Sankha Subhra Bhattacharjee,Andreas Jonas Fuglsig,Flemming Christensen,Jesper Rindom Jensen,Mads Græsbøll Christensen
Abstract:Performance of sound zone control (SZC) systems deployed in practical scenarios are highly sensitive to the location of the listener(s) and can degrade significantly when listener(s) are moving. This paper presents a robust SZC system that adapts to dynamic changes such as moving listeners and varying zone locations using a dictionary-based approach. The proposed system continuously monitors the environment and updates the fixed control filters by tracking the listener position using audio signals only. To test the effectiveness of the proposed SZC method, simulation studies are carried out using practically measured impulse responses. These studies show that SZC, when incorporated with the proposed audio-only position tracking scheme, achieves optimal performance when all listener positions are available in the dictionary. Moreover, even when not all listener positions are included in the dictionary, the method still provides good performance improvement compared to a traditional fixed filter SZC scheme.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: in practical application scenarios, when the listener's position changes, the performance of the Sound Zone Control (SZC) system will decline significantly. Traditional methods rely on fixed control filters, which are pre - designed based on the acoustic characteristics of the environment and the position of the control points. However, environmental changes (such as temperature changes, movement of speakers and microphones, changes in zone size, and movement of the listener, etc.) will make these preset filters no longer optimal, resulting in a decline in system performance.
To solve this problem, this paper proposes a robust sound zone control system that can adapt to dynamic changes, such as moving listeners and changing zone positions. Specifically, this method uses a dictionary - based approach to track the listener's position only by using audio signals and updates the fixed control filters according to the tracking results. This enables the system to maintain good performance when the listener moves.
### Main contributions:
1. **Audio signals for position tracking**: This method monitors the sound field through a set of observation microphones (different from the control points in the bright and dark zones) and uses audio signals to estimate the listener's position.
2. **Dictionary matching to select the best filter**: The system maintains a pre - measured impulse response (IR) dictionary. These IRs are responses at different listener positions from the speaker to the observation microphones. By matching the observed sound signal with the estimated signal in the dictionary, the most similar listener position is selected, and the optimal control filter is selected accordingly.
3. **Frame - level real - time adaptation**: This method is executed on each time frame, ensuring that the system can adapt to the movement of the listener in real - time, thereby maintaining the best sound zone control effect.
### Experimental verification:
Through simulation studies, this paper shows that when the dictionary contains all possible listener positions, the proposed method can accurately track the listener's position and switch to the optimal filter, thereby achieving the best control performance. Even when the dictionary does not contain all listener positions, this method still provides better performance improvement than the traditional fixed - filter method.
### Conclusion:
This research proposes a robust sound zone control method based on audio signals, which can maintain good performance when the listener's position changes. This method can enhance the robustness of sound zone control with only a small number of observation microphones and has broad application prospects.