Hand rehabilitation training and assessment require the frequent participation of professional doctors, which is time-consuming, laborious, and sometime nonquantitative. To liberate both doctors and patients, a finger motion monitoring glove is expected to give a helping hand by monitoring the speed and state of hand movement in real-time. However, the existing finger motion monitoring gloves suffer from low sensitivity, narrow detection range and lack of fabrication consistency. Herein, a highly sensitive finger motion monitoring glove with an intrinsic surface microstructure is fabricated based on the stable printing method, and hand rehabilitation training and assessment are realized with the help of machine learning method. The printed strain sensors achieve a high strain sensitivity (GF reaches 100.69 at 30–50% strain), wide response range (0.1–50%), fast response/recovery speed (71.43/178.49 ms), satisfied durability (function properly after 5000 cycles), and low batch-to-batch variation (within 0.10). These advantages enable the printed sensors to monitor the finger movements quickly and comprehensively, thus making it practicable for hand rehabilitation training and assessment. This work provides a simple and stable method to obtain highly sensitive finger motion monitoring glove, which is expected to make hand rehabilitation training and assessment more convenient and reliable.
Trust based approaches have been widely used to counter insider attacks in wireless sensor networks because traditional cryptography-based security mechanisms such as authentication and authorization are not effective against such attacks. A trust model, which is the core component of a trust mechanism, provides a quantitative way to evaluate the trustworthiness of sensor nodes. The trust evaluation is normally conducted by watchdog nodes, which monitor and collect other sensors' behavior information. Most existing works mainly focus on the design of the trust models and how these models can be used to defend against certain insider attacks. However, these studies are empirical with the implicit assumption that the trust models are secure and reliable. In this paper, we discuss several security vulnerabilities that watchdog and trust mechanisms have, examine how inside attackers can exploit these security holes, and finally propose defending approaches that can mitigate the weaknesses of trust mechanism and watchdog.
In cluster-based wireless sensor networks, cluster heads (CHs) gather and fuse data packets from sensor nodes; then, they forward fused packets to the sink node (SN). This helps wireless sensor networks balance energy effectively and efficiently to prolong their lifetime. However, cluster-based WSNs are vulnerable to selective forwarding attacks. Compromised CHs would become malicious and launch selective forwarding attacks in which they drop part of or all the packets from other nodes. In this paper, a data clustering algorithm (DCA) for detecting a selective forwarding attack (DCA-SF) is proposed. It can capture and isolate malicious CHs that have launched selective forwarding attacks by clustering their cumulative forwarding rates (CFRs). The DCA-SF algorithm has been strengthened by changing the DCA parameters (Eps, Minpts) adaptively. The simulation results show that the DCA-SF has a low missed detection rate of 1.04% and a false detection rate of 0.42% respectively with low energy consumption.
Many studies have shown that clustered Wireless Sensor Networks (WSNs) have a better performance in terms of the balance of energy and lifetime. However, due to the harsh environment and open communication, the clustered WSNs are easy to be attacked. The selective forwarding attack is one of the most difficult attacks to be detected. When a malicious sensor node launches the selective forwarding attacks, it drops part of or all the data packets it received. In this paper, we propose a Noise-Based Density Peaks Clustering (NB-DPC) algorithm for detecting selective forwarding attacks. It can detect selective forwarding attacks by clustering the Cumulative Forwarding Rates (CFRs) of all sensor nodes. The NB-DPC algorithm has been improved by defining noise points specifically for identifying malicious behavior and deleting the unnecessary steps in Density Peaks Clustering (DPC) for faster detection speed. The NB-DPC has a low Missed Detection Rate (MDR) and False Detection Rate (FDR) of below 1% according to the simulation results.