Indian Journal of Science and Technology

Abstract

Objectives: Authenticating the MobiCloud users using traditional passwords, biometrics is prone to shoulder surfing attack and need external device. Pattern Based User Authentication (PBUA) mechanism is introduced to identify and authenticate the users in MobiCloud environment. Methods/ Statistical Analysis: To authenticate the users, the whole process is divided into two phases namely, identification and authentication.Dynamic patterns are generated using random mathematical function to select the patterns from the Identity Management (IdM) Server. Pattern matching operation is done using the hash-based technique to match the image pattern sequences quickly with the actual pattern. Findings: Experimental results show that the proposed PBUA is strong enough against shoulder surfing attack. The time taken for pattern matching is very less when compared to existing pattern matching mechanisms. Application/Improvement: PBUA does not need any intermediate table as the patterns are matched directly with original patterns. The complexity of the algorithm is reduced by , thus the efficiency is improved n2 times. Pattern updation is not required which reduces the user’s burden. 
Keywords: Authentication, Identity Management, Identification, Mobile Cloud, Pattern Matching