When fuel is vaporized and mixed well with air in the cylinder of premixed diesel engines, the mixture auto-ignites in one burst resulting in strong combustion noise, and combustion noise reduction is necessary to achieve high load premixed diesel engine operation.In this paper, an engine noise analysis was conducted by engine tests and simulations. The engine employed in the experiments was a supercharged single cylinder DI diesel engine with a high pressure common rail fuel injection system. The engine noise was sampled by two microphones and the sampled engine noise was averaged and analyzed by an FFT sound analyzer. The engine was equipped with a pressure transducer and the combustion noise was calculated from the power spectrum of the FFT analysis of the in-cylinder pressure wave data from the cross power spectrum of the sound pressure of the engine noise. The parameters investigated in the engine tests were the maximum rate of pressure rise, intake pressure by the supercharger, intake oxygen content by EGR, and the fuel injection timing, in all experiments the engine speed was maintained at 1600 rpm. The engine noise and combustion noise were sampled under 69 different test conditions and the data were compared with the results of the simulation. The engine test results show that the maximum rate of pressure rise is most strongly related to combustion noise, and that the combustion duration and the maximum value of the heat release rate (ROHRmax) are the second and third most important parameters in the generated noise.To discuss the results of the simulation, the heat release histories were approximated by Wiebe function, and the simulated combustion noise was calculated from the fitted curves of the heat release and coherent transfer function. To investigate the accuracy of the simulated combustion noise, the simulation data were compared with the engine test data. Further, the simulation makes it possible to determine, for example, changes in the CA50 with the same combustion phases, something that cannot be achieved by actual engine tests. The results of the simulations showed that the combustion noise has a maximum condition for CA50 and is closely related to the degree of constant volume conditions and the thermal efficiency.The simulation also suggests the possibility of reductions in combustion noise with extensions in the combustion duration by EGR. Further, the engine tests showed that a combustion noise of 6.3 dBA at 0.33 MPa IMEP can be achieved.