Design and optimization of intelligent sports IoT system based on big data and quantum computing

Abstract

In the digital age, technological advancements have transformed athletic training, competition, and performance analysis. The growing demand for better training techniques has resulted in the integration of big data, AI, IoT, and quantum computing, all of which play critical roles in human action recognition (HAR) for improving athlete performance and safety. With an increasing volume of data generated by various sensors in activity fields, efficient analysis becomes difficult. Conventional techniques frequently fall short of processing large-scale data, resulting in inefficiencies in real-time performance evaluation and decision-making. The Efficient Gazelle Optimized Dynamic Support Vector Machine (EGO-DSVM) is a novel model being developed to improve the accuracy and efficiency of activity recognition. It focuses on optimizing sensor data analysis, enhancing feature extraction, and using quantum computing to explore dynamic search spaces. The EGO-DSVM model uses a convolutional neural network (CNN) for feature extraction, which is tuned by the EGO algorithm for optimal hyperparameters. Quantum computing is utilized to represent gazelles' positions in quantum space, which improves dynamic exploration and optimization. Preprocessing techniques are used to improve data quality by eliminating noise from sensor-generated information. The proposed EGO-DSVM model surpassed conventional models in identifying six activity features. For the "Upstairs" activity, it attained an accuracy of 93.10%, precision of 90.18%, recall of 95.14%, and F1-score of 82.58%. These findings show significant improvements over baseline models such as traditional SVM, CNN, and transformer-based HAR techniques. The EGO-DSVM model improves activity recognition performance by combining big data approaches and quantum computing. The integration of these technologies establishes a new standard for real-time, scalable HAR, providing athletes with improved performance analysis and training methodologies.