To solve the sparse channel estimation problem in Orthogonal Frequency Division Multiplexing (OFDM) systems, this paper presents a pilot design algorithm based on multi-pilot parallel searching and binary coding. Under the Mutual Incoherence Property (MIP), we consider the deterministic pilot design problem by minimizing the coherence of the dictionary matrix. The pilot is also required to be placed within the effective subcarrier range for more practical purposes. In this scheme, the pilot set is binary coded. In each iteration, the sequence of the binary code segment is recombined to update the approximately optimal pilot position. The update of some coding segments can change multiple pilot positions at the same time. It can increase the diversity of pilots, and reduce the complexity of computation. Experimental results show that, compared with the existing pilot optimization methods, this algorithm can find a smaller MIP measurement matrix. Therefore, the corresponding pilot mode can significantly improve the performance of channel estimation.