I know that different attacks would have their own-unique- pattern, unless for some similarities which could be common among attacks of same category-this is based on the classification and/or grouping of attacks based on certain criteria.

But I'm looking for the way or approach to get that patterns.

Based of log data , which is huge in volume variety and velocity , the patterns can be found , and can be employed fr attacks

Take a look at CAPEC from MITRE

http://capec.mitre.org/data/graphs/1000.html

Robin Gandhi but these are the patterns of the attacks. However,I have visited the site before. But I'm looking for the analysis processes/methods to get these patterns out of the attacks and the attackers.

Thanks anyway, appreciate it

Hi Sergio,

I can understand that my focus should be directed to DOS attack patterns, as no specific ones for other attacks!!!

Buddy your question was not narrowed down there are zillions of types of attacks like some attacks are on the architecture, some are external. There are different intents, motivations and different vulnerabilities work on them by exploiting different parts of your network/hardware. Which part are you discussing here?

Fully depends on the nature of your attack.WEB attack will be totally different then attacking a proprietary based communication protocol.

basically you have to thoroughly analyse your target

and fully agreed with @abdus saboor

We have developed an approach to automatically generate attack scenarios to Test security tools (mainly the IDS). Despite the approach considers generic attacks, it may help. The following papers explain the approach:

1- https://www.researchgate.net/publication/264669837_A_model-driven_approach_for_experimental_evaluation_of_intrusion_detection_systems

2- https://www.researchgate.net/publication/4360194_Execution_Patterns_in_Automatic_Malware_and_Human-Centric_Attacks?ev=prf_pub

Article A model-driven approach for experimental evaluation of intru...

Conference Paper Execution Patterns in Automatic Malware and Human-Centric Attacks