Computer Viruses – Analysis of Detection Techniques and their Limitations

Abstract

A Virus is basically developed to disturb the working of a computer and intentionally cause damage to its working. A virus can do so by file corruption, user document damage such as movie files, photographs, text documents, music files and in such a way to make the computer completely useless. A virus is not like a spyware developed for some mysterious behavior, it is developed to arise problems to use the computer and damage something. A computer virus replicates itself similar to a human virus and grows itself with the help of file sharing or email attachments. There are various detection techniques available and used by antivirus software to identify virus infected files and mark those files as infected and also try to clean them if possible. These techniques are successful to detect some specific category of viruses but may fail to detect other category of viruses. Some of the main categories of virus detection techniques are Static Signature Scanning Technique, Generic Signature Scanning Technique, Heuristic Analysis Technique and Integrity Checking Technique. But each of these techniques has its own limitations and the techniques fail to detect viruses with properties beyond the scope of the technique used. So, each antivirus software uses mixed set of techniques to detect virus infections in files. This paper discusses various types of virus threats and analyses various detection techniques in view of their limitations.