Data Fusion is one of the attractive topic in sonar signal processing. Decision level data fusion of multi-sensor (multi-array) system is described in this paper. The optimum linear data fusion algorithm for N independent observation data is derived. It is proved that the estimation error of optimum data fusion is not greater than that of individual components. The expression of estimation error and weight coefficients are presented. The results of numerical calculation and some examples are illustrated. The effect of input signal to noise ratio for the data fusion is described. Data fusion is one of the attractive topics in sonar signal processing. It is a data fusion of multi-sensor (multi-array) system described in this paper. that the estimation error of optimum data fusion is not greater than that of individual components. The expression of estimation error and weight coefficients are presented. The results of numerical calculation and some examples are illustrated. The effect of input signal to noise ratio for the data fusion is described.