A practical distributed active noise control algorithm overcoming communication restrictions

Junwei Ji, Dongyuan Shi, Zhengding Luo, Xiaoyi Shen, Woon‐Seng Gan

Abstract

By assigning the massive computing tasks of the traditional multichannel active noise control (MCANC) system to several distributed control nodes, distributed multichannel active noise control (DMCANC) techniques have become effective global noise reduction solutions with low computational costs. However, existing DMCANC algorithms simply complete the distribution of traditional centralized algorithms by combining neighbour nodes' information but rarely consider the degraded control performance and system stability of distributed units caused by delays and interruptions in communication. Hence, this paper develops a novel DMCANC algorithm that utilizes the compensation filters and neighbour nodes' information to counterbalance the cross-talk effect between channels while maintaining independent weight updating. Since the neighbours' information required barely affects the local control filter updating in each node, this approach can tolerate communication delay and interruption to some extent. Numerical simulations demonstrate that the proposed algorithm can achieve satisfactory noise reduction performance and high robustness to real-world communication challenges.

Introduction

Active Noise Control (ANC) systems operate by generating "anti-noise" to destructive interfere with unwanted acoustic signals. In large-scale environments, Multichannel ANC (MCANC) systems utilize multiple secondary sources (speakers) and error sensors (microphones) to achieve global noise reduction. Traditionally, these systems rely on a centralized processor, which faces significant scalability issues and high computational costs as the number of channels increases.

Distributed Multichannel ANC (DMCANC) offers a solution by spreading the computational load across multiple nodes. However, distributed systems face two primary challenges: cross-talk—where anti-noise from one node interferes with the error sensors of another—and communication constraints. In real-world wireless networks, latency and intermittent connectivity can destabilize traditional distributed algorithms that require tight synchronization between nodes. This research proposes a novel DMCANC algorithm designed to maintain high performance even under severe network degradation.

Proposed Methodology

The core innovation of this work is the decoupling of the local weight-update process from the global output generation using compensation filters ($c_{km}$).

In the proposed architecture, each node maintains a local control filter ($\psi_k$) that is updated independently based solely on the local error signal. To account for cross-talk from neighboring nodes, the system employs compensation filters learned during an offline modeling phase. These filters represent the relationship between a node's own secondary path and the cross-secondary paths from its neighbors.

The internal structure of Node k in the proposed DMCANC algorithm, highlighting the decoupled updating mechanism and the integration of neighbor information through compensation.

The final output of a node is a combination of its local filter and the compensation-filtered versions of its neighbors' filters. By moving the interaction with neighbor data out of the weight-update loop, the algorithm becomes inherently robust to network jitter; if a neighbor's data is delayed or lost, the node continues to update its local weights using the last known state without stalling the convergence process.

Experimental Setup

The algorithm was validated through extensive numerical simulations. The primary noise source was a broadband signal (100 Hz to 1000 Hz) sampled at 16,000 Hz. Acoustic paths were modeled as band-pass filters (50 Hz to 5000 Hz) with tap lengths of 320, while control filters used 512 taps. To test robustness, the researchers simulated various network conditions, including latencies up to 4000 sampling periods and communication frequencies as low as 0.01%.

Results and Discussion

Numerical results demonstrate that the proposed DMCANC algorithm achieves performance parity with the gold-standard centralized McFXLMS algorithm under ideal conditions. More importantly, the system exhibits remarkable resilience to communication impairments.

Mean Squared Error (MSE) performance of Node 1 under varying communication frequencies, demonstrating the algorithm's stability even at 0.1% communication frequency.

As shown in the performance analysis, the system maintains near-optimal noise reduction even when data is exchanged only ~16 times per second (0.1% frequency). Furthermore, the algorithm remained stable with latencies up to 187.5 ms (3000 sampling periods), which exceeds the typical latency of standard Wi-Fi hardware. This robustness stems from the "Aha" moment of the research: treating cross-talk as a deterministic path difference that can be compensated for, rather than random noise to be filtered out.

Conclusion and Future Work

This paper presents a practical DMCANC algorithm that successfully overcomes the limitations of real-world communication networks. By utilizing compensation filters to decouple node updates, the system achieves centralized-level performance with significantly higher robustness to latency and packet loss.

Future research will focus on transitioning from offline to online modeling of compensation filters to support dynamic environments where sensors or speakers may move. Additionally, the authors plan to implement the algorithm on physical distributed hardware to verify its effectiveness in complex 3D acoustic spaces using standard wireless protocols.

Summary updated

Abstract

By assigning the massive computing tasks of the traditional multichannel active noise control (MCANC) system to several distributed control nodes, distributed multichannel active noise control (DMCANC) techniques have become effective global noise reduction solutions with low computational costs. However, existing DMCANC algorithms simply complete the distribution of traditional centralized algorithms by combining neighbour nodes' information but rarely consider the degraded control performance and system stability of distributed units caused by delays and interruptions in communication. Hence, this paper develops a novel DMCANC algorithm that utilizes the compensation filters and neighbour nodes' information to counterbalance the cross-talk effect between channels while maintaining independent weight updating. Since the neighbours' information required barely affects the local control filter updating in each node, this approach can tolerate communication delay and interruption to some extent. Numerical simulations demonstrate that the proposed algorithm can achieve satisfactory noise reduction performance and high robustness to real-world communication challenges.
BibTeX
@inproceedings{ji2023practical,
  author = {Junwei Ji and Dongyuan Shi and Zhengding Luo and Xiaoyi Shen and Woon‐Seng Gan},
  title = {{A practical distributed active noise control algorithm overcoming communication restrictions}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2023},
  doi = {10.1109/icassp49357.2023.10097013},
}

Loading related content...

Figure