A Digital Twin Architecture for Wireless Networked Adaptive Active Noise Control

• Published in: IEEE/ACM transactions on audio, speech, and language processing Vol. 30; pp. 2768 - 2777
• Main Authors: Shi, Chuang, Du, Feiyu, Wu, Qianyang
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Active control; Active noise control; Adaptation; Adaptive control; Adaptive filters; Cloud computing; Computer architecture; Controllers; digital twin; Digital twins; human-centered computing; Least mean squares; Least mean squares algorithm; LF noise; Loudspeakers; Microphones; networked control systems; Noise control; Noise generation; Signal generation; Speech processing; Wireless communication; wireless sensor network; Wireless sensor networks
• ISSN: 2329-9290; 2329-9304
• DOI: 10.1109/TASLP.2022.3199992

The active noise control (ANC) is a complementary technique to the passive noise control (PNC) to reduce the low frequency noise. The ANC controller can be implemented by pre-trained filters or adaptive filters. The adaptive ANC controller is advantageous in its adaptation to environmental changes. However, the algorithm complexity of the adaptive ANC controller increases with the scale of ANC applications, making it difficult to be carried out on low-cost processors. To resolve this problem, cloud computing should be utilized in ANC systems, and thus the wireless networked ANC system is proposed. Since it is crucial for ANC controllers to generate the anti-noise wave in real time, this paper formulates a digital twin architecture that implements the control filter adaptation in the cloud and the anti-noise signal generation on the local controller, respectively. A digital twin filtered-reference least mean squares (DT-FxLMS) algorithm is proposed to coordinate the digital twin with the local controller. Simulation and experiment results demonstrate the effectiveness and efficiency of the wireless networked ANC system based on the digital twin architecture.