Donghui Guo

Multimodal signal fusion is a cornerstone of biomedical engineering and intelligent sensing, enabling holistic analysis of heterogeneous sources such as electroencephalography (EEG), peripheral signals, speech, and imaging data. However, integrating diverse modalities in a computationally efficient and biologically plausible manner remains a significant challenge. Transformer-based fusion architectures rely on global cross-attention to integrate multimodal information but incur high computational costs. In contrast, STDP-driven fully connected layers adopt local learning rules, which restrict their ability to autonomously form efficient sparse topologies for complex multimodal tasks. To address these issues, we propose a novel end-to-end framework—the Multimodal Spiking Neural Network (MSNN)—featuring a fusion module grounded in the Generalized Distributive Law (GDL). This principled mechanism provides an efficient and interpretable means of integrating heterogeneous biomedical and sensory signals. The MSNN further incorporates structure-adaptive leaky integrate-and-fire (SALIF) neurons, enabling dynamic optimization of sparse connectivity to enhance fusion efficiency. The proposed MSNN is validated on a range of datasets, demonstrating strong versatility: it achieves binary classification accuracies of 92.29% (valence) and 91.08% (arousal) on the DEAP dataset for affective state decoding and 99.77% on the WESAD dataset for stress detection, while delivering state-of-the-art performance on standard pattern recognition tasks (MNIST & TIDIGITS: 99.01%) and event-driven neuromorphic datasets (MNIST-DVS & N-TIDIGITS: 99.98%). These results demonstrate that MSNN offers an effective and energy-efficient solution for multimodal sensor fusion in biomedical and intelligent sensing applications.

The consumer electronics industry has seen rising demand for passive sensor chips, particularly high-precision, low-power variants, in recent years. Responding to industry needs, this paper proposes a low-power, high-precision CMOS temperature sensor intended for RFID temperature sensing, which is compliant with the Electronic Product Code (EPC) Gen 2 standard. The design incorporates a multivibrator-based current-to-frequency converter that maintains high linearity and ultra-low power consumption across extended temperature ranges. Furthermore, this work presents an energy-efficient CMOS rectifier circuit with a fully complementary cross-coupled topology for enhanced RF energy harvesting, which utilizes diode-connected transistor ON-resistance to minimize leakage currents. The proposed EPC Gen-2 RFID tag is fabricated in a 0.18μm CMOS process occupying a silicon area of 0.39 mm2. Measurement results show that the sensor achieves the inaccuracy value of ±0.31∘C after one-point calibration, and the temperature sensor tag chip achieves a communication sensitivity of -17dBm, and the power conversion efficiency (PCE) of the proposed rectifier circuit can reach up to 62.4%.

This paper presents a low-power fourth-order single-bit switched capacitor sigma-delta modulator implemented by 180[ ]nm CMOS technology. In the modulator, the Floating Inverter Amplifier (FIA) with Correlated Level Shift (CLS) technique increasing gain is employed to decrease the modulators power consumption, and a two-tap Finite Impulse Response (FIR) is used in a feedback loop to reduce the integrator output swings. With the help of a simple digital circuit, the mismatches influence on the first-order integrators capacitor is mitigated because of reusing the sampling capacitor. Additionally, the work exploits the charge-sharing module to reduce the integrating capacitors size. The proposed modulator achieves 97.8[ ]dB peak Signal-to-Noise and Distortion Ratio (SNDR) at 102.4[ ]kHz over a bandwidth of 400[ ]Hz under the supply voltage of 1.2[ ]V VDD and 1.2[ ]V reference voltage while consuming a total power consumption of 3.94[ ][ ]W.