19 August 2025 in IEEE Transactions on Big Data
The dynamics of multivariate time series (MTS) data are jointly characterized by its nonlinear temporal dependencies and complex variable dependencies, making unsupervised time series anomaly detection a challenging task. Existing methods primarily rely on prediction or reconstruction errors, neglecting the valuable information within the variable dependencies. In this paper, we propose a variable dependency discrepancy-based Transformer (VDDFormer) for unsupervised MTS anomaly detection. VDDFormer comprises a variable correlation encoder, a temporal dependency encoder, and a reconstruction decoder. The variable correlation encoder capitalizes on a variable dependency attention mechanism, which employs self-attention to learn the global variable dependencies; meanwhile, the local variable dependencies are captured by the adaptive correlation matrix. The global and local variable dependencies are then used to compute the variable dependency discrepancy as a new intrinsic property to distinguish between normal and abnormal patterns. By integrating this new discrepancy with the reconstruction error, the model effectively enhances its anomaly differentiation capability. Extensive experiments on five real-world anomaly detection datasets demonstrate that VDDFormer effectively and robustly detects group anomaly patterns by leveraging the variable dependency discrepancy and achieves state-of-the-art performance on four out of the five datasets.
19 September 2025 in IEEE Communications Letters
Frequency modulation (FM) is widely used in broadcasting and IoT audio for its simplicity and low cost, yet remains susceptible to noise. We present DeepFM, a hybrid digital–analog system that supplements FM with a low-rate semantic side channel. A neural encoder compresses audio into quantized features sent alongside the FM signal. A channel-aware decoder fuses semantic bits, noisy analog input, and RSSI to restore high-fidelity audio. Evaluated on our SFR dataset at 0.5–12 kbps, DeepFM outperforms leading neural codecs in mainstream speech-quality metrics, particularly at low–medium bitrates (0.5–3 kbps). DeepFM demonstrates robust, real-time performance compatible with legacy FM infrastructure.
7 November 2025 in Knowledge-Based Systems