Dynamic measurement of sound fields using compressed sensing

本研究的目的是利用压缩感知理论，探索用移动麦克风测量声场（全声函数）的可能性，并开发相应的工具。首先，考虑单个全向麦克风，假设其空间位置在整个测量过程中相对于初始位置是已知的。在实践中，这可以通过以受控的方式移动麦克风（例如，通过机械臂）或通过其他方式进行随机运动并使用快速跟踪设备观察麦克风位置来实现。通过利用基于压缩感知的测量中的位置信息，所提出的方法有可能仅使用准随机移动的手持麦克风来捕获给定音量内的声场。与使用空间分布麦克风的大阵列捕捉声场的方法相比，预计有许多优点。例如，一个麦克风，不需要精确的麦克风校准。实际应用是多方面的，从最佳扬声器放置的听音室补偿到建筑声学，在那里可以用最少的设备获得可靠的声场测量。该项目的第二阶段将致力于优化空间适应性，并为多个同时有源的多通道情况提供解决方案，包括使用移动麦克风阵列。

The objective of the proposed research is to explore the possibilities of measuring sound fields (plenacoustic functions) with moving microphones by making use of the theory of compressed sensing and to develop the corresponding tools. At first, a single omnidirectional microphone is considered for which the spatial location is assumed to be known during the entire measurement with respect to the initial position. This can be achieved in practice by either moving the microphone in a controlled way (e.g., by a robot arm) or by carrying out random movement by some other means and observing the microphone position with a fast tracking device. By exploiting the position information within a compressed-sensing based measurement, the proposed approach has the potential to capture the sound fields within a given volume by using only a hand-held microphone that is quasi randomly moved. A number of advantages is expected over methods that capture sound fields with large arrays of spatially distributed microphones. For example, with one microphone, no exact microphone calibration is needed. The practical applications are manifold, from listening-room-compensation over optimum loudspeaker placement to architectural acoustics, where reliable sound-field measurements could be obtained with a minimum of equipment. The second phase of the project will be directed toward optimizing spatial adaptivity and providing solutions for the multichannel case with several simultaneously active sources, including the use of moving microphone arrays.