This paper presents a Wavelet based alienation technique to detect and classify various faults on transmission line. The proposed scheme analyses the absolute values of three phase current signals over a half cycle to obtain detail coefficients. These detail coefficients of half a cycle are compared with those of previous half cycleto compute alienation coefficients which are further utilized to detect and classify the faults. The proposed technique was able to discriminate non-fault transients such as capacitance, inductance and load switching, from fault transients. The increase in the sensitivity of protection scheme, due to utilization of wavelet based detail decomposition, has been established by case studies. The proposed algorithm is tested for different locations and various types of faults. The algorithm is proved to be successful in detecting and classifying various types of faults in a half cycle. © 2014 IEEE.