Experimental study on the effects of increasing number of system components and sampling rate of an active noise blocker for a tilted window

To reduce the impact of noise entering the room through a tilted window active noise control in general and active noise blocking in particular is subject of current research. This concept uses several loudspeakers and microphones to generate sound pressure nodes. They prevent the transfer of acoustic energy. The loudspeakers and microphones are evenly distributed in the transmission path through the tilted window to achieve a global effect by a local measure. The system is driven by a multiple input multiple output feedforward filtered reference (x) least mean square (MIMO feedforward FxLMS) controller and implemented on a real time system. Using parallel computing on two processors with eight cores in total the number of loudspeakers and error microphones as well as the sampling rate can be increased. The idea behind these changes is to receive a better coverage of the gap and to increase the limit of controllable frequencies and to accelerate and stabilize the algorithm. This paper presents the results of an experimental study comparing the results of the adapted measurements to previous ones. Configurations with 2 reference microphones, 14 loudspeakers and 14 error microphones and with 2 reference microphones, 20 loudspeakers and 20 error microphones with different sampling rates have been investigated. It has been shown that for the 2x14x14 system an increased sampling rate has only minor impact. By contrast, the 2x20x20 configuration with more system components shows a significant improvement of noise reduction. On this case an increased sampling rate has additional impact.