In this paper, we present a novel signal to noise ratio (SNR) estimation technique using Constant Amplitude Zero Autocorrelation (CAZAC) time synchronization preamble in cooperative communication system. More specifically, a SNR estimation technique is proposed based on CAZAC preamble which was originally designed for time synchronization in orthogonal frequency division multiplexing (OFDM) system. Therefore, there is no extra load or throughput penalty on the system as the preamble used is already present in the system for time synchronization in OFDM systems and can be adapted for SNR estimation. In this work, we consider single-input-single-output (SISO)-OFDM cooperative with decode-and-forward (DAF) single relay and non-cooperative communication systems, when communicating over multipath channel in MATLAB®. At the receiver of the destination node in the SISO-OFDM cooperative system, the proposed SNR estimator exhibits around 0.5dB improvement over SNR estimation based on Schmidl's preamble technique. Whilst, the proposed SISO-OFDM cooperative system that invokes the proposed SNR estimator exhibits 2 dB improvements in comparison with non-cooperative system, when communicating over multipath channel. However, for transmission over AWGN channel the SISO-OFDM cooperative system that employs the proposed SNR estimator exhibits about 1 dB improvement over the non-cooperative system.