1990 Fiscal Year Final Research Report Summary
Statistical Properties of Noise
| Project/Area Number |
01550258
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
電子通信系統工学
|
|---|
| Research Institution | University of Electro-Communications |
|---|
Principal Investigator |
MIMAKI Tadashi University of ElectroーCommunications, Depart. of Communication Systems, Professor, 電気通信学部・電子情報学科, 教授 (60017358)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MUNAKATA Tsutomu Tamagawa University, Depart. of Electronics, Lecturer, 工学部・電子工学科, 講師 (20190861)
SATO Hirosi University of ElectroーCommunications, Depart. of Information Engineering, Profes, 電気通信学部・情報工学科, 教授 (00017279)
|
|---|
| Project Period (FY) |
1988 – 1990
|
| Keywords | Noise / Level-Crossing / First passage time / Gaussian process / Rayleigh process / Butterworth spectrum |
|---|
| Research Abstract |
a. First Passage Time Problem : The first passage time is composed of two kinds : The first one is the directーpassage time, wherein after a random process has crossed the first level it arrives the barrier without any crossing with the first level ; and the second one is the usual passage time, whereinafter a random process has crossed the first level it arrives the target barrier with aribitrary number of crossings with the first level therebetween. The properties of these two kinds of first passage times are experimentally studied for Rayleigh and Gaussian random processes.
b. Level-Crossing Problem of The Phase of Sinewave Plus Noise : Different from the level-crossing problem of the amplitude of a random process, the up-crossing and down-crossing does not necessarily occur alternately. The problem is investigate for Gaussian processes having Gaussian and Butterworth power spectrum densities.
c. Estimation of Auto-correlation Function : A new method is derived to estimate the auto-correlation function of a random process by using four Rice functions, which are numerically calculated or experimentally determined.
d. CurveーCrossing Problem of Gaussian Process : If the barrier curve is properly selected for the curveーcrossing problem, the Rice function does not change for a group of the curves. Using this fact, the estimation of the autoーcorrelation function of a random process is derived.
|
Research Products
(18 results)
