Analysis and realization of an active noise control system on DSP hardware using an fMRI bore model

The performance of Active Noise Control (ANC) Algorithms (epsilon-nLMS and APA) are tested on floating point DSP hardware with high sampling rates to reveal their effectiveness in cancelling noise at high sound pressure levels. The results presented use a realistic primary path of an fMRI bore simulated by a half cylindrical acrylic bore of 0.76m (D) x 1.52m (L). Detailed results of the performance of the ANC system are presented in the paper. The results were obtained by using an automated system, designed to systematically test any frequency range with high precision. Our experiments show that the LMS algorithm gives the best performance over all frequencies with cancellation of over 50dB(A) over a 5kHz frequency band.