Research On The DSP Based Real-time Sound Intensity Measurement System By Use Of The Two-microphone Cross Spectral Method

Compared with the developed sound pressure measurement technology, sound intensity measurement technology is still a young acoustic measurement method. Sound intensity is defined as the sound power per area in the process of the sound wave promulgation. Its direction is the same as the direction of sound wave promulgation. Integrating the sound intensity vector on the area which surrounds a sound source gets the sound power of this sound source. Since the integral of the sound power of the sound sources which are not enclosed by the area is zero, one could measure the sound power in ordinary environments correctly by measuring sound intensity. Using the direction information of sound intensity, one could easily identify the sound source and study the distribution and promulgation of the sound source. Sound intensity measurement technology has played a very important role in measuring sound power, identifying the main sound sources, measuring the acoustical characters of materials and so on. Therefore, studying the sound intensity measurement theory and developing a real-time sound intensity measurement system which meets the situation of our country is very significant.Through comparing the advantages and disadvantages of the current sound intensity systems, a real-time sound intensity measurement system is developed in this dissertation. This system is composed of a sound intensity probe, a DSP based multi-channel signal analysis system and a personal computer. The sound intensity probe is used to get the sound pressure signal of the two microphones. The DSP based multi-channel signal analysis system is used to conduct the real-time processing of the sound intensity signal, which includes the FFT computing, the sound intensity octave spectrum computing, A weighed sound intensity spectrum computing and so on. The personal computer is used for the data storage, results display, and other post processing. This hardware configuration uses the fast computing ability of DSP and the powerful graphics display ability of the personal computer to make the best system.In the software algorithm, the system combines the sound pressure digital signals to make a complex number before the FFT computing. First compute the FFT of this complex number and then separate the result to get the FFT results of the sound pressure signals. Using this method, the system gets the FFT results of the sound pressure signals by just one FFT computing and the simple addition and subtraction. This greatly reduces the computing time of the system. On the aspect of error compensation, the system using the geometric average sound pressure instead of the arithmetic average sound pressure to get the sound pressure of the measurement point. This method not only reduces the high frequency error in the sound intensity measurement, which compensates the finite difference approximation error, but also greatly reduces the computation of the sound pressure spectrum. The last experiments have proved that this system meets the design requirements on both precision and real-time ability.