A low power, real time acoustic source localization (ASL) network based on intelligent microsystems was proposed and realized by an array of 1.68-cm${}^3$, miniaturized nodes. The spherical interpolation algorithm based on time difference of arrival (TDOA) was employed on the network to achieve high localization accuracy and topology adaptability with low, micro controller unit (MCU) level computing resources. Optimized localization accuracy of 0.12 m in 5 × 5 ${\mathrm{m}}^2$ area was achieved by a square-shape layout model out of three topologies of four nodes. Experiment results illustrate that the developed ASL network may achieve comparable accuracy to that of traditional ASL systems with significant advances in miniaturizing the device size, lowering the power consumption and increasing the topology flexibility.

提出了一种基于智能微系统的低功耗、实时声源定位（ASL）网络，并通过1.68-cm阵列实现${}^3$，小型化节点。网络采用基于到达时间差（TDOA）的球面插值算法，以低微控制器（MCU）级计算资源实现高定位精度和拓扑适应性。在 5 × 5 中优化定位精度为 0.12 m${\mathrm{米}}^2$面积是通过四个节点的三个拓扑结构的方形布局模型来实现的。实验结果表明，所开发的 ASL 网络可以达到与传统 ASL 系统相当的精度，并且在减小设备尺寸、降低功耗和增加拓扑灵活性方面取得了重大进展。