Guillermo Garcia-Barrios,Eduardo Latorre Iglesias,Juana M. Gutierrez-Arriola,Ruben Fraile,Nicolas Saenz-Lechon,Victor Jose Osma-Ruiz
Abstract:Acoustic reverberation is one of the most relevant factors that hampers the localization of a sound source inside a room. To date, several approaches have been proposed to deal with it, but have not always been evaluated under realistic conditions. This paper proposes exploiting spatial diversity as an alternative approach to achieve robustness against reverberation. The theoretical arguments supporting this approach are first presented and later confirmed by means of simulation results and real measurements. Simulations are run for reverberation times up to 2 s, thus providing results with a wider range of validity than in other previous research works. It is concluded that the use of systems consisting of several, sufficiently separated, small arrays leads to the best results in reverberant environments. Some recommendations are given regarding the choice of the array sizes, the separation among them, and the way to combine SRP-PHAT maps obtained from diverse arrays.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to improve the robustness of the sound source localization system in reverberant environments. Specifically, reverberation is one of the main factors affecting indoor sound source localization. Although many existing methods have proposed solutions, they are often not fully evaluated under actual conditions. This paper proposes a method using spatial diversity to enhance the system's robustness to reverberation, and verifies the effectiveness of this method through theoretical analysis, simulation results and actual measurements.
### Main problem statements:
1. **The influence of reverberation on sound source localization**:
- Sound source localization algorithms usually rely on estimating the time - delay - of - arrival difference (TDOA) between two microphones. Under non - reverberant conditions, the TDOA can be accurately estimated by the generalized cross - correlation (GCC) function. However, in a reverberant environment, the acoustic signal will experience delay spread, causing the GCC function to have multiple peaks, thus affecting the estimation accuracy of the TDOA.
- Specifically, reverberation will cause the main peak of the GCC function to widen and secondary peaks may appear, and these secondary peaks may exceed the height of the main peak, resulting in TDOA estimation errors.
2. **Limitations of existing methods**:
- Many existing methods have been tested under low - reverberation conditions, but the reverberation time in the actual environment is usually long (0.5 seconds to 3 seconds), so it is necessary to evaluate the performance of the system under a wider range of reverberation conditions.
- Some methods need to know the information of the acoustic channel in advance, which may be difficult to achieve in practical applications.
- The influence of the spatial layout on the performance of the microphone array also lacks systematic research.
### Solutions:
- **Using spatial diversity**:
- This paper proposes to improve the robustness of the system by using multiple sufficiently separated small arrays. Specifically, by combining the SRP - PHAT diagrams of different arrays, the influence of reverberation can be reduced.
- The author has verified the effectiveness of this method through theoretical analysis and experiments, especially under longer reverberation times (up to 2 seconds).
### Key contributions:
1. **Theoretical analysis**:
- Analyzed in detail the influence of reverberation on the GCC function and derived the relevant mathematical model.
- Proved that in an ideal situation, if the reverberation responses of two microphones are proportional, the reverberation will not affect the GCC function and the TDOA estimation.
2. **Simulation and actual measurement**:
- Verified the effectiveness of the method using spatial diversity in a reverberant environment through simulation and actual measurement.
- The simulation results show that using multiple separated small arrays can significantly improve the robustness of the system.
3. **Recommended configurations**:
- Provided specific suggestions on array size, array spacing and how to combine the SRP - PHAT diagrams of different arrays.
### Conclusion:
This paper has proved through theoretical analysis and experimental verification that using spatial diversity can effectively improve the robustness of the sound source localization system in a reverberant environment. This method is of great significance in practical applications, especially in scenarios where high - precision sound source localization is required.