Wireless sensor nodes (WSNs) operating with low-duty-cycle consume considerable energy during hibernation state due to high standby power, significantly limiting their lifespan. This study highly integrated a flexible antenna, a matching network, a rectifier, and an embedded wireless sensor to constitute a conformable WSN with a compact size of 65 mm × 54 mm. The proposed WSN realizes periodical wake-up by radio frequency (RF) energy transmission that reduces the standby current to 12 nA. A flexible asymmetric dipole antenna is designed to harvest RF energy and make the WSN conformable to curved surfaces with raii exceeding 50 mm. The experimental results indicate the WSN has a maximum wake-up distance of 17.38 m from an RF energy source and a lifespan of 2.7 years under bihourly data acquisition cycles when powered by a 3.7 V, 50 mAh battery. The proposed WSN reduces reliance on the hazardous, inefficient manual maintenance by eliminating battery replacement, significantly enhancing practicality in the appliance of Internet of Things.

Gesture recognition is a popular human-computer interaction technology, which has been widely applied in many fields (e.g., autonomous driving, medical care, VR and AR). However, 1) most existing gesture recognition methods focus on the fixed recognition scenarios with several gestures, which could lead to memory consumption and computational effort when continuously learning new gestures; 2) Meanwhile, the performance of popular class-incremental methods degrades significantly for previously learned classes (i.e., catastrophic forgetting) due to the ambiguity and variability of gestures. To tackle these challenges, we propose a novel class-incremental gesture recognition method with out-of-distribution (OOD) detection, which can continuously adapt to new gesture classes and achieve high performance for both learned and new gestures. Specifically, we construct an episodic memory with a subset of learned training samples to preserve the previous knowledge from forgetting. Moreover, the OOD detection-based memory management is developed for exploring the most representative and informative core set from the learned datasets. When a new gesture recognition task with strange classes comes, rehearsal enhancement is adopted to increase the diversity of memory exemplars for better fitting the real characteristics of gesture recognition. After deriving an effective class-incremental gesture recognition strategy, we perform experiments on two representative datasets to validate the superiority of our method. Evaluation experiments demonstrate that our proposed method substantially outperforms the state-of-the-art methods with about 2.17%-3.81% improvement under different class-incremental learning scenarios.

As affected by climate conditions, e.g. solar irradiation and temperature, the output power of photovoltaic (PV) generation is generally random and intermittent. Energy storage devices like batteries are thus necessary to be incorporated in a PV system to smooth the output power. In order to connect the PV-battery to the DC grid, a three-port bidirectional DC/DC converter based on the triple-active-bridge (TAB) topology is a superior choice. Such three-port converter is a strongly coupled system, which demands decoupling control methods to obtain independent control of input/output power in each port. However, the conventional control strategy with decoupling control cannot suppress the filter resonance in DC-grid-port effectively and could lead to the converter output current instability. Therefore, a stability control design for TAB-based three-port bidirectional DC/DC converters is proposed in this paper. It can manage the power flow among the ports with reliability and maintain the system stable operation. Simulation results are finally presented to verify the proposed control design.