Tag anticollision has long been an important issue in RFID systems. To accelerate tag identification, some researchers have recently adopted bit tracking technology that allows the reader to detect the locations of collided bits in a collision slot. However, these methods still encounter the problem of too many collisions occurring at the beginning of identification. This paper proposes an optimal query tracking tree protocol (OQTT) that tries to separate all of the tags into smaller sets to reduce collisions at the beginning of identification. Using bit tracking technology, OQTT mainly adopts three proposed approaches, bit estimation, optimal partition, and query tracking tree. Bit estimation first estimates the number of tags based on the locations of collided bits. Optimal partition then determines the optimal number of the initial sets based on this estimation. Query tracking tree splits a set of collided tags into two subsets using the first collided bit in the tag IDs. This paper analyzes the efficiency of OQTT, which represents how many tags can be identified in a slot. Results show that its efficiency is close to 0.614, the highest efficiency published to date. The simulation results further show that OQTT outperforms other existing algorithms.

No abstract

Traditional methods for demand forecasting only focus on modeling the temporal dependency. However, forecasting on spatio-temporal data requires modeling of complex nonlinear relational and spatial dependencies. In addition, dynamic contextual information can have a significant impact on the demand values, and therefore needs to be captured. For example, in a bike-sharing system, bike usage can be impacted by weather. Existing methods assume the contextual impact is fixed. However, we note that the contextual impact evolves over time. We propose a novel context integrated relational model, Context Integrated Graph Neural Network (CIGNN), which leverages the temporal, relational, spatial, and dynamic contextual dependencies for multi-step ahead demand forecasting. Our approach considers the demand network over various geographical locations and represents the network as a graph. We define a demand graph, where nodes represent demand time-series, and context graphs (one for each type of context), where nodes represent contextual time-series. Assuming that various contexts evolve and have a dynamic impact on the fluctuation of demand, our proposed CIGNN model employs a fusion mechanism that jointly learns from all available types of contextual information. To the best of our knowledge, this is the first approach that integrates dynamic contexts with graph neural networks for spatio-temporal demand forecasting, thereby increasing prediction accuracy. We present empirical results on two real-world datasets, demonstrating that CIGNN consistently outperforms state-of-the-art baselines, in both periodic and irregular time-series networks.

No abstract

No abstract