Intelligent Diversified Smart Elderly System from the Perspective of AI+ and IoT 
Li Yan  Bu Yanbin  Xing Tianxia  Xu Xiaoke  Wang Ziqing  Han Yutong  Tu Huanqi  Tian Rongxi

Communication University of China Nanjing, Nanjing Jiangsu Province, 211100；

Abstract：Driven by the accelerating trend of population aging and the iterative advancement of digital technology, traditional elderly care models face bottlenecks such as insufficient service precision and lagging response times. This paper proposes an intelligent diversified smart elderly system deeply integrated with AI and IoT. By constructing a full-chain technical system of "Perception-Transmission-Decision-Application," it integrates multimodal perception, intelligent decision-making, and scenario-based service functions to provide diversified support for health management, safety protection, daily life services, and emotional companionship. The system is characterized by passive sensing and active intervention, breaking through the limitations of traditional "passive care" and building a linked ecosystem involving "Family-Community-Institution-Medical care." The research elaborates on the system architecture design, core technology integration paths, and application logic, providing engineering references for the implementation of smart elderly care technologies.

Keywords: Smart Elderly System; Multimodal Perception; Active Intervention

References

[1]Antunes, Mário, Gomes D , Aguiar R .Towards IoT data classification through semantic features[J].Future Generation Computer Systems, 2017, 86(SEP.):792-798.

[2]Zhang L .Landscape layout characteristics and evaluation of smart buildings based on deep learning algorithms[J].Discover Internet of Things, 2025, 5(1):1-21.

[3] Mokhov A , Yakovlev A .Conditional Partial Order Graphs: Model, Synthesis, and Application[J].IEEE Transactions on Computers, 2010, 59(11):1480-1493.

[4] Yang X , Liu J , Wu Q ,et al.Big Data Dynamic Aggregation and Intelligent Service Model for Multimodal Healthcare and Eldercare[J].Journal of Library & Information Science in Agriculture, 2025, 37(4).