基於現場可程式化邏輯閘陣列之空氣超音波發射與接收雙向功能之大規模相位陣列系統設計與實現

Abstract

本研究針對空氣超音波相位陣列的發展需求，設計並實現了一套具備大規模擴充能力與雙向功能的新型系統。傳統系統雖已廣泛應用於醫學影像，但在空氣中應用時往往受限於輸出聲壓不足、僅能發射無法接收、以及更新率偏低等問題，難以支撐懸浮操控、觸覺回饋與定向音響等新興技術的進一步發展。為克服這些挑戰，本研究採用FPGA為核心控制平台，建立Master–Slave架構並搭配模組化拼接設計，使陣列通道數可由百級擴展至千級，並能依需求靈活配置，同時實現每一通道獨立相位控制以精準塑造三維聲場。類比驅動電路則能於保持低功耗的同時提供高達40V的操作電壓，使系統能輸出更強的聲壓，確保在多點聚焦與大範圍聲場操控下仍具穩定性。另一方面，系統亦首次整合高取樣率的接收功能，為後續聲學成像、姿態追蹤與手勢辨識等應用奠定基礎。實驗結果證實，本系統在最大擴充陣列下輸出超過7kPa的聲壓，展現出足以支援懸浮與觸覺合成的效能。整體而言，本研究所提出的架構不僅突破了傳統系統在輸出功率與接收能力上的限制，更強調透過擴大陣列規模來提升聲壓的重要性，提供一個兼具高擴充性與高效能的開發平台，為超音波相位陣列在人機互動、三維聲場控制及智慧感測等領域的未來應用開啟更多可能性。

This research addresses the development needs of airborne ultrasonic phased arrays by designing and implementing a novel system with large-scale scalability and bidirec tional functionality. While conventional systems have been widely applied in medical imaging, their use in air is often limited by insufficient acoustic output pressure, lack of receiving capability, and relatively low update rates, which constrain further advance ments in emerging technologies such as acoustic levitation, haptic feedback, and direc tional audio. To overcome these challenges, this work employs an FPGA as the central control platform, establishing a Master–Slave architecture combined with a modular tiling design. This enables the array to scale flexibly from hundreds to thousands of channels while allowing adaptive configuration according to application needs, with each channel supporting independent phase control for precise three-dimensional sound field shaping. The analog driving circuit maintains low power consumption while providing up to 40 V operating voltage, thereby enabling stronger acoustic output pressure and ensuring stabil ity under multi-focus and large-area field control. Furthermore, the system integrates, for the first time, a high-sampling-rate receiving function, laying the foundation for applica tions such as acoustic imaging, posture tracking, and gesture recognition. Experimental results demonstrate that the system achieves an output pressure exceeding 7 kPa with the maximum expanded array, sufficient to support levitation and haptic synthesis. Overall, the proposed architecture not only overcomes the limitations of conventional systems in terms of output power and reception capability, but also emphasizes the importance of scaling up array size to enhance acoustic pressure. It provides a highly scalable and high performance development platform, opening new possibilities for the future application of ultrasonic phased arrays in human–machine interaction, three-dimensional sound field control, and intelligent sensing.