Autism Spectrum disorder (ASD) is one of the fastest growing developmental disorder worldwide. Recently, the number of people diagnosed with ASD has increased Tremendously. ASD should be diagnosed as early as possible to provide effective intervention or atleast a way to live with. Even though many neuro-imaging and diagnostic techniques are available, no such technique is accurate or practically implemented by medical practitioners. Thus, there is urgency for the development of an accurate and practically implementable autism diagnosis system. Unique EEG patterns are widely used to differentiate autistic and normal children. This study focuses on various aspects of EEG data processing which will assist the development of an automated autism diagnosis system.