Real-Time Data Cloud Transmission and Early Warning Algorithm for Outdoor Sports with Smart Glasses

Abstract

Through the review analysis, it is evident that most of the current big data mining methods are based on known abnormal characteristics for big data mining. Existing algorithms ignore relevant information and prior information, which reduces the reliability and efficiency of big data mining and increases the overhead of processing big data, resulting in a decrease in the overall availability and performance of big data. Hence, real-time data cloud transmission and the early warning algorithm for outdoor sports with smart glasses is studied. This research study presents 2 aspects of novelty: (1) For the data cloud transmission, the UDT is selected, the application program also uses the UDT socket interface to transmit data, and the UDT calls UDP through the Socket interface provided by the operating system. Then, the OBEX is combined to improve the efficiency. (2) For the early warning algorithm, we consider using the grid space as the clustering area to then reduce the time spent on retrieving the clustering area, ensuring high-precision picking of the clustering center and greatly improving the efficiency. Through the comparison simulation under different data sets, the missing and false alarm tests are both satisfactory.