Active Noise Cancellation systems require an accurate estimate of the secondary path between the loudspeakers and microphones. Whenever the secondary paths change over time, due to occupancy, obstruction, temperature/humidity, transducer aging, or window opening, the estimations become erroneous since they are static. These errors may cause instabilities whereby noise is generated instead of cancelled. To reduce the effects of instability, it is first necessary to detect it. A method has been developed where the error signal, simulated disturbance signal, and loudspeaker (via secondary path) output are queried from the system in real time and combined to create new normalized variables to evaluate the stability of the adaptive filter. These variables are analyzed during system tuning, which include intentionally induced instabilities, so that thresholds can be determined for a given car model.
https://www.aes.org/e-lib/browse.cfm?elib=16420
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Learn more about the AES E-Library
Start a discussion about this paper!