Athlete body fat rate monitoring and motion image simulation based on SDN data center network and sensors

Abstract

With the development of artificial intelligence technology, new software is also emerging in an endless stream. On the basis of sensors, the new software realizes the separation of network control layer and data layer, thereby improving network throughput and link utilization. With the gradual maturity of deep reinforcement learning technology, the redefined network architecture can be managed and controlled through software, making the network evolve toward a more intelligent direction. By providing data support for the intelligent control of the network, the network controller can obtain the data transmission status in real time, so that more ideas can become reality. Now, on the basis of motion sensors, through data fusion technology, athletes' physical conditions can be planned more effectively, so as to achieve scientific management and reasonable planning, obtain more accurate body fat rates, and customize corresponding data flow routing strategies., To achieve the combination of technology and technology. This paper proposes a scheduling strategy based on machine learning, combining the reinforcement learning algorithm in machine learning and deep reinforcement learning algorithm, setting the key factors of reinforcement learning, and applying it to real-time sports images of athletes, combining the sports characteristics of athletes, Set the action and reward value. Then use the algorithm to allocate a reasonable path for data transmission according to the real-time status to reduce network delay. This article will use sensor technology and data center network to provide a new method for athletes' real-time motion images and body fat percentage.