High-Precision Harmonic Distortion Level Measurement of a Loudspeaker Using Adaptive Filters in a Noisy Environment

Abstract

The harmonic distortion level (HDL), an important criterion used to evaluate loudspeaker performance, can usually be measured using spectral analysis. However, spectral analysis entails great computational complexity. Furthermore, long measurement times are necessary to measure HDL accurately in a noisy environment. Therefore, we proposed the HDL measurement technique for a loudspeaker using an adaptive filter. This measurement technique can measure HDL as accurately as spectral analysis using fast Fourier transform (FFT). Furthermore, its computational complexity is much less than that of spectral analysis. Nevertheless, this measurement technique requires a long measurement time to assess the convergence of filter coefficients. This paper presents a description of the new HDL measurement technique of a loudspeaker using plural adaptive filters. The proposed measurement technique can measure HDL more accurately than spectral analysis in a noisy environment. We evaluated the performance of the proposed measurement technique using computer simulations. Results based on computer simulations show that the proposed measurement technique is more effective than spectral analysis in an acoustic environment with background noise.