Biometrics provides a better solution for increasing security requirements and privacy protection than traditional recognition methods such as passwords and PINs. Figure 2.1 shows a biometric system that uses a single biometric trait to establish identity is known as unibiometric system. Various limitations imposed by such biometric systems are noisy data, non-universality, intra-class variations, inter-class similarities and spoof attacks. These limitations can be alleviated by fusing the information presented by multiple sources. A system that consolidates the evidence presented by multiple biometric sources is known as a multibiometric system. Figure 2.2 shows a typical multibiometric system. Integration of evidences is called information fusion. If fusion is appropriately done then it can enhance the matching accuracy of a recognition system. These systems are also expected to be more reliable due to the availability of multiple pieces of evidence. Combining multiple sources of information into a system, improves matching performance, increases population coverage and deters spoofing activities.Mere using multiple biometrics does not imply better system performance rather degrades the performance of individual biometric traits when used in poorly designed system [HON, 1999].