The transition toward Healthcare Industry 5.0 requires personalized, intelligent, and privacy-preserving solutions. Additionally, the proliferation of home monitoring in the current Healthcare 4.0 framework has been embracing the challenge for invasiveness/privacy issues brought by home monitoring sensors and devices, potentially forsaking the conceived human-centric attributes of Smart Healthcare Industry 5.0. At the intersection of Healthcare Industry 5.0 and secured home monitoring, we propose a multi-service non-invasive solution for distributed home monitoring. A new infrastructure based on Fiber Bragg Grating (FBG) accelerometers is designed, realizing non-invasive/high-privacy distributed home monitoring by solely monitoring floor vibrations. To achieve home monitoring based on the low-dimensional data collected by FBG accelerometers, a multi-class hierarchical support vector machine-based algorithm (H-SVM) is proposed, which achieves simultaneous multi-service classification with FBG noise/interference mitigation. Moreover, an enhanced time difference of arrival (TDoA)-based algorithm (E-TDoA) is proposed for real-time localization of users, with the enhanced accuracy specifically based on FBG-collected data. Experimental results show an overall accuracy of 93.62% for multi-service classification, with a low processing time of 4.1 ms, meanwhile comprising a localization accuracy of 0.715 m (sufficient for home monitoring multi-service delivery).