DOI: 10.1192/j.eurpsy.2026.11346 ISSN: 0924-9338

Adaptive Federated Smart Homes with In-Home Displays for Mental Wellness and Sustainable Lifestyle Enhancement

A. Oveysi Kian, M. Oveysi Kian
Show PDF Cite

Introduction

Conventional smart home platforms focus primarily on energy efficiency and comfort optimization, often overlooking the psychosocial dimensions of well-being. Recent advances in Federated Learning (FL) and edge computing enable privacy-preserving behavioral modeling and decentralized adaptation. Within this context, In-Home Displays (IHDs) provide an interactive channel for real-time feedback, supporting self-awareness and behavioral self-regulation. However, few studies have examined how adaptive federated intelligence can actively promote mental wellness and lifestyle consistency within everyday domestic settings. This study addresses this gap by proposing an Adaptive Federated Smart Home framework that couples federated personalization with behavioral feedback loops, aiming to improve psychological indices such as the MWI, SRI, and LCS, while SEF.

Objectives

The specific aims are: To design a multi-layer adaptive architecture integrating federated learning, local behavioral adaptation, and IHD-based feedback.To quantify behavioral and mental wellness improvements using normalized indices (SRI, MWI, LCS, SEF).To evaluate the feasibility, behavioral trends, and potential effect sizes in a six-week pilot study across 15 smart homes.

Methods

A four-layer architecture was implemented, comprising the Cloud (Federated Aggregation), Local Adaptive Edge, Sensing, and In-Home Display Feedback layers. Each local node (smart home) learned user-specific behavioral features and fine-tuned the global federated model received from the cloud, preserving data privacy. Environmental sensors captured parameters such as light exposure, noise level, and temperature, while mood and wellness self-reports were collected through the IHD interface. The IHD also provided real-time behavioral feedback, visualizing daily metrics and trends. Behavioral indices were normalized to [0, 1], and weekly progress was evaluated across SRI, MWI, LCS, and SEF.

Results

Table 1.

Table 1. Long description.

Metric
Baseline Mean
After 6 Weeks
Change (%)
Sleep Regularity Index (SRI)
0.64
0.81 +26.6
Mental Wellness Index (MWI)
0.71 0.87 +22.5
Lifestyle Consistency Score (LCS)
0.68 0.79 +16.2
Stress Event Frequency (SEF)
0.56 0.46 −18.0
Image 1:

Image 1: Long description.

Image 2:

Image 2: Long description.

Image 3:

Image 3: Long description.

Conclusions

This feasibility study demonstrates that adaptive federated smart homes can foster psychological well-being and sustainable living through personalized, privacy-preserving behavioral feedback. The synergy of edge intelligence and federated coordination enables local adaptation without compromising data confidentiality, offering a scalable pathway for real-world deployment.

Disclosure of Interest

None Declared

More from our Archive