Channel characterization is essential to develop a reliable communication system. In this article, a time-domain cyclic correlation-based channel impulse response (CIR) estimation technique is proposed for frequency-domain pilot aided orthogonal frequency division multiplexing (OFDM) sounding systems. The proposed system provides an improved mean squared error (MSE) as compared to least square (LS) and minimum least square (MMSE) techniques. In this paper, periodic comb-type pilots are considered. The time-domain cyclic correlation results of the up-sampled frequency-domain pilot sequence contain multiple copies of CIR with a length equal to the sequence length. The aggregated response from various copies of CIR in one symbol results in a better CIR estimation. The proposed system provides better performance with very less computational complexity. The performance of the system is evaluated in the Rayleigh channel. The results show that the proposed method can be a better alternative to LS and MMSE technique for OFDM sounding systems. © 2019 IEEE.