Spiking neural networks (SNNs) have emerged as a promising model for energy-efficient, event-driven processing of asynchronous event streams from Dynamic Vision Sensors (DVSs), a class of neuromorphic image sensors with microsecond-level latency and high dynamic range. Nevertheless, challenges persist in optimising training and effectively handling spatio-temporal complexity, which limits their potential for real-time applications on embedded sensing systems such as object tracking and recognition. Targeting this neuromorphic sensing pipeline, this paper proposes the Dynamic Tracking with Event Attention Spiking Network (DTEASN), a novel framework designed to address these challenges by employing a pure SNN architecture, bypassing conventional convolutional neural network (CNN) operations, and reducing GPU resource dependency, while tailoring the processing to DVS signal characteristics (asynchrony, sparsity, and polarity). The model incorporates two innovative, self-developed components: an event-driven multi-scale attention mechanism and a spatio-temporal event convolver, both of which significantly enhance spatio-temporal feature extraction from raw DVS events. An Event-Weighted Spiking Loss (EW-SLoss) is introduced to optimise the learning process by prioritising informative events and improving robustness to sensor noise. Additionally, a lightweight event tracking mechanism and a custom synaptic connection rule are proposed to further improve model efficiency for low-power, edge deployment. The efficacy of DTEASN is demonstrated through empirical results on event-based (DVS) object recognition and tracking benchmarks, where it outperforms conventional methods in accuracy, latency, event throughput (events/s) and spike rate (spikes/s), memory footprint, spike-efficiency (energy proxy), and overall computational efficiency under typical DVS settings. By virtue of its event-aligned, sparse computation, the framework is amenable to highly parallel neuromorphic hardware, supporting on- or near-sensor inference for embedded applications.

Piezoelectric vibration energy harvesters (PVEHs) have demonstrated their potential for sustainable energy generation from diverse ambient vibrations for low-power devices and systems. However, great challenges remain concerning harvesting more energy from low-frequency input sources and broadband random excitations. In this paper, a novel PVEH featuring a lead zirconate titanate (PZT) hollowed trapezoidal cantilever beam is proposed, simulated, optimized and fabricated to effectively broaden its output bandwidth at low frequency ranges. Under 1 g acceleration, the traditional solid PVEH showed a resonant frequency of 47.80 Hz and a maximum output power density of 14.22 mW/cm³, while the proposed PVEH showed two resonant frequencies of 21.30 Hz and 50.40 Hz. Compared to the traditional solid PVEH, the first-order resonant frequency was reduced by 55.44% and the corresponding maximum output power density was 3.3 times higher in the proposed PVEH. Furthermore, a parallel synchronized switch harvesting inductor quadruple voltage rectifier (P-SSHI-QVR) circuit is designed to extract energy from the proposed PVEH. For the proposed PVEH incorporating the P-SSHI-QVR circuit, the maximum stored voltage was 20.49 V at a first-order resonant frequency of 21.30 Hz and 5.68 V at a second-order resonant frequency of 50.40 Hz, with corresponding maximum stored powers of 36.89 μW and 2.97 μW, respectively. This study verified the feasibility of the optimized design through simulation and experimental comparison.