In a semiconductor fabrication plant, various types of sensors are installed at various equipment of various processes to monitor the quality of the products and the progress of various processes. These sensors generate a huge volume of time-series waveform data, which are used for failure prognostics, failure diagnosis, and anomaly detection using supervised or unsupervised classifiers. For this purpose, automatic extraction of features from a huge volume of waveform data is needed; however, extraction of effective features automatically from a compound waveform having multiple spikes, complex states or transitions is difficult. If a compound waveform is properly segmented into multiple sub-waveforms, effective features from sub-waveforms can be extracted. In this paper, we propose a new waveform segmentation method based on two-step state and change-point detection using Kernel Density Estimation (KDE) and clustering techniques and apply IEEE Standard-based methods with our state determination technique to extract features from sub-waveforms. We apply our proposed method to waveform data of real sensors installed at one of our semiconductor fabrication plants and compare the performance with the conventional step-based segmentation technique and another pattern matching technique. The experimental results show that our proposed method enables extraction of effective features, which result in higher accuracy of anomaly detection compared to conventional techniques.

We propose a robust scheme for detecting vehicles approaching from several directions to prevent crossing collisions. The scheme consists of three sequential operations. The first operation computes cross-power spectrum phase (CSP) coefficients of sound signals from a microphone array; the second operation classifies the information obtained by the CSP coefficients into multiple ranges of directions so as to separate multiple approaching vehicles. The last one detects approaching vehicles through a threshold operation of the clustered information. In order to evaluate the performance of the scheme, we constructed a large-scale database based on the acoustic and related information concerned with approaching vehicles. Using the database, we show that the scheme is robust for detecting not only a single but also multiple approaching vehicles.

Packet detection is one of the most important problems in packet communication systems. In a CDMA Unslotted ALOHA system, which is one of the wireless packet communication systems, multiple access interference (MAI) makes the performance of the packet detection worse. To reduce the effect of MAI, we propose a new packet detection scheme in this paper. In this scheme, Successive Interference Cancellation(SIC) is applied. The packet's signal is detected after the cancellation of MAI using SIC. This proposed scheme gives good performance of the packet detection.

We compare CDMA and FDMA (or TDMA) schemes in multibeam non-geostationary satellite systems, taking the effects of antenna patterns on inter-cell interference into consideration. The maximum acceptable number of users per cell in the uplink is employed as a measure of the system capacity. We have found that in FDMA (or TDMA) systems the maximum acceptable number of users varies according to the altitude of the satellites. While the performance is insensitive to the altitude in CDMA systems. For this reason, the altitude of the satellites is found to be an important factor for selecting multiple access schemes in non-geostationary satellite systems.

In wireless communication systems, as one of the effective techniques for combating fading, transmit diversity has attracted much attention, especially when receive diversity is expensive or impractical due to the constraint of terminal size. In this study we consider an adaptive carrier-by-carrier basis antenna selection scheme for transmit diversity in OFDM systems under the assumption that a transmitter has priori knowledge of the channel state information (CSI). In the proposed schemes transmit antennas are selected adaptively for each subcarrier according to CSI.

A modified version of the SAGE algorithm applied for joint estimation of delays, incidence azimuths and attenuation parameters characterizing the channel is presented. The modifications are based on using different observation intervals in optimizing the parameters and using time correlation in estimating delays, for the purpose of a further reduction in SAGE's computational weight, and improving its accuracy in delay estimation. We show analytically that short observation intervals are sufficient for optimizing the delays and the azimuths, and verify through simulation the reliability of using time correlation. Two other estimators are proposed combining the modified SAGE with a sequential estimation of the attenuation parameters suitable for slowly varying channels.

In CDMA cellular systems, multicode transmission is an effective way of accommodating multimedia traffic having various transmission rates. We proposed a common searcher configuration at a base station to successfully handle multicode signals. The number of searchers needed in a base station can be reduced because a multicode signal needs only one searcher. We evaluated the call blocking probability and communications quality when the number of searchers is restricted.

In an attempt to improve the performance under a frequency selective fading environment, we propose an orthogonal frequency division multiple access (OFDM) system in which MTCM (multiple trellis coded modulation) is applied to a set of multiple subcarriers. The code is designed to enlarge the summation of the Euclidean distance and the product distance. We also propose a scheme for assigning the symbols to subcarriers adaptively based on the channel state information (CSI) fed back from the reception receive end, and evaluate the performance applying these two codes and the proposed scheme under a multipath Rayleigh fading environment.

This paper proposes a transmission power control scheme for downlinks in CDMA/shared-TDD cellular packet systems that experience inter-cell interference between uplinks and downlinks. The proposed scheme increases the downlink transmission power for re-transmission packets and thus improves the signal-to-interference ratio for the packets. Simulation results show that the proposed scheme improves the downlink throughput without sacrificing the uplink throughput.

A CDMA unslotted ALOHA system using MMSE multi-user detection is proposed. In the CDMA unslotted ALOHA system, a user station can transmit a packet asynchronously, and so the number of simultaneously transmitted packets fluctuates moment by moment. This fluctuation affects the performance of the MMSE multi-user detection. At first, we employ the Wiener filter detector assuming perfect information as an upper bound of the performance of the proposed system. Then, we apply an adaptive filter to CDMA unslotted ALOHA to compensate for the birth/death of a packet.

This paper examines the inter-cell interference between the uplinks and downlinks when a shared-TDD scheme is applied to CDMA packet communication systems in cellular environments. The CDMA/shared-TDD cellular packet systems experience two additional types of interference: one is the interference between base stations and the other is between mobile stations. The effect of these interferences on the throughput performance is evaluated. Computer simulation demonstrated that interference between base stations rarely degrades uplink throughput, but interference between mobile stations substantially degrades downlink throughput.

This paper introduces a variable classification technique to support the maintenance of COBOL application systems. It classifies program variables according to what domain entities they represent. It supports various types of software maintenance. In a system modification, it is necessary to identify the source code related to the problem at hand properly. This maintenance phase, called the impact analysis, involves significant costs in large-scale business application systems. The variable classification presented supports finding all variables to be checked and modified. In a system restructuring, the variable classification is needed to restructure data declarations. The behavior of the classification can be changed interactively with several parameters. This paper also presents these parameters and how to set them for each type of maintenance task.

We propose the introduction of space diversity techniques to the code acquisition of a direct-sequence spread-spectrum signal. In this scheme, both a transmitter and a receiver have multiple antennas and the signals corresponding to all the combinations of the transmitter/receiver antennas are combined at the acquisition circuit of the receiver. The performance is evaluated for indoor packet radio communication systems from the viewpoints of average time for acquisition, probability of success of acquisition, and necessary preamble length. As the result, we show great performance improvements by the proposed scheme under slow and flat Rayleigh fading environment.

In this paper, we propose a CDMA slotted-ALOHA (CDMA S-ALOHA) system with low-bit rate and high-bit-rate classes of data users applying the multi code CDMA (MC-CDMA) scheme. Performance evaluation is obtained in the respect of the throughput. Contrary to one might expect, we show that the throughput of the class with higher bit rate can be improved by the MC-CDMA scheme. Moreover, we show that applying the modified channel load sensing protocol (MCLSP) to the system gives further improvement in the throughput of the higher bit rate class, as well as to the total throughput of the channel.

In this paper, we analyze the CDMA slotted ALOHA systems with finite buffer and finite population assumptions by expanding the linear approximate solution based on restricted occupancy urn models. We evaluate the system performance in terms of throughput, average delay, and rejection probability. Moreover, we clarify the effect of buffer capacity by changing the buffer size.

In this paper, we discuss the performance of a system in which multiprocessing gain (MPG) spread spectrum signals are transmitted to a base station within the same bandwidth. At the receiver of the base station, an adaptive array antenna is employed. We evaluate the system performance in terms of signal to interference-and-noise ratio (SINR) and also of outage probability of low-rate users. We find that the employment of the adaptive array antenna may result in serious degradation of the outage probability of low-rate users. We then show that the cause of this serious degradation is the inadequate guidelines of conventional adaptive array antennas.

We consider a receiver with satellite diversity by using DS/SS signals from different LEO satellites. Since the signals from the different LEO satellites may have different frequency offsets due to the Doppler effect, we propose an acquisition scheme which can estimate the code timings and carrier frequencies of these signals at once. We evaluate the characteristics of the proposed scheme by the mean acquisition time.

The effect of multiple-access interference on the throughput performance of a direct-sequence spread-spectrum low-Earth-orbiting satellite communications network is discussed. To recognize the effect of interference when their sources are either inside or outside the service area of a satellite, we develop a stochastic model for the location of users. We show that the effect of interference on the performance degradation from users with large propagation distance to their connecting satellites is the dominant factor. Hence, to improve the performance of the system, we propose a method in which the transmissions of packets are controlled according to their distances to connecting satellites as well as the traffic distribution.

In multipath fading channels, the direct-sequence spread spectrum (DS/SS) technique is often employed because of the capability to resolve signals from different paths and realize path diversity at the receiver. This advantage of DS/SS, however, is only when the time difference of propagation delay between paths is greater than a chip duration. In indoor communications, the delay spread is often very small and the above condition on the delay and chip duration is no longer satisfied, and this performance improvement by the path diversity with conventional DS/SS systems cannot be expected. In this paper we propose a new diversity technique for indoor wireless communication systems where delay spread is smaller than a chip duration. In this system, the transmitter sends signals at the same time from multiple antennas with intentional time delays, which allow a receiver to distinguish and combine the signals from different antennas.

We evaluate the bit-error rate performance of direct-sequence spread spectrum (DS/SS) system using /spl pi//4-shift QPSK and /spl pi//2-shift BPSK and observe the effect of the nonlinearity of the high power amplifier (HPA) of transmitters which occurs when the HPA is operated near its saturation region. The hard limiter is chosen as the example of the nonlinear amplifier and the numerical results are reported by using coherent as well as differential detection at the receiving end. With the employment of either /spl pi//4-shift QPSK or /spl pi//2-shift BPSK, the DS/SS signal is not much influenced by the nonlinearity of the HPA as the two modulation schemes have the advantage of nearly constant envelope.

In this paper, we present the throughput performance of CDMA slotted ALOHA systems with modified channel load sensing protocol (MCLSP). In MCLSP, a hub station continuously senses channel load for a certain number of slots, and computes the probability with which each user station should transmit its packet, then broadcasts the computed value. As the access control is based on this probability, we can apply it to slotted system. We show that MCLSP can achieve significant improvement in throughput performance of CDMA slotted ALOHA systems even in large offered load. One of the issues in such a protocol is the access timing delay, which is the sum of process time and propagation time for the packet access control. We also show the effect of the access timing delay on the throughput performance.

A transmit permission control method for improving the throughput characteristics of a low Earth orbit (LEO) satellite communication system employing spread-slotted ALOHA multiple-access scheme is proposed. Both nonfading and fading satellite links are considered. The basic idea of the proposed scheme is to decrease the level of interference at each satellite and, hence, to increase the probability of packet success, by prohibiting the packet transmission from the users with relatively high propagation loss to their connecting satellites. It is shown that the method has the ability to improve the throughput performance in heavy traffic loads and the peak value of the throughput, significantly. It is also shown that the average delay performance of the system employing the proposed scheme is superior to that of the conventional system at heavy traffic loads.

Throughput analysis of direct-sequence spread spectrum multiple access (DS/SSMA) unslotted ALOHA with fixed packet length is presented. As the levels of multi-user interference fluctuate during the packet transmission, we calculate the packet error probability and the throughput by considering not only the number of overlapped packets but also the amount of time overlap. On the assumption that packet generation is Poisson, the system can be thought as the queueing system M/D//spl infin/. With Gaussian approximation of multi-user interference, we obtain the throughput as the function of the number of chips in a bit, the packet length, and the offered load of the system. We also analyze the channel load sensing protocol (CLSP), and obtain the optimum threshold of CLSP.

We analyze a system which simultaneously transmits spread-spectrum signals with different processing gains. The main purpose of this study is to give an analytical framework that describes the influence of the interfering signals with different processing gains. The results show that the power of the interference component after the despreading procedure, the power ratio of the desired to undesired component, and the bit error rate are not constant but vary symbol by symbol. We also show that the processing gain of the interfering signal affects the performance.

We propose a coordinate-by-coordinate interleaved trellis coded QPSK with the T-algorithm. A rate 1/2 convolutional code with constraint lengths of 7 and 15 are evaluated on Rayleigh fading channels in terms of the bit error rate (BER) by an analysis and a computer simulation. As the effective code length (ECL) is equal to the code constraint length for the case of a QPSK system, the proposed system achieves a satisfactory BER performance even in the situation where no channel state information (CSI) is available at the decoder. Further, in order to reduce the decoding complexity, we employ the T-algorithm. It is shown that as the T-algorithm can track the fluctuating gains introduced by the fading channel, we can achieve a satisfactory BER performance with 0.01% of the decoding complexity of the Viterbi algorithm.

We present a new direct-sequence spread-spectrum demodulator using the block signal processing in the presence of large carrier frequency offset such as Doppler shift in low Earth-orbital satellites systems. The proposed demodulator employs the initial acquisition scheme which can roughly estimate both the timing and the carrier frequency offset. This scheme realizes the maximum likelihood estimation with the help of the FFT. As the result of the analytical discussion, it is shown that the proposed scheme makes the initial acquisition possible even in the presence of a large frequency offset where the conventional demodulator cannot achieve the acquisition.

In some recent studies, the use of low Earth-orbit satellites in various applications is considered. In all of these studies, uniform distribution of traffic load is assumed. In this paper, the performance of a low Earth-orbit satellite communication system which is designed to service to two kinds of users; i.e., voice users and data users is estimated. The distribution of population of these users is assumed to be nonuniform. According to the simulation results, it is shown that the nonuniformity in traffic affects the performance of the system by decreasing the signal quality at the satellite which has to service to populated areas and increasing it superfluously at the satellite with not so populated service area, in a given period of time. By modeling the satellites during their movements, the change in signal quality while experiencing a peak of traffic load in their route is also determined. A modified power control method based on the amount of traffic load of each satellite is also examined and is shown that this method can make some performance improvements in signal quality, which is limited by special features of low Earth-orbit satellite systems.>

This paper investigates the bit error rate (BER) performance of hybrid direct-sequence/slow-frequency-hopping spread-spectrum multiple-access (DS/SFH-SSMA) operating in the environment with the near-far problem. Rate 1/2 convolutional code of constraint length 7 is applied with an appropriate interleaver to increase the diversity effect not only on time domain but also on frequency domain. As a result, we found that the BER performance of DS/SFH system was always satisfactory in the presence of near-far problem. On the other hand, poor BER performance of the direct-sequence system is found regardless of E/sub b//N/sub 0/ or number of users.

This paper presents a novel spread slotted Aloha system with channel load sensing protocol (CLSP). CLSP is an effective scheme to improve the throughput performance in spread unslotted Aloha systems. In spread slotted Aloha systems, however, it does not make sense to utilize CLSP because the slot size is usually the same as the packet size. The slot size of the proposed system is set less than the packet size, thereby enabling the authors to apply CLSP and improving the throughput performance. Another feature of the proposed system is that the system is not likely affected by the time difference between channel load sensing and timing of packet access, which they call the access timing delay. Throughput performance of the proposed system is evaluated in the presence of the access timing delay and a significant increase of the throughput is shown compared with that of spread unslotted Aloha with CLSP.>

Digital transmission techniques are expected to spread throughout the fixed-satellite service in the near future. The introduction of digital satellite systems will lead to a new requirement regarding how to connect the satellite link with terrestrial digital networks. This paper presents a satellite/terrestrial digital interface that has been developed to interconnect 2.048 Mbit/s PCM terrestrial and INTELSAT TDMA links which are in the plesiochronous relationship. Major functions of the interface include frame slip control, compensation for path length variation, and rate and format conversions. This paper also describes the design and implementation of the developed equipment, and experimental results for signal impairment due to the frame slips and PCM multiplex standard conversion. The experimental results demonstrate that a large humber of bit errors are induced by frame slips for voiceband data transmission above 4800 bits/s.

A TDM/FDM conversion algorithm for realizing transmultiplexers with an FFT processor and a set of digital subfilters is proposed which provides a significant saving in multiplication rate. In the proposed algorithm in which an FDM signal is directly sampled, the dimension of the FFT processor is reduced by provision of pre- and postprocessing and the multiplication rate in radix-3 is reduced by refinement of the processing. In addition, the multiplication rate in the subfilters is reduced by the adoption of symmetrical coefficients. Although the direct sampling of an FDM signal has been considered to increase the multiplication rate compared with the sampling with frequency shift of FDM signal. The authors' estimation shows that the increase of multiplication rate in the proposed algorithm keeps within 20 percent compared with that in a typical algorithm based on sampling with frequency shift of FDM signal. Furthermore, it shows that the proposed algorithm reduces the multiplication rate by 30 percent compared with a previously proposed algorithm based on direct sampling of FDM signal. Based on the proposed algorithm, a digital 60 channel transmultiplexer has been implemented. System configuration of the developed equipment is outlined and its measured performance is described.

Recently, digital systems such as digital telephone switching systems and digital transmission paths are being introduced more and more into communications networks in which signals are handled with Time Division Multiplex (TDM). There still exists, on the other hand, a number of analog transmission paths in which signals are multiplexed with Frequency Division Multiplex (FDM). Therefore, it is important to develop a scheme to interconnect efficiently TDM and FDM systems. Transmultiplexer, which is an equipment directly converting PCM.TDM signals and SSB.FDM signals with each other, is expected to provide an efficient interconneetion between TDM and FDM systems. Authors have developed an experimental transmultiplexer which can provide mutual conversion between a 24-channel PCM primary group (1.544 Mbits/s) and two 12-channel FDM basic groups (60-108 kHz). The developed equipment is designed so as to achieve reduction of the computation rate and simplicity of the hardware by the following techniques: (1) The FFT (Fast Fourier Transform) technique is used to optimize the filtering and multiplexing processes. (2) The spectrum reversion of odd-numbered channel signals is achieved by inverting the sign bits of the PCM signals every other frame. (3) The bandpass sampling technique is applied to obtain SSB.FDM signals directly without frequency shifts at analog FDM stages. In order to evaluate the characteristics of the developed transmultiplexer, experiments have been conducted by connecting it with FDM terminal equipments. The experimental results showed that the transmultiplexer can offer better performance than the tandem connection of standard PCM multiplex equipments and standard FDM terminal equipments with reasonable hardware.