An Intelligent Optimization Strategy Based on Deep Reinforcement Learning for Step Counting

Abstract

With the popularity of Internet of things technology and intelligent devices, the application prospect of accurate step counting has gained more and more attention. To solve the problems that the existing algorithms use threshold to filter noise, and the parameters cannot be updated in time, an intelligent optimization strategy based on deep reinforcement learning is proposed. In this study, the counting problem is transformed into a serialization decision optimization. This study integrates the noise recognition and the user feedback to update parameters. The end-to-end processing is direct, which alleviates the inaccuracy of step counting in the follow-up step counting module caused by the inaccuracy of noise filtering in the two-stage processing and makes the model parameters continuously updated. Finally, the experimental results show that the proposed model achieves superior performance to existing approaches.

Wide-angle monostatic radar cross section enhanced metasurface based on deep deterministic policy gradient

“…In contrast, corner reflectors and Luneburg lenses have specific geometries and structural requirements, making their integration more challenging and limiting their adaptability to complex or irregular surfaces [21,22,23,24]. This paper designs a 1-bit reconfigurable element that achieves 0°or 180°degree reflection phase flipping by controlling PIN diodes through 0/1 coding [25,26,27,28]. A 12x12 reconfigurable reflection array that can be attached to the surface of smooth objects is introduced, and the combination of the DDPG algorithm and electromagnetic simulation software is utilized to explore different coding schemes.…”