| JaLCDOI | 10.18926/15833 |
|---|---|
| FullText URL | Mem_Sch_Eng_OU_17_2_31.pdf |
| Author | Taniguchi Takeo| Ikuta Shusaku| |
| Abstract | New estimation method of the optimum relaxation factor for the successive overrelaxation method (SOR) is proposed, and the efficiency of the new method is surveyed through a number of numerical experiments. This method can a priori determine the value of the factor by using only the topological properties of the problem, and it is valid for a sparse set of linear equations obtained by using the five-point difference scheme for any rectangular area with arbitrary boundary conditions. The experiments clarify that the method can estimate good approximate value of the factor. |
| Publication Title | Memoirs of the School of Engineering, Okayama University |
| Published Date | 1983-06-30 |
| Volume | volume17 |
| Issue | issue2 |
| Start Page | 31 |
| End Page | 46 |
| ISSN | 0475-0071 |
| language | English |
| File Version | publisher |
| NAID | 120002307996 |
| JaLCDOI | 10.18926/15828 |
|---|---|
| FullText URL | Mem_Sch_Eng_OU_17_2_15.pdf |
| Author | Tanada Yoshihiro| Sano Hiroya| Xu Hai| |
| Abstract | Optical-fiber multi-level PCM transmission system is investigated on the transmission distance and bit rate. A communication channel is established as follows: a light signal is modulated in intensity with a completely-balanced M-Ievel code at a transmitter, and is propagated over an optical glass fiber, and is demodulated at a receiver where the signal with an additive noise is processed through a PD, an equalizing amplifier, a matched filter, (M-l) comparators and a decision circuit. The relative power of the noise from the amplifier is increased in accordance with the reciprocal of fiber's transmittance, where shot noise and thermal noise are smoothed and decreased in power by the matched filter. The relation between the BER and the SNR leads the transmission distance. The product of the transmission distance and bit rate takes the large maximum at the large values of SNR and M. The large value of M is suitable for low-speed and high-rate system, and the interference system. |
| Publication Title | Memoirs of the School of Engineering, Okayama University |
| Published Date | 1983-06-30 |
| Volume | volume17 |
| Issue | issue2 |
| Start Page | 15 |
| End Page | 29 |
| ISSN | 0475-0071 |
| language | English |
| File Version | publisher |
| NAID | 120002307714 |
| JaLCDOI | 10.18926/15826 |
|---|---|
| FullText URL | Mem_Sch_Eng_OU_17_2_1.pdf |
| Author | Nakanishi Sen-ichiro| Katsuyama Yoshiaki| Himei Toyoji| |
| Abstract | The paper presents the computer aided design (CAD) method, the program, the design and the experimental results of inverse parallel thyristor phase control circuits. The calculated values agree well with the measured. The CAD program contains the next two methods which are inquired carefully by authors, such as (i) the optimization by SUMT (Sequential Unconstrained Minimization Technique) method, and (ii) the combined use of the gradient and the cramp calculation methods. |
| Publication Title | Memoirs of the School of Engineering, Okayama University |
| Published Date | 1983-06-30 |
| Volume | volume17 |
| Issue | issue2 |
| Start Page | 1 |
| End Page | 14 |
| ISSN | 0475-0071 |
| language | English |
| File Version | publisher |
| NAID | 120002307323 |
