Echo state network (ESN) as a novel artificial neural network has drawn much attention from time series prediction in edge intelligence. ESN is slightly insufficient in long-term memory, thereby impacting the prediction performance. It suffers from a higher computational overhead when deploying on edge devices. We firstly introduce the knowledge distillation into the reservoir structure optimization, and then propose the echo state network based on improved knowledge distillation (ESN-IKD) for edge intelligence to improve the prediction performance and reduce the computational overhead. The model of ESN-IKD is constructed with the classic ESN as a student network, the long and short-term memory network as a teacher network, and the ESN with double loop reservoir structure as an assistant network. The student network learns the long-term memory capability of the teacher network with the help of the assistant network. The training algorithm of ESN-IKD is proposed to correct the learning direction through the assistant network and eliminate the redundant knowledge through the iterative pruning. It can solve the problems of error learning and redundant learning in the traditional knowledge distillation process. Extensive experimental simulation shows that ESN-IKD has a good time series prediction performance in both long-term and short-term memory, and achieves a lower computational overhead.

Staff attendance information has always been an important part of corporate management. However, some opportunistic employees may consign others to punch their time cards, which hampers the authenticity of attendance and effectiveness of record keeping. Hence, it is necessary to develop an innovative anti-cheating system for office attendance. Radio-Frequency IDentification (RFID) offers new solutions to solve such problems because of its strong anti-interference capability and non-intrusiveness. In this paper, we present a smart attendance system that extracts distinguishable phase characteristics of individuals to enable recognition of various targets. A frequency distribution histogram is extracted as a fingerprint for recognition and the K-means clustering method is utilized for more fine-grained recognition of targets with similar features. Compared with traditional attendance mechanisms, RFID-based attendance systems are based on living biological characteristics, which greatly reduces the possibility of false records. To evaluate the performance of our system, we conducted extensive experiments. The results of which demonstrate the efficiency and accuracy of our system with an average accuracy of $\mathrm{92}\mathrm{\%}$. Moreover, the system evaluation shows that our design is robust against differences in the clothing worn and time of day, which further verifies the successful performance of our system.

Device-free Passive (DfP) detection has received increasing attention for its ability to support various pervasive applications. Instead of relying on variable Received Signal Strength (RSS), most recent studies rely on finer-grained Channel State Information (CSI). However, existing methods have some limitations, in that they are effective only in the Line-Of-Sight (LOS) or for more than one moving individual. In this paper, we analyze the human motion effect on CSI and propose a novel scheme for Robust Passive Motion Detection (R-PMD). Since traditional low-pass filtering has a number of limitations with respect to data denoising, we adopt a novel Principal Component Analysis (PCA)-based filtering technique to capture the representative signals of human motion and extract the variance profile as the sensitive metric for human detection. In addition, existing schemes simply aggregate CSI values over all the antennas in MIMO systems. Instead, we investigate the sensing quality of each antenna and aggregate the best combination of antennas to achieve more accurate and robust detection. The R-PMD prototype uses off-the-shelf WiFi devices and the experimental results demonstrate that R-PMD achieves an average detection rate of 96.33% with a false alarm rate of 3.67%.

Coverage is an important issue in the area of wireless sensor networks, which reflects the monitoring quality of the sensor networks in scenes. Most sensor coverage research focuses on the ideal two-dimensional (2-D) plane and full three-dimensional (3-D) space. However, in many real-world applications, the target field is a 3-D complex surface, which makes conventional methods unsuitable. In this paper, we study the coverage problem in directional sensor networks for complex 3-D terrains, and design a new surface coverage algorithm. Based on a 3-D directional sensing model of nodes, this algorithm employs grid division, simulated annealing, and local optimum ideas to improve the area coverage ratio by optimizing the position coordinates and the deviation angles of the nodes, which results in coverage enhancement for complex 3-D terrains. We also conduct extensive simulations to evaluate the performance of our algorithms.