Abstarct: Individual signature is one of the identification procedures, especially in the field of economic relations. The task of a signature verification system is to accept the original signature and reject the false ones. Online signatures are the ones that are recorded by electronic devices, such as digitized page, and saved in the computer in the form of time series. In this thesis, pre-processing such as normalization, removing the rotation, smoothing, and etc. has been applied on the original and false signatures of SVC2004 databaxse, and 72 features has been extracted from them. Then by utilizing PSO optimized algorithm, original signature is verified from the false ones. First by utilizing 72 features, Equal Error Rate of verification system of 11.5% is achieved. Removing unimportant features by PSO algorithm, and utilizing more important features equal to 48, results to EER=10.25%. It is noteworthy that, this results compared with the results of participant teams in the first international signature verification competition, which in the best case, got the 5th rank.