With the widespread of Internet of Things (IoT), the wireless localization technology has provided various location-based applications and services. Radio frequency identification (RFID) has been widely used in objects identification and tracking. Combining with its excellent identification ability, RFID is also a promising technique for indoor localization. This paper provides the state-of-art of UHF RFID localization and gives an introduction of methods and applications. The major methods of UHF RFID localization have two types: range-based and range-free reported in previous literature. The range-based methods obtain the distance or angle between the antenna and tag utilizing the received signal strength indicator (RSSI), time information or signal phase and estimate tag location further combining the scheme geometrical characteristics. The methods based on fingerprinting and non-fingerprinting are mainly discussed in the range-free domain. Finally, a comparative study of these surveyed works are presented and concluded with a future trend.

In recent years, emerging intelligent algorithms have achieved great success in the domain of fault diagnosis due to effective feature extraction and powerful learning ability. However, the current models can only handle the data in Euclidean space, ignoring latent structure relationships of the signal, which can provide additional helpful information to distinguish diverse fault patterns. To address this issue, a graph convolution network (GCN) incorporating the weighted horizontal visibility graph (WHVG) is proposed for bearing faults diagnosis. The WHVG is utilized to transform time series to graph data from a geometric perspective. Edges are weighted by the difference between the sampling indexes to weaken the influence of remote nodes that are considered as noise. Furthermore, the graph isomorphism network (GIN) is improved as GIN+ to learn the graph representation and perform fault classification. Finally, the validity of WHVG and GIN+ is testified by three real-world bearing datasets. Meanwhile, the GIN+ model is compared with other machine learning models, multilayer perceptron (MLP), long short-term memory (LSTM), and two GCN models. The experimental results show that GIN+ boosts the performance and the internal structure relationships of the data contribute to the bearing faults diagnosis.

With the appearance of the new concepts “smarter planet,” “industry 4.0” and “self-service supermarket,” radio frequency identification (RFID), a key technology of Internet of Things, has been a hot research topic in recent years. Owing to the long read range, fast read speed and low cost, RFID in the ultrahigh frequency using passive tags has achieved great development in the management and tracking of asset and warehousing, personnel location, and robot navigation. In the meantime, there is a growing demand for the accurate localization indoors. In this paper, a method of similarity measurement of virtual reference tags using phase and receive signal strength indicator (RSSI) is designed to estimate the position of tags. The attributes of virtual reference tags, such as phase and RSSI, are calculated by the natural neighbor interpolation. Moreover, the Laiyite criterion with variable coefficient is presented to remove the multipath error. A 2-D platform is deployed to test the validity of the method. Compared with a k-nearest neighbor and multilayer perceptron, similarity measurement of virtual reference tags has a better performance for localization, which can meet the demand of the practical application.