Our system is currently under heavy load due to increased usage. We're actively working on upgrades to improve performance. Thank you for your patience.
2003
DOI: 10.1007/978-3-540-39927-8_29
|View full text |Cite
|
Sign up to set email alerts
|

A Study on the Covert Channel Detection of TCP/IP Header Using Support Vector Machine

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
39
0
1

Year Published

2005
2005
2016
2016

Publication Types

Select...
6
3
1

Relationship

0
10

Authors

Journals

citations
Cited by 52 publications
(41 citation statements)
references
References 6 publications
0
39
0
1
Order By: Relevance
“…Classifier models are then trained based on provided examples of features of covert channels and normal traffic. Sohn et al demonstrated that simple covert channels encoded in the IP ID or TCP ISN field can be discovered with high accuracy by Support Vector Machines (SVMs) [Sohn et al 2003]. Tumoian et al showed that a neural network can detect Rutkowska's TCP ISN covert channel [Rutkowska 2004] with high accuracy [Tumoian and Anikeev 2005] (both Random Value pattern).…”
Section: Countermeasures For Patternsmentioning
confidence: 99%
“…Classifier models are then trained based on provided examples of features of covert channels and normal traffic. Sohn et al demonstrated that simple covert channels encoded in the IP ID or TCP ISN field can be discovered with high accuracy by Support Vector Machines (SVMs) [Sohn et al 2003]. Tumoian et al showed that a neural network can detect Rutkowska's TCP ISN covert channel [Rutkowska 2004] with high accuracy [Tumoian and Anikeev 2005] (both Random Value pattern).…”
Section: Countermeasures For Patternsmentioning
confidence: 99%
“…This means that we do not handle cases where, for example, a single key is sent in one data package. In this way, the proposed technique is similar to other techniques which use machine learning techniques, such as those that employ neural networks (e.g., Tumoian and Anikeev 2005) or support vector machines (e.g., Sohn et al 2003), to analyse data streams for anomalies indicating the possible existence of a covert channel. Since machine learning techniques require training data, it is acknowledged in Tumoian and Anikeev (2005) that it is impossible to discover a covert channel using a single data package and that the more packets that are recorded, the more precise the technique will be.…”
Section: Discussionmentioning
confidence: 93%
“…Several covert channels can be eliminate by blocking protocols/ports by firewalls (Loki [22,23], ICMPTX [113], Skeeve [128], ICMP-Chat [84]) or ingress/egress filtering (B0CK [120], Skeeve [128], false IPv6 Source Address Covert−TCP [98] can be detected using a Support Vector Machine (SVM) [109], and together with NUSHU [99] by [85] anomaly tests, because covert headers are easily distinguished from those generated by a genuine TCP/IP stack.…”
Section: Defence Mechanismsmentioning
confidence: 99%