2012 Fiscal Year Final Research Report
P2P Video Delivery Method and Its Evaluation on PlanetLab
| Project/Area Number |
22500055
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Computer system/Network
|
|---|
| Research Institution | Shiga University |
|---|
Principal Investigator |
MORI Masaaki 滋賀大学, 経済学部, 教授 (90024942)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SHIBATA Naoki 奈良先端科学技術大学院大学, 情報科学研究科, 准教授 (40335477)
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
YASUMOTO Keiichi 奈良先端科学技術大学院大学, 情報科学研究科, 教授 (40273396)
SUN Weihua 奈良先端科学技術大学院大学, 情報科学研究科, 助教 (40517520)
|
|---|
| Project Period (FY) |
2010 – 2012
|
| Keywords | ネットワーク / ビデオマルチキャスト / トランスコード / Planetlab / 3D画像 |
|---|
| Research Abstract |
In order to estimate P2P video delivery method in real environments, we proposed a method based on supervised learning, so-called the weighted k-nearest neighbor (WKNN) algorithm, for estimation of overlay link qualities from qualities observed in the past. Our experiments on PlanetLab showed that our method has good performance on path latency estimation, and there is no correlation between overlay links with close geographical similarity on bandwidth estimation.
|
