INTRUSION DETECTION AND PREVENTION FRAMEWORK IN A HONEYPOT CLOUD NETWORK USING HIDDEN MARKOV MODEL

Abstract

With the fast expansion in the quantity of users, there is an ascent in issues connected with hardware failure, web hosting, space and memory allocation of data, which is straight forwardly or in a roundabout way prompting the deficiency of data. With the goal of offering services that are reliable, quick and low in cost, Cloud computing is the most appropriate and suitable environment. With a gigantic improvement in this innovation, steadily expanding chance of its security is being undermined by Honeypot. A method for redirecting vindictive traffic from frameworks is by utilizing Honeypot. An enormous methodology has given indications of progress in security of frameworks. Remembering the different legitimate issues one might look while sending Honeypot on third-party cloud vendor servers, the idea of Honeypot is carried out in a record sharing application which is conveyed on cloud server. This paper examines the detection attacks in a cloud-based environment as well as the utilization of Honeypot for its security, subsequently proposing a Intrusion detection framework design utilizing Hidden Markov Model Algorithm. This framework considers each stage utilized by ongoing intrusion and applies them to the Hidden Markov Model algorithm to figure out which intrusion is utilized in the audit data. This architecture diminishes overheads of intrusion agents and raises efficiency of the entire framework.

Keywords: Honeypot, Hidden Markov Model (HMM), Intrusion Detection System (IDS), Intrusion Prevention System (IPS), MD5.