will affect the gain characteristics, which has been included in the present study. Steady state response of the Er doped Garnet crystal waveguide amplifiers has been analyzed in order to optimize the gain characteristics, which are further used in the dynamic response analysis. Accordingly, it is shown that a high gain of 20 dB/cm is possible to be achieved. Experimentally determined parameters such as waveguide loss, absorption and emission cross-sections have been used for the simulations. Comparisons of the present simulation results with our earlier reported results of quasi-two-level laser model and other reported results are also presented. Understanding the dynamic characteristics of the integrated optic waveguide amplifiers is necessary when the input signal is modulated in various formats. Because of the slower gain dynamics of the Er doped Garnet amplifier medium, it is shown that the longer signal input pulses are observed to be distorted upon amplification. Very short single pulse of nano- and pico-second duration are amplified without change in the pulse shape. Input pulses of square, Gaussian and Lorentzian shapes have been considered for the numerical examples.

A Pragmatic Trellis Coded MPSK on a Rayleigh fading channel is analyzed. This scheme allows bandwidth expansion ratio to be varied aiming at an optimization between complexity of the system design and improvement of coding gain. In order to vary the bandwidth expansion ratio, a punctured convolutional code is used. The performance of the proposed TC-2mPSK on a Rayleigh fading channel is theoretically analyzed. In the test examples, the BER performances of TC-QPSK and TC-8PSK are evaluated by theoretical analyses and computer simulations at the encoder parameters of K3 and r3/4. The results show that the proposed scheme can attain better performance not only over the uncoded scheme but over the conventional Pragmatic TCM.

A fuel cell energy system is under development for supply of generated electrical energy to telecommunications equipment. It is a cogeneration system; the heat energy recovered is used to cool the telecommunications equipment. For this system, a method is described for controlling a new DC interconnection converter. Its DC interconnection characteristics are also discussed. The new converter controls its input current to the fuel cell rated current at maximum and can operate stably even when the fuel cell voltage decreases. This allows good DC interconnection characteristics to be obtained in both the steady state and the transient state.

Cursor RAMs have been composed of two memory planes. A cursor pattern is stored in these planes with 2-bit data depth. While the pixel port requires data from both planes at the same time, the MPU port accesses either one of the planes at a time. Since the address space is defined differently between the ports, conventional cursor RAMs could not have dealt with these different access ways at real time. This paper proposes a dual port cursor RAM with a dynamic data alignment architecture. The architecture processes the different access ways at real time, and reduces a large amount of control circuitry. Conventional cursor RAMs have been organized with a single port memory because dual port memory cells have been large. We have applied the port swap architecture which has reduced the cell size. The control block is further simplified because the controller no longer emulate a dual port memory. The cursor RAM with these architectures is fabricated with a double metal 0. 5 µm CMOS process technology. The active area is 1. 51. 6 mm2 including a couple of shift registers and a control block. It operates up to 263 MHz at the supply voltage of 3. 3 V.

Initial leveling of strapdown inertial navigation system is a prerequisite work for distinguishing between gravity and acceleration effects in the accelerometer sensing's. This study presents an on-line methodology to resolve the initial leveling problem of a vehicle, which is subject to a large, long duration, and abrupt disturbance input with a deterministic nature under noisy circumstances. The developed method herein is the Kalman filter based scheme with a robust input estimator, generalized M estimator, and a testing criterion. The generalized M estimator identifies the unexpected disturbance inputs in real time. In addition, hypothetical testing based on the least-squares estimator is devised to detect the input's onset and presence. A required regression equation between the observed value of the residual sequence with an unknown input and theoretical residual sequence of the Kalman filter with no input is formulated. Input estimation and detection are then provided on the basis of the derived regression equation. Moreover, Monte Carlo simulations are performed to assess the superior capabilities of the proposed method in term of rapid responses, accuracy, and robustness. The efficient initial leveling can facilitate the entire alignment of the inertial system.

This letter proposes a new retransmission control scheme for slow-frequency-hopped communication systems, in which the number of (re)transmitted packets is adaptively decreased in a certain period. Performance of the proposed scheme is analyzed and compared with that of the conventional scheme in terms of the normalized throughput and the 98% packet transmission delay. The numerical results show the superiority of the proposed scheme.

Array redistribution is required very often in programs on distributed memory parallel computers. It is essential to use efficient algorithms for redistribution, otherwise the performance of programs may degrade considerably. In this paper, we focus on automatic generation of communication routines for multi-dimensional redistribution. The principal advantage of this work is to gain the ability to handle redistribution between arbitrary source and destination processor sets and between arbitrary source and destination distribution schemes. We have implemented these algorithms using Parallelware communication library. Some experimental results show the efficiency and flexibility of our techniques compared to the other redistribution works.

A partial buffer sharing scheme is proposed as loss-priority control for a finite buffer with batch inputs. A partial batch acceptance strategy is used for a batch arriving at a finite buffer. Customer loss probabilities for high- and low-priority customers are derived under this batch acceptance strategy, using a supplementary variable method that is a standard tool for queueing analysis. A comparison of the partial buffer sharing scheme and a system without loss-priority control is made in terms of admissible offered load.

One of the objectives of ABR traffic control is a fair allocation of transmission rate to every connection. Algorithms which have been proposed can achieve the fair allocation of transmission rate in persistent traffic, even though the connections have different number of hops. In bursty traffic, however, the transmission rate during the burst period, called average burst throughput, may become degraded when the number of hops increases. In this paper, we evaluate the average burst throughput characteristics of ERICA switch algorithm and investigate the effect of the number of hops. We propose a new bandwidth allocation method which can improve the performance degradation by taking the number of hops into account. The effectiveness of our method is evaluated by computer simulation.

In the ITS (Intelligent Transport Systems), it is an essential condition (mixed environment) that vehicles that have communication equipment and vehicles that do not have it simultaneously run in the same road. In this paper, a vehicular driving assistant system that is applicable to the mixed environment is proposed. The proposed system uses spread spectrum techniques and consists of several new systems such as a PN code assignment system, new vehicle position systems, and a vehicle map update system. In the proposed system, the wireless broadcast CDMA is used for inter-vehicle communications. This paper also shows preexaminations of the proposed system by using an autonomous traffic flow simulator including inter-vehicle communications. It is shown that the traffic safety can be improved by using inter-vehicle communications.

QoS of a multimedia interactive satellite communications system, which uses a satellite circuit for downlinks and a terrestrial network for uplinks, has been measured and evaluated under inferior satellite circuit conditions. Our attention in this paper is focused on TCP throughput as the most significant QoS, and experiments and analysis on TCP throughput were performed. The measured results are in good agreement with theoretical values calculated by taking congestion avoidance algorithms of TCP into consideration. As a result, we were able to verify that the TCP throughput declines radically when the low signal quality of a satellite circuit causes packet retransmission to occur. In this case, congestion avoidance works to reduce network congestion. The congestion avoidance increases the value of a retransmission timer and decreases the window size even if the network is not congested. Assuming that throughput deterioration is caused by rain fade, the total time when the throughput declines is less than 4 hours a year in Tokyo, and the results show the system can be put to practical use.

This paper formulates problems of fitting two corresponding sets of points by translation, rotation and scaling, and proposes efficient algorithms for the fitting. The algorithms are based on the theory of lower envelopes, or Davenport-Schinzel sequences, and linearization techniques in computational geometry, and are related to dynamic furthest Voronoi diagrams.

In this paper, we study unequal error protection (UEP) capabilities of punctured convolutional codes. For constructing the good UEP convolutional codes, the conditional weight distributions of UEP convolutional codes are defined and evaluated. The conditional weight distributions are computed by using the transfer functions of time-varying trellis structures of punctured convolutional codes. The best UEP convolutional codes from the viewpoint of the weight distributions are listed.

A turbo TCM system is applied to a channel with overall noise which is equal to the additive combination of impulsive Gaussian noise and Additive White Gaussian Noise (AWGN). By taking the distribution of the previously mentioned overall noise into account, a decoding algorithm for Poisson occurrence impulsive noise is derived as an extension of that for AWGN. A simulation result shows that Eb/N0 difference from Shannon limit to realize BER=10-4 is 0. 493 dB. To investigate the effect of burst noise, we discuss the case of additive impulsive noise with Markovian occurrence which is represented by Hidden Markov Model. A decoding algorithm for Markovian noise is proposed. In the iterative decoding for the Markovian channel, the decoding algorithms for Markovian and Poisson noise are applied separately to the two component decoders. The decoding algorithm for Markovian noise is used in the component decoder wherein received signal is directly fed, while the decoding algorithm for Poisson noise is used in the component decoder wherein received signal is fed after passing an interleaver. This paper also shows simulation results that include the effects of varying the noise parameters in the decoding. In the Markovian case, when smaller value of variance of impulsive noise is used, the observed flattening of BER performance is more serious compared to the effect in the Poisson noise channel. No flattening is observed when large value is used.

We discuss the security of the improved knapsack cryptosystem that Kobayashi and Kimura have proposed. Two attacking methods for their cryptosystem are proposed; one is the method for obtaining secret keys from public keys by using the continued fraction, and the other is for decrypting the ciphertext without knowing secret keys. We show that their cryptosystem is not secure against these attacks.

This paper studies how the self-similarity of traffic changes through shaper (buffered leaky bucket) and switch in ATM networks by numerical experiments. Further the applicability of CAC algorithm to shaped self-similar traffic is also investigated. Numerical experiments show self-similarity of total output traffic from shapers and switch is kept while connection-wise self-similarity is broken.