The third generation partnership project has employed SCFDMA for its uplink transmission because of its low PAPR. SCFDMA signal while travelling through channel is affected by noise contained in the channel. Various channel estimation techniques has been given in the literature. This paper presents a channel estimation based on LMS with its parameters being optimised using PSO. The paper also compares result with existing LMS algorithm based systems. It has been observed that the proposed technique provides improvement in Bit Error Rate as compared to other technique.