The Industrial Internet of Things (IIoT) integrates interconnected sensors
and devices to support industrial applications, but its dynamic environments
pose challenges related to data drift. Considering the limited resources and
the need to effectively adapt models to new data distributions, this paper
introduces a Continual Learning (CL) approach, i.e., Distillation-based
Self-Guidance (DSG), to address challenges presented by industrial streaming
data via a novel generative replay mechanism. DSG utilizes knowledge
distillation to transfer knowledge from the previous diffusion-based generator
to the updated one, improving both the stability of the generator and the
quality of reproduced data, thereby enhancing the mitigation of catastrophic
forgetting. Experimental results on CWRU, DSA, and WISDM datasets demonstrate
the effectiveness of DSG. DSG outperforms the state-of-the-art baseline in
accuracy, demonstrating improvements ranging from 2.9% to 5.0% on key datasets,
showcasing its potential for practical industrial applications.

基于蒸馏的自主引导 (DSG) 模型利用生成式回放机制，通过知识蒸馏从先前基于扩散的生成器向更新的生成器传输知识，从而提高了生成器的稳定性和再现数据的质量，有效缓解了在工业流数据中出现的挑战，并在 CWRU、DSA 和 WISDM 数据集上取得了优于现有基准模型的实验结果，展示了在实践工业应用中的潜力。

基于扩散式生成回放的工业流数据连续学习

Continual Learning with Diffusion-based Generative Replay for Industrial  Streaming Data

Traditional communications focus on regular and orthogonal signal waveforms
for simplified signal processing and improved spectral efficiency. In contrast,
the next-generation communications would aim for irregular and non-orthogonal
signal waveforms to introduce new capabilities. This work proposes a spectrally
efficient irregular Sinc (irSinc) shaping technique, revisiting the traditional
Sinc back to 1924, with the aim of enhancing performance in industrial Internet
of things (IIoT). In time-critical IIoT applications, low-latency and
time-jitter tolerance are two critical factors that significantly impact the
performance and reliability. Recognizing the inevitability of latency and
jitter in practice, this work aims to propose a waveform technique to mitigate
these effects via reducing latency and enhancing the system robustness under
time jitter effects. The utilization of irSinc yields a signal with increased
spectral efficiency without sacrificing error performance. Integrating the
irSinc in a two-stage framework, a single-carrier non-orthogonal frequency
shaping (SC-NOFS) waveform is developed, showcasing perfect compatibility with
5G standards, enabling the direct integration of irSinc in existing industrial
IoT setups. Through 5G standard signal configuration, our signal achieves
faster data transmission within the same spectral bandwidth. Hardware
experiments validate an 18% saving in timing resources, leading to either
reduced latency or enhanced jitter tolerance.

通过提出一种新的不规则 Sinc (irSinc) 塑形技术来提高工业物联网 (IIoT) 的性能，减少延迟和增强抖动容忍度，进而实现更高的数据传输速率和节省时间资源。