| Presentation | 2017-08-24 A DDoS Attack Detection Based on Group Sparsities of OD Flow Matrices Masaya Endo, Masao Yamagishi, Isao Yamada, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | To realize effective detection schemes applicable to diverse network traffic anomalies, we first propose to use group sparsities of the {em extended OD flow matrix}, expressing time transitions of traffic matrices, for flexible characterizations of major pathological behaviors of network traffic. We also formulate an approximate decomposition problem of an extended OD flow matrix into a group sparse and a low-rank matrix, and show that this problem, {em generalized robust principal component analysis for anormaly detection}, can be solved by efficiently with proximal splitting methods. Moreover, we present a novel characterization of typical DDoS (Distributed Denial of Service) attacks with a certain sparsity of OD flows after being assigned to groups of the same destinations. The proposed characterization is combined with the generalized robust principal component analysis to establish a DDoS attack detection scheme. Numerical experiments demonstrate effectiveness of the proposed DDoS attack detection scheme. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | DDoS attack detection / Extended OD flow matrix / Group Sparsitiy / Generalized robust principal component analysis |
| Paper # | SIP2017-52 |
| Date of Issue | 2017-08-17 (SIP) |
| Conference Information | |
| Committee | SIP |
|---|---|
| Conference Date | 2017/8/24(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Tokyo Denki University |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Masahiro Okuda(Univ. of Kitakyushu) |
| Vice Chair | Shogo Muramatsu(Niigata Univ.) / Naoyuki Aikawa(TUS) |
| Secretary | Shogo Muramatsu(Chiba Inst. of Tech.) / Naoyuki Aikawa(Takushoku Univ.) |
| Assistant | Masayoshi Nakamoto(Hiroshima Univ.ひろ) |
| Paper Information | |
| Registration To | Technical Committee on Signal Processing |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A DDoS Attack Detection Based on Group Sparsities of OD Flow Matrices |
| Sub Title (in English) | |
| Keyword(1) | DDoS attack detection |
| Keyword(2) | Extended OD flow matrix |
| Keyword(3) | Group Sparsitiy |
| Keyword(4) | Generalized robust principal component analysis |
| 1st Author's Name | Masaya Endo |
| 1st Author's Affiliation | Tokyo Institute of Technology(Tokyo Inst. of Tech.) |
| 2nd Author's Name | Masao Yamagishi |
| 2nd Author's Affiliation | Tokyo Institute of Technology(Tokyo Inst. of Tech.) |
| 3rd Author's Name | Isao Yamada |
| 3rd Author's Affiliation | Tokyo Institute of Technology(Tokyo Inst. of Tech.) |
| Date | 2017-08-24 |
| Paper # | SIP2017-52 |
| Volume (vol) | vol.117 |
| Number (no) | SIP-180 |
| Page | pp.pp.21-26(SIP), |
| #Pages | 6 |
| Date of Issue | 2017-08-17 (SIP) |