Presented at BlackHat\n
Host-based Intrusion Detection Systems traditionally compare observable data to pre-constructed models of normal behavior. Such models can either be automatically learnt during a training session, or manually written by the user. Alas, the former technique suffers from false positives, and therefore repeatedly requires user intervention, while the latter technique is tedious and demanding.\n
In this talk we discuss how static analysis can be used to automatically construct a model of application behavior. We call this methodology \"Code-based Intrusion Detection\", and show that its derived model can prevent future or unknown code injection attacks (such as Buffer Overflows) with guaranteed zero false alarms. We will then present Korset, a Linux prototype that implements this approach with negligible runtime overheads.\n
\nOhad Ben-Cohen is a Linux Kernel developer and consultant, bringing years of Information Security expertise and Free and Open Source Software know how. His recent Open Source work includes writing Texas Instruments' Bluelink Linux driver, Bluetooth power management support for the kernel and the Linux port of TI's FM stack. He teaches System Programming at Tel-Aviv University, where he conducts his research and develops Korset. \n
In collaboration with Avishai Wool, Tel-Aviv University\n
\n\n\n\n\n\n"},"langlinks":[],"categories":[],"links":[],"templates":[],"images":[],"externallinks":[],"sections":[{"toclevel":1,"level":"2","line":"Korset: Code-based Intrusion Detection System for Linux, Ohad Ben-Cohen","number":"1","index":"1","fromtitle":"OWASP_Israel_2008_Conference_Ohad_Ben_Cohen","byteoffset":0,"anchor":"Korset:_Code-based_Intrusion_Detection_System_for_Linux.2C_Ohad_Ben-Cohen"},{"toclevel":1,"level":"2","line":"Bio","number":"2","index":"2","fromtitle":"OWASP_Israel_2008_Conference_Ohad_Ben_Cohen","byteoffset":930,"anchor":"Bio"}],"displaytitle":"OWASP Israel 2008 Conference Ohad Ben Cohen","iwlinks":[],"properties":[]}}