In this paper, a novel variable-rate vector quantizer (VQ) design algorithm using fuzzy clustering technique is presented. The algorithm, termed fuzzy entropy-constrained VQ (FECVQ) design algorithm, has a better rate-distortion performance than that of the usual entropy-constrained VQ (ECVQ) algorithm for variable-rate VQ design. When performing the fuzzy clustering, the FECVQ algorithm considers both the usual squared-distance measure, and the length of channel index associated with each codeword so that the average rate of the VQ can be controlled. In addition, the membership function for achieving the optimal clustering for the design of FECVQ are derived. Simulation results demonstrate that the FECVQ can be an effective alternative for the design of variable-rate VQs.

Equipment simulation can provide valuable support in reactor design and process optimization. This article describes the physical and chemical models used in this technique and the current state of the art of the available software tools is reviewed. Moreover, the potential of equipment simulation will be highlighted by means of three recent examples from advanced quarter micron silicon process development. These include a vertical batch reactor for LPCVD of arsenic doped silicon oxide, a multi station tungsten CVD reactor, and a plasma reactor for silicon etching.

Recent years have seen great advances in our understanding and modeling of the coupled diffusion of dopants and defects in silicon during integrated circuit fabrication processes. However, the ever-progressing shrinkage of device dimensions and tolerances leads to new problems and a need for even better models. In this review, we address some of the advances in the understanding of defect-mediated diffusion, focusing on the equations and parameters appropriate for modeling of dopant diffusion in submicron structures.

The characteristics of the decision feedback carrier recovery loop (DFL) for conventional QPSK signaling is evaluated experimentally through measurements of the carrier-to-noise ratio of the regenerated carrier, lock range, acquisition waveforms and bit error rates. The results show that the DFL hardly exhibits inferiority to the ideal synchronization by designing the loop natural frequency adequately small. The DFL is shown superb in carrier tracking.

In this paper there suggested is a bifurcated (or multiple) input-queued ATM switch in which a buffer for each input port is divided into multiple (m) buffer blocks, i. e. , bifurcated buffers, for enhancement of the limited throughput of the ordinary input-queued switch using a single FIFO. As the contention/arbitration rule for the bifurcated input-queued switching scheme, free and restricted contention rules are come up with and discussed. The free rule allows an input port to switch up to m cells at the cost of internal speedup. With the restricted rule, on the other hand, an input port can switch no more than one cell in a time slot so that the switch operates at the same speed as the external link speed. The throughput bound for the bifurcated input-queued switch is analyzed for both rules through the generalization of the analysis by Karol et al. The throughput bound approaches to 1.0 as m becomes large enough, irrespective of the contention/arbitration rule.

Frequency chirping induced in an electorabsorption (EA) modulator can degrade transmission performance because of the chromatic dispersion of fiber. This letter studies the frequency chirping in an EA modulator from the viewpoint of the influence of the modulation bandwidth. Both simulations and experiments, in which fiber transmission was carried out applying modulation signals of different bandwidths to an EA modulator, show that a large bandwidth causes small degradation in the transmission performance. This result is attributed to the short chirping time that occurs when a large bandwidth signal is applied.

We are concerned with a CDMA cellular system with unbalanced traffic environment such that a cell with high traffic should be assigned a frequency additional to the common one shared with its neighboring cells. To remove quality-dropping inter-cell hard handoffs, the cell with high traffic was first partitioned into three regions. Then, the common frequency is assigned in the outermost region, and the additional frequency is operated in the innermost region, whereas both the common frequency and the additional frequency are operated in the middle region. This frequency assignment strategy is shown not only to remove inter-cell hard handoffs without requiring extra hardware, but also to reduce intra-cell hard handoffs.

This paper proposes a new distortion analysis method for frequency-dependent FET amplifiers, which uses a novel Frequency-Dependent Complex Power Series (FDCPS) model. This model consists of a frequency-independent nonlinear amplifier represented by an odd-order complex power series and frequency-dependent input and output filters. The in-band frequency characteristics of the saturation region are represented by the frequency-dependent output filter, while the in-band frequency characteristics of the linear region are represented by the frequency-dependent input and output filters. In this method, the time-domain analysis is carried out to calculate the frequency-independent nonlinear amplifier characteristics, and the frequency-domain analysis is applied to calculate the frequency-dependent input and output filter characteristics. The third-order intermodulation (IM3) calculated by this method for a GaAs MESFET amplifier is in good agreement with the numerical results obtained by the Harmonic Balance (HB) method. Moreover, the IM3 calculated by this method also agrees well with the measured data for an L-band 3-stage GaAs MESFET amplifier. It is shown that this method is effective for calculating frequency-dependent distortion of a nonlinear amplifier with broadband modulation signals.

The architecture and control procedure for a direct conversion receiver are investigated for a linear modulation scheme. The proposed design techniques maintain receiver linearity despite various types of signal distortion. The techniques include the fast gain control procedure for receiving a control channel for air interface connection, DC offset canceling in both analog and digital stages, and 2nd-order intermodulation distortion canceling in an analog down-conversion stage. Experimental and computer simulation results on PHS (Personal Handy-phone System) parameters, showed that required linear modulation performance was achieved and thus the applicability of the proposed techniques was demonstrated.

We propose an H-plane manifold-type triplexer with closely arranged junctions. Broadband characteristics for each bands are obtained by arranging filters closely near the end of the common waveguide. Three fundamental and sufficient parameters are introduced for numerical optimizations to determine the configuration of the broadband triplexer. The configuration including closely arranged junctions requires an generalized scattering matrix (GS matrix) of an asymmetric cross junction to simulate and design. We expand the mode matching technique (MMT) to be able to analyze this kind of discontinuities by joining two asymmetric steps discontinuities to a symmetric cross junction. This is suitable expressions for numerical calculations. The characteristics of the whole triplexer are obtained by cascading GS matrices of the corresponding discontinuities. The experimental results of the fabricated triplexer were compared with the simulated data, and the results agree well with the simulated one. The characteristics of the fabricated triplexer satisfy the request of the broad band operation and high power-handling capability.

An efficient ARQ scheme based on the packet combining technique is investigated for multi-carrier modulation systems. In multi-carrier modulation systems, several sub-carriers are used for high data rate transmission and their individual received signal quality becomes different from one sub-carrier to others in a frequency selective fading channel. Therefore by changing the assignment of data to the sub-carriers in the retransmission packets, the distortion between the previous transmitted packet and the newly retransmitted one will be different. This is the principle of the proposed adaptive data order rearrangement for a packet combining ARQ scheme, which can achieve more diversity gain in packet combining and improve the ARQ performance. From the results of the theoretical analysis and the computer simulation, it is confirmed that the proposed packet combining ARQ with the proposed operation can achieve the better performance in terms of the average packet transmission success probability. In addition, this proposed scheme is also compared with the conventional multi-carrier modulation ARQ scheme based on the partial retransmission of a packet. The computer simulation results demonstrate that the proposed scheme has also advantage against the latter one, and it is considered to be as a more efficient ARQ scheme for multi-carrier modulation systems.

This paper proposes a newly developed dual-frequency antenna for 800 MHz and 1500 MHz band use. A uniformly spaced array configuration, originally designed for a 800 MHz analog system, is extended to yield dual frequencies operations. An important characteristic of a base station antenna is low sidelobe level in order to suppress inter-cell interference. In the case of a uniformly spaced array configuration, sidelobe levels are increased by the emergence of grating lobes at higher frequencies. Electrical beam tilt also degrades the sidelobe level. As does the change in the excitation coefficients of the array elements at higher frequencies. These three factors are studied theoretically to yield practical sidelobe levels. One more important beam characteristic is the sector beam in the horizontal plane. The same beam width in two frequency bands is achieved by designing the novel reflector shape and determining the proper array element positions in front of the reflector. Practical antenna characteristics are confirmed by designing, manufacturing, and testing a base station antenna.

Multicasting has become an important part of current networking technologies. To accommodate the multicasting traffic, switches in ATM (Asynchronous Transfer Mode) networks need to be able to support multicasting. In multicast ATM switches, usually there are many cells contending for the same output port resources at the same time. This may lead to serious performance degradation. In the past, many special output port components are proposed to solve this problem. However, most of these approaches alleviate the output port overflowing problem without considering the output port congestion or the internal routing blocking problem. Especially when the switch fabric is based on the banyan-like network, these two problems will seriously degrade the switch performance. In this paper, we propose a pre-scheduling mechanism to precede the multicast switch. This mechanism will completely eliminate the problems mentioned above. Two architectures are designed to implement this mechanism. Analysis and simulations show that our design performs quite well.

An Acoustic echo canceller has problems adaptating under noisy or double-talk conditions. The adaptation process requires a precise identification of the temporarily changed room impulse response. To do this, both minimizing the step size parameter of the Least Mean Square (LMS) method to be as small as possible and giving up on updating the adaptive filter coefficients have been considered. This paper describes an adaptive cross-spectral technique that is robust to adaptive filtering under noisy or double-talk conditions and for colored signals such a speech signal. The cross-spectral technique was originally developed to measure the impulse response in a linear system. Here we apply in the adaptive cross-spectral technique to solve the acoustic echo cancelling problem. This cross-spectral technique takes the ensemble average of the cross spectrum between input and error signals and the averaged cross spectrum is divided by the averaged power spectrum of the input signal to update the filter coefficients. We have confirmed that the echo signal is suppressed by about 15 dB even under double-talk conditions. We also explain that this method has a systematic error due to using a short time block for estimating the room impulse response. Then we investigate overlapping every last half block by the following first half block in order to reduce the effect of the systematic error. Finally, we compare our method with the Frequency-domain Block LMS (FBLMS) method because both methods are implemented in the frequency domain using a short time block.

This paper proposes a phase-rotating phase-shift keying (PSK) modulation and shows that its narrow-band version is suitable for Viterbi equalization. The proposed PSK has the following features: 1) a spectrum shaping of the transmit/receive filters does not need to be restricted to the Nyquist criterion; 2) the transmitted data sequence is rotated for every symbol in order to reduce noise-correlation at the receiver. First, this paper discusses a performance degradation of bit error rate of Viterbi equalizers in the presence of the sampling timing offset or under time-dispersive frequency selective fading. Next, computer simulation confirms that π/2-shifted binary PSK with narrow-band spectrum shaping filter, which includes offset QPSK for its special case, solves the above mentioned performance degradation, keeping good spectrum efficiency equal to M-ary PSK.

In the basic user/network interface of ISDN (ITU-T Recommendation I. 430), the D-channel is shared by up to 8 terminals for signal and data packets. An analytical model is proposed to reveal the performance characteristics of the access control scheme for the D-channel. Numerical and simulation results are shown to demonstrate the performance differentiation of the terminals with different priorities. It is observed that the mean signal delay at low load may become large because of long service time for packets, and that the priority mechanism may not work properly when the loads at terminals are very asymmetric.

The frequency offset estimation is used to correct any frequency error of the local reference oscillator. In this letter, a frequency offset estimation algorithm utilizing the peak phase error detection and frequency offset smoother is proposed for burst data transmission. The basic idea of frequency offset estimator is to use a curve fitting method. The proposed peak phase error detection avoids a large phase error which yields a bad value for FOE. In order to control the AFC, frequency offset smoother by a simple filter is used. Simulation results show that the proposed algorithm is adequate for frequency offset estimation of burst data transmission.

An ATM buffer management scheme with logically separated buffers is proposed to guarantee various Quality of Services such as CBR, rt-VBR, nrt-VBR, ABR, and UBR and to make efficient use of system resources in ATM switching systems. By assigning proper priorities based on the count indicator (CI) and the time indicator (TI), respectively, the scheme can afford ABR services without cell loss and real-time services with controllable delay.

This paper proposes a method of improving the accuracy of the attenuation constant estimate obtained by using the cross-spectral technique. In the cross-spectral technique, the envelope of the estimated impulse response is deformed due to the use of a time window. As a result, the estimated impulse response decays more rapidly than the real impulse response does, and the attenuation constant obtained by the estimated impulse response becomes larger than the real value. This paper first describes how the attenuation constant changes in the process of impulse response estimation. Next, we propose a method of improving the accuracy of the estimation. The effect of the proposed method is confirmed by computer simulation.

The conventional linear prediction can be viewed as a constrained blind equalization problem that has gained a lot of interests along with development of telecommunication networks. Because the blind equalization or deconvolution is a general framework of the inverse problem, the reliable and faster algorithm is requested in many applications. This paper proposes an orthogonal wavelet transform domain realization of a blind equalization technique termed as EVA, and presents an application to speech analysis. An orthogonal transformation has no influence to the equalization result in general, but we show that a particular wavelet makes the matrix in EVA nearly lower triangular that promotes the faster convergence in the estimation of maximum eigenvalue and its associate vector in EVA iteration. The experiments with the Japanese vowels show that the the proposed method effectively separates the glottis and vocal tract information, hence is promising for speech analysis.