Search for: All records

Machine unlearning is becoming increasingly important as deep models become more prevalent, particularly when there are frequent requests to remove the influence of specific training data due to privacy concerns or erroneous sensing signals. Spatial-temporal Graph Neural Networks, in particular, have been widely adopted in real-world applications that demand efficient unlearning, yet research in this area remains in its early stages. In this paper, we introduce STEPS, a framework specifically designed to address the challenges of spatio-temporal graph unlearning. Our results demonstrate that STEPS not only ensures data continuity and integrity but also significantly reduces the time required for unlearning, while minimizing the accuracy loss in the new model compared to a model with 0% unlearning. 

Due to the insufficient transient amount of energy supplied from ambient energy sources and constrained amount of energy storage in super-capacitors, energy harvesting (EH) nodes are limited with operations and vulnerable to frequent faults due to energy scarcity. Consequently, such faults will reduce reliability and energy utility due to data collisions, lost data, or idle listening. To address these challenges, this work implements a novelty task scheduling scheme to minimize energy waste and maximize throughput under these scenarios and constraints. To demonstrate the effectiveness, we use a green test bed using LoRa nodes for evaluation. 

The emerging unmanned aerial vehicle (UAV) such as a quadcopter offers a reliable, controllable, and flexible way of ferrying information from energy harvesting powered IoT devices in remote areas to the IoT edge servers. Nonetheless, the employment of UAVs faces a major challenge which is the limited fly range due to the necessity for recharging, especially when the charging stations are situated at considerable distances from the monitoring area, resulting in inefficient energy usage. To mitigate these challenges, we proposed to place multiple charging stations in the field and each is equipped with a powerful energy harvester and acting as a cluster head to collect data from the sensor node under its jurisdiction. In this way, the UAV can remain in the field continuously and get the data while charging. However, the intermittent and unpredictable nature of energy harvesting can render stale or even obsolete information stored at cluster heads. To tackle this issue, in this work, we proposed a Deep Reinforcement Learning (DRL) based path planning for UAVs. The DRL agent will gather the global information from the UAV to update its input environmental states for outputting the location of the next stop to optimize the overall age of information of the whole network. The experiments show that the proposed DDQN can significantly reduce the age of information (AoI) by 3.7% reliably compared with baseline techniques. 

As the next-generation battery substitute for IoT system, energy harvesting (EH) technology revolutionizes the IoT industry with environmental friendliness, ubiquitous accessibility, and sustainability, which enables various self-sustaining IoT applications. However, due to the weak and intermittent nature of EH power, the performance of EH-powered IoT systems as well as its collaborative routing mechanism can severely deteriorate, rendering unpleasant data package loss during each power failure. Such a phenomenon makes conventional routing policies and energy allocation strategies impractical. Given the complexity of the problem, reinforcement learning (RL) appears to be one of the most promising and applicable methods to address this challenge. Nevertheless, although the energy allocation and routing policy are jointly optimized by the RL method, due to the energy restriction of EH devices, the inappropriate configuration of multi-hop network topology severely degrades the data collection performance. Therefore, this article first conducts a thorough mathematical discussion and develops the topology design and validation algorithm under energy harvesting scenarios. Then, this article developsDeepIoTRouting, a distributed and scalable deep reinforcement learning (DRL)-based approach, to address the routing and energy allocation jointly for the energy harvesting powered distributed IoT system. The experimental results show that with topology optimization,DeepIoTRoutingachieves at least 38.71% improvement on the amount of data delivery to sink in a 20-device IoT network, which significantly outperforms state-of-the-art methods. 

Molecular doping can increase the conductivity of organic semiconductors and plays an increasingly important role in emerging and established plastic electronics applications. 4-(1,3-Dimethyl-2,3-dihydro-1 H -benzimidazol-2-yl)- N , N -dimethylaniline (N-DMBI-H) and tris(pentafluorophenyl)borane (BCF) are established n- and p-dopants, respectively, but neither functions as a simple one-electron redox agent. Molecular hydrogen has been suggested to be a byproduct in several proposed mechanisms for doping using both N-DMBI-H and BCF. In this paper we show for the first time the direct detection of molecular hydrogen in the uncatalysed doping of a variety of polymeric and molecular semiconductors using these dopants. Our results provide insight into the doping mechanism, providing information complementary to that obtained from more commonly applied methods such as optical, electron spin resonance, and electrical measurements. 

Recently, brushless motors with especially high torque densities have been developed for applications in autonomous aerial vehicles (i.e. drones), which usually employ exterior rotortype geometries (ER-BLDC motors). These motors are promising for other applications, such as humanoids and wearable robots; however, the emerging companies that produce motors for drone applications do not typically provide adequate technical specifications that would permit their general use across robotics-for example, the specifications are often tested in unrealistic forced convection environments, or are drone-specific, such as thrust efficiency. Furthermore, the high magnetic pole count in many ER-BLDC motors restricts the brushless drives able to efficiently commutate these motors at speeds needed for lightly-geared operation. This paper provides an empirical characterization of a popular ER-BLDC motor and a new brushless drive, which includes efficiencies of the motor across different power regimes, identification of the motor transfer function coefficients, thermal response properties, and closed loop control performance in the time and frequency domains. The intent of this work is to serve as a benchmark and reference for other researchers seeking to utilize these exciting and emerging motor geometries.