Federated Learning (FL) is a transformative decentralized approach in machine learning and deep learning, offering enhanced privacy, scalability, and data security. This review paper explores the foundational concepts, and architectural variations of FL, prominent aggregation algorithms like FedAvg, FedProx, and FedMA, and diverse innovative applications in thermal comfort optimization, energy prediction, healthcare, and anomaly detection within smart buildings. By enabling collaborative model training without centralizing sensitive data, FL ensures privacy and robust performance across heterogeneous environments. We further discuss the integration of FL with advanced technologies, including digital twins and 5G/6G networks, and demonstrate its potential to revolutionize real-time monitoring, and optimize resources. Despite these advances, FL still faces challenges, such as communication overhead, security issues, and non-IID data handling. Future research directions highlight the development of adaptive learning methods, robust privacy measures, and hybrid architectures to fully leverage FL’s potential in driving innovative, secure, and efficient intelligence for the next generation of smart buildings.

The surveillance of indoor air quality is paramount for ensuring environmental safety, a task made increasingly viable due to advancements in technology and the application of artificial intelligence and deep learning (DL) tools. This paper introduces an intelligent system dedicated to monitoring air quality and categorizing activities within indoor environments using a DL approach based on 1D Convolutional Neural Networks (1D-CNNs). Our system integrates six diverse sensors to gather measurement parameters, which subsequently train a 1D CNN model for activity recognition. This proposed model boasts a lightweight and edge-deployable design, rendering it ideal for real-time applications. We conducted our experiments utilizing an air quality dataset specifically designed for Activity of Daily Living (ADL) classification. The results illustrate the proposed model's efficacy, achieving a remarkable accuracy of 97.00%, a minimal loss value of 0.15%, and a swift prediction time of 41 milliseconds.

Occupancy detection plays a pivotal role in optimizing energy consumption, enhancing occupant comfort, and enabling intelligent decision-making in smart building environments. While traditional centralized learning methods often compromise user privacy and scalability, federated learning (FL) offers a promising privacy-preserving alternative. In this study, we propose Federated Cascade Learning with Long Short-Term Memory and Attention (FCL-LSTMA), a novel decentralized architecture that synergistically combines FL and cascade learning to enable accurate and efficient occupancy detection across distributed environments. The FCL-LSTMA framework operates through a two-stage training process: initially, Block B1 leverages 1D convolution, batch normalization, and a parallel LSTM layer for foundational feature learning. Once stabilized, the block is frozen and extended to Block B2, which incorporates attention mechanisms and an additional LSTM layer to refine temporal feature extraction. This progressive training ensures improved stability, model depth, and generalization without compromising data privacy. Experimental results on four benchmark datasets demonstrate the superiority of the proposed framework. FCL-LSTMA achieves outstanding accuracy for binary occupancy detection: 99.99% on the Living Room dataset, 98.95% on the UCI Occupancy Detection dataset, 99.95% on both the UCI Room Occupancy Estimation and IEQ datasets. For multi-class tasks, the model achieves 99.98% and 99.95% on the Living Room and UCI Room Occupancy Estimation datasets, respectively. Furthermore, it demonstrates computational efficiency with inference times ranging from 0.2028 to 0.562 milliseconds and memory usage between 0.17 MB and 0.49 MB. These results confirm the robustness, scalability, and privacy-preserving capabilities of FCL-LSTMA, making it a compelling solution for real-time occupancy detection in smart building systems.

Recent advancements in technology, particularly in artificial intelligence and privacy-preserving tools, have facilitated the development of more sophisticated and secure approaches to addressing this challenge. This paper introduces FedWell, a novel privacy-preserving Federated occupant Stress monitoring approach for smart building environments. Our system integrates physiological data from smart yoga pillows (SaYoPillow) and wearable environmental sensors to train a lightweight, edge-deployable Artificial Neural Network (ANN) model. The FedWell framework ensures data privacy while enabling collaborative learning across distributed clients using Federated Averaging (FedAvg) aggregation. We conducted experiments using a comprehensive dataset of physiological parameters for stress level detection. The results demonstrate the global model’s exceptional performance, achieving 99.95% accuracy in stress level recognition, with precision of 99.95%, recall of 99.93%, and an F1-score of 99.94%. The global model exhibits a minimal loss of 0.0019% and a low communication cost of 0.08 Mb, highlighting its efficiency for real-time applications.

Pandemics place extraordinary pressure on healthcare systems, particularly in identifying and prioritizing high-risk groups such as the elderly, pregnant women, and individuals with chronic diseases. Existing Artificial Intelligence models often fall short, focusing on single diseases, lacking interpretability, and overlooking patient-specific vulnerabilities. To address these gaps, we propose an Explainable Deep Learning approach for identifying at-risk patients during pandemics (AI-RiskX). AI-RiskX performs risk classification of patients diagnosed with COVID-19 or related infections to support timely intervention and resource allocation. Unlike previous models, AI-RiskX integrates five public datasets (asthma, diabetes, heart, kidney, and thyroid), employs the Synthetic Minority Over-sampling Technique (SMOTE) for class balancing, and uses a hybrid convolutional neural network–long short-term memory model (CNN–LSTM) for robust disease classification. SHAP (SHapley Additive exPlanations) enables both individual and population-level interpretability, while a post-prediction rule-based module stratifies patients by age and pregnancy status. Achieving 98.78% accuracy, AI-RiskX offers a scalable, interpretable solution for equitable classification and decision support in public health emergencies.

This paper presents a novel privacy-preserving approach, Fed-CNN1D (Federated Convolutional Neural Network 1D), based on federated learning for monitoring air quality and classifying different activities of daily living in indoor spaces. The system employs six different types of sensors to collect measurement parameters, which are then used to train a 1D CNN model locally for activity recognition. The proposed model is lightweight and edge deployable, making it suitable for real-time applications. Experiments were conducted using an air quality dataset specifically curated for Activity of Daily Living (ADL) classification. The results demonstrate that our approach Fed-CNN1D achieves 96.50% accuracy, 96.27% F1-Score, 96.54% precision, 96.21% recall, 0.11% loss, with low communication cost of 0.099Mb, and a swift detection time of 15 ms.