Nowadays, Multiple-Input Multiple-Output (MIMO) is the technique of wireless communication that utilizes the multiple antennas in between the transmitter and receiver that improves the quality and reliability of signals, the technique of sparse channel estimation relays the dense pilot symbols that leads to the high overhead and reduces the spectral efficiency. However, the sparse channel estimation failed difficult in obtaining the practice and required with the prior knowledge of the channel sparsity level that leads to high latency. To overcome this drawback, in this research proposed and improved Pilot Pattern Optimization (IPPO) recovers model low coherence by Bacteria Foraging Algorithm (BFA) and this technique provides interference through pilot pattern that improves the robustness to noise by leveraging the foraging behaviors. The system model which are considered for optimization techniques are Block Diagonal Channel State (BDCS), Orthogonal Frequency Division Multiplexing (OFDM), Basis Expansion Model (BEM), and finally, the proposed system model Pilot Pattern are employed for sparse channel estimation. The proposed model gained better results in terms of Normalized Mean Squared Error (NMSE) and Signal-to-Noise Ratio (SNR) metrics the values obtained are 0.13 dB and 37 dB when compared with the existing method of Zero-Attracting Least Mean Squares (ZALMS), respectively.