一种可适应运动声源的分布式汽车鸣笛声实时定位系统

A Distributed Real-Time Location System for Automobile Whistle Adaptive to Moving Sound Source

针对违章鸣笛车辆的定位问题,提出了一种基于分布式传声器阵列的运动声源快速定位系统。采用全球导航卫星系统(GNSS)时钟实现传声器间的时间同步,并将同步采集的声音信息传送到云端数据库,应用云计算技术实现声源定位算法。相比于集中式传声器阵列,该系统可大幅降低传声器的数量和运算资源,具有成本经济、部署灵活的优点。采用基于到达时间差-到达频率的快速定位算法,充分利用多普勒效应导致的分布式传声器之间到达频率差异信息来克服到达时间差法难以适应运动声源的瓶颈,避免了计算复杂、运算量大的消除多普勒效应过程,具有运算复杂度低、且能适应高速运动声源的优点。系统仿真和现场实验结果均表明该系统能够实现对高速运动声源的快速精确定位,可较好地适用于汽车鸣笛声定位场景。

Aiming at the localization problem of illegal whistle vehicles, a rapid location system of moving sound source based on distributed microphone array is proposed. GNSS clock is used to realize the time synchronization between microphones, and the sound information collected synchronously is transmitted to the cloud database.Meanwhile, cloud computing technology is applied to realize the sound source localization algorithm. Compared with the centralized microphone array, it can greatly reduce the number of microphones and computing resources, and has the advantages of cost economy and flexible deployment. In addition, the proposed system adopts a fast location algorithm based on the arrival time difference and arrival frequency, which can make full use of the arrival frequency difference information between distributed microphones caused by Doppler effect to overcome the bottleneck of the arrival time difference method that is difficult to adapt to moving sound source. This proposed method can avoid the process of eliminating Doppler effect with complex calculation and large amount of calculation, and has low computational complexity and can adapt to high-speed moving sound source. The simulation and field experiment results show that the proposed system can realize fast and accurate positioning of high-speed moving sound source, and can be applied to car whistle positioning well.