Showing papers by "Beijing University of Posts and Telecommunications published in 1993"
TL;DR: In this article, a letter by Agbinya misused the method of interpolation using the type I discrete cosine transform (DCT-I), and did not give a correct interpolation scheme using the conventional DCT.
Abstract: A letter by Agbinya misused the method of interpolation using the type I discrete cosine transform (DCT-I), and did not give a correct interpolation scheme using the conventional discrete cosine transform (DCT). The correct procedure of interpolation using the DCT-I is clarified, and an appropriate approach of interpolation using the DCT is given in the Letter.
21 citations
TL;DR: In this article, the exact expression of the bit error rate for optical soliton communication systems with lumped amplifiers by considering the combined effect of timing jitter and energy fluctuation was proposed.
Abstract: We propose the exact expression of the bit-error rate for optical soliton communication systems with lumped amplifiers by considering the combined effect of timing jitter and energy fluctuation. Some numerical examples are given, and the dependence of the optimal fiber dispersion on the amplification period and the transmission distance is discussed.
14 citations
TL;DR: Forty-four NH(3) absorption spectral lines of the combination tones of the vibration-rotation transitions were observed at 1.4988-1.5520 microm with a 1.5-microm grating-external-cavity semiconductor laser to measure linewidth and absorption rate.
Abstract: Forty-four NH(3) absorption spectral lines of the combination tones of the vibration-rotation transitions were observed at 1.4988-1.5520 microm with a 1.5-microm grating-external-cavity semiconductor laser. The linewidth and absorption rate of NH(3) absorption spectral lines were measured. Also, the changes of center frequency and linewidth of the NH(3) absorption spectral line with the our knowledge the first time.
9 citations
TL;DR: Using complex numbers, a special case of a spatial 7R mechanism is discussed in this article, where the computer calculations show that the input-output equation of this mechanism is still 16th degree in tan-half-angle of the output angular displacement.
9 citations
TL;DR: A two-dimensional (2-D) search flow scheme is proposed for the incremental backprojection algorithm and it is found that back projection of four views can be performed at the same time, and the search operation is only required for one of them if the number of views is even.
Abstract: A two-dimensional (2-D) search flow scheme is proposed for the incremental backprojection algorithm. In this scheme, the searching direction in a beam is determined only by the distance from the current pixel to the corresponding ray; no pixels outside the beam are involved in the procedure. In addition, the geometrical relationships between the image grids and the projection rays in different views are fully exploited to further reduce the search operations. It is found that backprojection of four views can be performed at the same time, and the search operation is only required for one of them if the number of views is even. Thus, backprojection is accomplished on a quartet-by-quartet of views basis as well as on beam-by-beam basis. These approaches not only minimize the number of search operations but also possess a simple algorithm structure. Implementation results obtained with an AST-386 computer for the incremental backprojection algorithm using this search flow scheme and the relationship among projection views show that the processing time can be reduced by a factor of about 2. >
9 citations
03 May 1993
TL;DR: Analysis and simulations show that the momentum misadjustment is a nonnegligible factor affecting the convergence precision of the BP network and an improved algorithm with stochastic attenuation momentum factor /splalpha/(k) can effectively cancel the negative effect of /spl alpha/ on a network and improve the convergence performance of the network.
Abstract: The authors analyze the misadjustment of the conventional backpropagation (BP) network introduced by the momentum factor /spl alpha/ and its disadvantageous effect on the network, and propose an improved algorithm with stochastic attenuation momentum factor /spl alpha/(k). Analysis and simulations show that the momentum misadjustment is a nonnegligible factor affecting the convergence precision of the BP network. Compared with the conventional BP algorithm, the improved algorithm can effectively cancel the negative effect of /spl alpha/ on a network and improve the convergence performance of the network. >
7 citations
TL;DR: Based on the geometrical relationships between image pixels and projection rays at different viewing angles, a multi-orientation simultaneous backprojection approach is proposed in this article. But it is not suitable for multi-view applications.
Abstract: Based on the geometrical relationships between image pixels and projection rays at different viewing angles, a multi-orientation simultaneous backprojection approach is proposed. Backprojections for four views theta , pi /2- theta , pi /2+ theta and pi - theta are performed at the same time; the locating operations of pixels for interpolation are only required for one of them when the number of views is even. By using this method in the Shepp and Logan backprojection algorithm, implementation results on an Intel 80386-based computer show that the processing time is reduced by a factor of approximately 2.7 for linear interpolation and approximately 4.2 for nearest-neighbour interpolation. >
5 citations
TL;DR: In this article, the authors described the research work on coherent optical fiber communications in China and described several 1.55 μm systems with various modulation schemes such as FSK, DPSK, and MSK.
Abstract: Research work on coherent optical fiber communications in China is described. Experimental achievements both in basic components and systems have been made since 1981. Several 1.55 μm systems with various modulation schemes such as FSK, DPSK, and MSK have been demonstrated at 140-154 Mbit/s. Two-channel FDM transmissions have been realized. A field test of 622 Mbit/s coherent transmission system and other applications are under planning.
5 citations
TL;DR: The significant activity that has taken place in China on source coding is reviewed, including three of the most important source coding techniques used for speech compression-waveform coders, vocoders, and hybrid coders-and a number of techniques used to implement image compression.
Abstract: The significant activity that has taken place in China on source coding is reviewed. This includes three of the most important source coding techniques used for speech compression-waveform coders, vocoders, and hybrid coders-and a number of techniques used to implement image compression. Two other areas of active communications research, channel coding and cryptography, are discussed. >
4 citations
19 Oct 1993
TL;DR: It is shown that the switch with LQFS policy has better performance with smaller memory compared with the one with round-robin policy.
Abstract: Crossbar switches have always been attractive to switch designers because they are internally nonblocking and they are simple For switching in the ATM mode, this fabric has many advantages and the main drawback is that the total memory required is large In order to reduce the large buffer memory the longest queue first service (LQFS) policy is used The performance of internal buffered crossbar-based ATM switches with LQFS policy is studied It is shown that the switch with LQFS policy has better performance with smaller memory compared with the one with round-robin policy >
3 citations
19 Oct 1993
TL;DR: The paper presents a novel Viterbi decoder hardware implementation, with the code rate R=1/2 and the constraint length K=5, for the GSM land mobile system with the use of one XILINX FPGA 3090 chip to replace the ACSs' function.
Abstract: The paper presents a novel Viterbi decoder hardware implementation, with the code rate R=1/2 and the constraint length K=5, for the GSM land mobile system. With the use of one XILINX FPGA 3090 chip to replace the ACSs' function, which results in a much simpler hardware and routing structure, this system makes the final ASIC fabrication a very direct consequence. >
TL;DR: A chain vector-quantization clustering clustering (CVQC) algorithm for realtime speech recognition that delivers faster training and recognition speeds and requires smaller memory locations.
TL;DR: The period distributions of R-S codes, extended R- S codes, and cyclic codes, in general, will be calculated in this paper.
Abstract: The period distribution of an error-correcting code tells how many codewords in the code have a specific minimum period, where the minimum period of a codeword is the minimum number of cyclic shifts necessary to obtain the same codeword. The period distributions of R-S codes, extended R-S codes, and cyclic codes, in general, will be calculated in this paper. The period di stribution problem for the other noncyclic codes still remains unknown.
TL;DR: The effects of the initial chirp of the control soliton pulse on the time shift of the soliton-dragging logic gate (SDLG) is investigated numerically and the physical interpretation for this phenomenon is presented.
Abstract: The effects of the initial chirp of the control soliton pulse on the time shift of the soliton-dragging logic gate (SDLG) is investigated numerically in this article. We find that the positive initial chirp increases the time shift and the negative initial chirp decreases the time shift. The physical interpretation for this phenomenon is presented. © 1993 John Wiley & Sons, Inc.
TL;DR: In this paper, a simple analytical expression for the optimal fiber dispersion which yields minimum bit error rate was proposed for an optical soliton communication system with lumped amplifiers, and the maximum rate-distance product of the system limited by the energy fluctuation and the timing jitter resulting from accumulated ASE noise of the amplifiers was given.
17 Jan 1993
TL;DR: The PSP sequence with the smallest density is found and is solved by a few new binary sequences with perfect staircase profile of linear complexity.
19 Oct 1993
TL;DR: Based on neural computing theory, an algorithm to solve the Steiner Tree Problem, an NP-complete problem, is proposed that is parallel and its efficiency is verified through examples.
Abstract: Based on neural computing theory, we propose an algorithm to solve the Steiner Tree Problem, an NP-complete problem. The algorithm proposed is parallel and its efficiency is verified through examples. However, two problems remain unsolved: the determination of parameters in the algorithm; and the optimal solution cannot always be found. >
19 Oct 1993
TL;DR: The paper proposes a dynamic recognition neural network model (DRNN), which has the advantages of some neural networks, so it is learns fast, calculates simply, and constructs flexibly, suitable for dynamic recognition, such as speech recognition.
Abstract: The paper proposes a dynamic recognition neural network model (DRNN), which has the advantages of some neural networks, so it is learns fast, calculates simply, and constructs flexibly The model is suitable for dynamic recognition, such as speech recognition Experiments on the Chinese isolated digital recognition were carried out The results show that the model has potential >
TL;DR: A computer-aided planning tool for local telephone networks (CAPTLN), composed of three software modules that provide traffic forecasting, network optimization, and network economic analysis for the local network of the public switched telephone network (PSTN), is discussed.
Abstract: The methods used in local network planning for the public telecommunications network in China are discussed. Network modeling, computer simulation, and planning software used in the planning process are described. A computer-aided planning tool for local telephone networks (CAPTLN), composed of three software modules that provide traffic forecasting, network optimization, and network economic analysis for the local network of the public switched telephone network (PSTN), is also discussed. Other tools are similarly used for all other parts of the PSTN network, such as international access, toll switching and trunking, signaling and synchronization. >
TL;DR: Research and development in China on optical fiber communications systems, including traditional high bit rate, coherent, and optical soliton systems is described and their potential for telecommunications systems applications is outlined.
Abstract: Research and development in China on optical fiber communications systems, including traditional high bit rate, coherent, and optical soliton systems is described. Their potential for telecommunications systems applications are outlined. >
19 Oct 1993
TL;DR: The authors propose neural networks architecture for two kinds of queueing in packet switch networks to gain maximum throughput by introducing a new energy function for the input queueing and a modified input smoothing scheme, which can obtain higher throughput than the standard input smoothed scheme.
Abstract: Packet switching is an important technology in broadband ISDN (B-ISDN). The authors propose neural networks architecture for two kinds of queueing in packet switch networks to gain maximum throughput. First they introduce a new energy function for the input queueing. Then they propose a modified input smoothing scheme, which can obtain higher throughput than the standard input smoothing scheme, and its control by neural network. Finally, computer simulation results are given. >
19 Oct 1993
TL;DR: This new DCME frame structure is described and it is shown that NEC'S NCM-501 DCME fully complies with INTELSAT IESS-501, DCME multiframe length is is longer than newDCME frame length.
Abstract: Digital circuit multiplication equipment (DCME) permits the concentration of a number of 64kbit/s PCM encoded input trunk channels on a reduced number of transmission channels It has two basic techniques; one is digital speech interpolation (DSI), the other is adaptive differential pulse code modulation (ADPCM) Essential facilities, interface conditions and overall performance requirements are given by recommendation G763, 1988 However, frame structure didn't detail assignment Many corporations such as ECI, NEC and AT and T have produced products They have used different DCME frame structures NEC'S NCM-501 DCME fully complies with INTELSAT IESS-501, DCME multiframe length is is longer than new DCME frame length This new DCME frame structure is described >
23 May 1993
TL;DR: A model is proposed for PRMA systems with channel fading by the Markov analysis method, and the numerical analysis shows close agreement with computer simulation.
Abstract: Packet reservation multiple access (PRMA) performance is evaluated in terms of dropping probability when the transmission channel is subject to fading. A model is proposed for PRMA systems with channel fading by the Markov analysis method. The numerical analysis shows close agreement with computer simulation. For a particular example, when the header error rate is less than 1%, 37 speech terminals can share a PRMA system channel with 20 slots per frame, within the 1% packet dropping limit. >
19 Oct 1993
TL;DR: The Hopfield/Tank model used for the communication network routing problem is described and it is found that the model has the fastest rate to converge to the optimal or near-optimal solutions from an initial value.
Abstract: A description of the Hopfield/Tank model used for the communication network routing problem is given. Then, through "static" (i.e. the eigenvalue and the eigenspace of the connection matrix) and the dynamic analyses of the model, it is found that the model has the fastest rate to converge to the optimal or near-optimal solutions from an initial value. Therefore the quality of the final solutions can be guaranteed. The influence of the initial value on the final solutions is also considered. Simulation results are given at the end of the paper. >
19 Oct 1993
TL;DR: This work uses discrete the hidden Markov model with noise-adaptive codebook for noisy speech recognition to improve the recognition accuracy of recognizer in a noisy environment.
Abstract: At present, many researchers turn their attention to automatic speech recognition in noisy environments. The main reason is that speech recognizers trained in quiet conditions but operated in a noisy environment usually have poor performance. We use discrete the hidden Markov model with noise-adaptive codebook for noisy speech recognition. The goal is to improve the recognition accuracy of recognizer in a noisy environment. When testing with noise-contaminated utterances at an SNR of 20 dB, the system has a recognition accuracy of 35%, by using the noise-adaptive codebook, the system has an accuracy of up to 90%. >
TL;DR: In this paper, a method of designing an optical soliton communication system with lumped amplifiers is presented, and an exact bit-error-rate expression of the system is derived.
Abstract: A method of designing an optical soliton communication system with lumped amplifiers is presented in this article. The design requirements and the design procedure are discussed. An exact bit-error-rate expression of the system is derived. Some design examples of the optimal systems are given. © 1993 John Wiley & sons, Inc.
19 Oct 1993
TL;DR: In this article, the problem of estimating the parameters of a moving average model from the cumulant statistics of the noise observations of the system output is addressed by setting up the relation between the bispectrum and tricepstrum.
Abstract: Addresses the problem of estimating the parameters of a moving average model from the cumulant statistics of the noise observations of the system output. The authors set up the relation between the bispectrum and tricepstrum. Using this equation, a simple closed form solution for FIR system identification can be obtained. The approach permits that the measurement noise sequence is colored Gaussian. >
TL;DR: In this article, a commenter pointed out that the nonexistence of the 4*3*3 *3*9 perfect binary array (PBA) can be easily proved, without the need for a lengthy computer search.
Abstract: For original papers see ibid., vol.29, p.99-101 (1993) and ibid., vol.27, p.1252-1254 (1991). The commenter suggests that the nonexistence of the 4*3*3*9 perfect binary array (PBA) can be easily proved, without the need for a lengthy computer search. He points out that the nonexistence of the 9*36 PBA and 18*18 PBA can also be proved. In reply, the original authors suggest that the commenter's analysis is based on an incorrect lemma.
16 Nov 1993
TL;DR: The probability distribution of random signals at circuit nodes is analyzed, and it is proved that controllability is in normal distribution, and the unbiasing estimate of fault detection probability can be obtained.
Abstract: STAFAN algorithm (1985) is regarded as a good alternative to fault simulation of digital circuits. However, this algorithm has two parameters /spl alpha/ and /spl beta/ to be determined by fault simulation and so it is of no practical uses. This article analyzes the probability distribution of random signals at circuit nodes, and proves that controllability is in normal distribution. Thus the unbiasing estimate of fault detection probability can be obtained. Moreover, according to the concept of observability, we can eliminate parameter /spl beta/. For actual circuits the fault coverage obtained from the modified STAFAN agrees favorably with the fault simulation results. >