| Project/Area Number |
15500080
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Media informatics/Database
|
|---|
| Research Institution | Tokyo National College of Technology |
|---|
Principal Investigator |
AONO Masahiro Tokyo National College of Technology, Department of Computer Science, Professor, 情報工学科, 教授 (30342471)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
ICHIMURA Hiroshi Tokyo National College of Technology, Department of Computer Science, Professor, 情報工学科, 教授 (10176307)
KOJIMA Tetsuya Tokyo National College of Technology, Department of Computer Science, Assistance Professor, 情報工学科, 助教授 (20293136)
|
|---|
| Project Period (FY) |
2003 – 2005
|
| Project Status |
Completed (Fiscal Year 2005)
|
| Budget Amount *help |
¥2,400,000 (Direct Cost: ¥2,400,000)
Fiscal Year 2005: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2004: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 2003: ¥400,000 (Direct Cost: ¥400,000)
|
|---|
| Keywords | data broadcast / scheduling / band compression / Prediction of Band / streaming / カルーセル放送 / ストリーム型データ |
|---|
| Research Abstract |
In video compression, the necessary bandwidth differs depend upon characteristics of the scene. When an available bandwidth is constant like radio broadcasting, useless time zones are left, even if it performs buffering. This useless bandwidth is applicable to carousel-type data broadcasting, on demand broadcasting and so on. We researched the utilization method of this non-stream type data and the scheduling algorism. The goal of the research was to discover the optimal multi-media broadcast scheduling algorithm and to offer the practical use. We started research from these viewpoints. At first, we measured the maximum bandwidth and the average bandwidth with various videos, and observed how to occur the left band. We examined the relationship of the delay by buffering of useless band between the smoothing of necessary band, and found out the characteristic. Next, if carousel type data broadcasting was assumed as a usage of the unused band with stream type data broadcasting, we resear
… More
ched optimal broadcast scheduling to compare with the optimal scheduling when supplied band is flat. As a result, it is cleared that used scheduling method should be changed from the optimal scheduling when flat band size is supplied. In addition, there was an issue that a useless bandwidth was unclear how much it was compressed to when the video server output broadcast data. We examined two methods for this solution. One is the method of delaying only 1 measurement unit at the time of video transmission timing. The unused bandwidth is correctly measurable by this method. Another is the method of predicting the required bandwidth of the following timing from change of the past required bandwidth. When the method of former is used for unidirectional image output like sport relay, it could judge that the delay of 1 measurement time unit is not big problem. When the method of latter is used for the transmission of the video with a large fluctuation of required bandwidth, the estimate error of predicted bandwidth is large. However, when there is little change of bandwidth like a meeting, the prediction is comparatively accurate and suitable for practical use. As an applied example, I took up manipulation of multi-Media record for an example and examined a protect method of an application. Less
|
