scholarly journals A Federated Learning Approach to Anomaly Detection in Smart Buildings

2021 ◽  
Vol 2 (4) ◽  
pp. 1-23
Author(s):  
Raed Abdel Sater ◽  
A. Ben Hamza
Keyword(s):  
Anomaly Detection ◽  
Learning Approach ◽  
Temporal Data ◽  
Energy Usage ◽  
Privacy By Design ◽  
Smart Buildings ◽  
Smart Building ◽  
Training Cost ◽  
Real World Datasets ◽  
Short Time

Internet of Things (IoT) sensors in smart buildings are becoming increasingly ubiquitous, making buildings more livable, energy efficient, and sustainable. These devices sense the environment and generate multivariate temporal data of paramount importance for detecting anomalies and improving the prediction of energy usage in smart buildings. However, detecting these anomalies in centralized systems is often plagued by a huge delay in response time. To overcome this issue, we formulate the anomaly detection problem in a federated learning setting by leveraging the multi-task learning paradigm, which aims at solving multiple tasks simultaneously while taking advantage of the similarities and differences across tasks. We propose a novel privacy-by-design federated learning model using a stacked long short-time memory (LSTM) model, and we demonstrate that it is more than twice as fast during training convergence compared to the centralized LSTM. The effectiveness of our federated learning approach is demonstrated on three real-world datasets generated by the IoT production system at General Electric Current smart building, achieving state-of-the-art performance compared to baseline methods in both classification and regression tasks. Our experimental results demonstrate the effectiveness of the proposed framework in reducing the overall training cost without compromising the prediction performance.

scholarly journals Assessing the smartness of buildings

Facilities ◽  
2015 ◽  
Vol 33 (9/10) ◽  
pp. 553-572 ◽  
Author(s):  
David Arditi ◽  
Giulio Mangano ◽  
Alberto De Marco
Keyword(s):  
Questionnaire Survey ◽  
Design Methodology ◽  
Years Of Experience ◽  
Smart Buildings ◽  
Content Type ◽  
Smart Building ◽  
Convenient Tool ◽  
Energy Issue

Purpose – This study aims at capturing the perspectives of construction professionals into a classified taxonomy of the various characteristics of smart buildings and at developing an index able to define their level of smartness. Design/methodology/approach – A questionnaire survey has been administrated to construction professionals in the service of designers, constructors and owners. Results have been analyzed with the Kruskal–Wallis test and they have been used to develop a smartness index. Findings – Designers and owners are more focused on the energy issue than constructors. The energy captures the attention of practitioners with less years of experience, confirming that the awareness of the energy topic is rather recent. Originality/value – The main characteristics of smart buildings have been structured in domains and subdomains. Their importance has been rated by construction professional and a smartness index for smart building has been developed to provide with a convenient tool for evaluation and benchmarking.

scholarly journals Unsupervised Anomaly Detection with Distillated Teacher-Student Network Ensemble

Entropy ◽  
2021 ◽  
Vol 23 (2) ◽  
pp. 201
Author(s):  
Qinfeng Xiao ◽  
Jing Wang ◽  
Youfang Lin ◽  
Wenbo Gongsa ◽  
Ganghui Hu ◽  
...  
Keyword(s):  
Anomaly Detection ◽  
Multivariate Data ◽  
Failure Detection ◽  
Superior Performance ◽  
Detection Algorithms ◽  
Teacher Student ◽  
Model Complex ◽  
Unsupervised Anomaly Detection ◽  
Real World Datasets ◽  
Complex Features

We address the problem of unsupervised anomaly detection for multivariate data. Traditional machine learning based anomaly detection algorithms rely on specific assumptions of normal patterns and fail to model complex feature interactions and relations. Recently, existing deep learning based methods are promising for extracting representations from complex features. These methods train an auxiliary task, e.g., reconstruction and prediction, on normal samples. They further assume that anomalies fail to perform well on the auxiliary task since they are never trained during the model optimization. However, the assumption does not always hold in practice. Deep models may also perform the auxiliary task well on anomalous samples, leading to the failure detection of anomalies. To effectively detect anomalies for multivariate data, this paper introduces a teacher-student distillation based framework Distillated Teacher-Student Network Ensemble (DTSNE). The paradigm of the teacher-student distillation is able to deal with high-dimensional complex features. In addition, an ensemble of student networks provides a better capability to avoid generalizing the auxiliary task performance on anomalous samples. To validate the effectiveness of our model, we conduct extensive experiments on real-world datasets. Experimental results show superior performance of DTSNE over competing methods. Analysis and discussion towards the behavior of our model are also provided in the experiment section.

Detecting cyberattacks using anomaly detection in industrial control systems: A Federated Learning approach

2021 ◽  
Vol 132 ◽  
pp. 103509
Author(s):  
Truong Thu Huong ◽  
Ta Phuong Bac ◽  
Dao Minh Long ◽  
Tran Duc Luong ◽  
Nguyen Minh Dan ◽  
...  
Keyword(s):  
Anomaly Detection ◽  
Control Systems ◽  
Learning Approach ◽  
Industrial Control ◽  
Industrial Control Systems

Anomaly Detection in Streaming Nonstationary Temporal Data

2019 ◽  
Vol 29 (1) ◽  
pp. 13-27 ◽  
Author(s):  
Priyanga Dilini Talagala ◽  
Rob J. Hyndman ◽  
Kate Smith-Miles ◽  
Sevvandi Kandanaarachchi ◽  
Mario A. Muñoz
Keyword(s):  
Anomaly Detection ◽  
Temporal Data

scholarly journals Grid-Based Routing Model for Energy Efficient and Secure Data Transmission in WSN for Smart Building Applications

2021 ◽  
Vol 11 (22) ◽  
pp. 10517
Author(s):  
V. Sivasankarareddy ◽  
G. Sundari ◽  
Ch. Rami Reddy ◽  
Flah Aymen ◽  
Edson C. Bortoni
Keyword(s):  
Routing Protocol ◽  
Data Transmission ◽  
Energy Efficient ◽  
Residual Energy ◽  
Smart Buildings ◽  
Smart Building ◽  
Secure Data ◽  
Secure Data Transmission ◽  
Average Residual ◽  
Grid Based

Presently, due to the establishment of a sensor network, residual buildings in urban areas are being converted into smart buildings. Many sensors are deployed in various buildings to perform different functions, such as water quality monitoring and temperature monitoring. However, the major concern faced in smart building Wireless Sensor Networks (WSNs) is energy depletion and security threats. Many researchers have attempted to solve these issues by various authors in different applications of WSNs. However, limited research has been conducted on smart buildings. Thus, the present research is focused on designing an energy-efficient and secure routing protocol for smart building WSNs. The process in the proposed framework is carried out in two stages. The first stage is the design of the optimal routing protocol based on the grid-clustering approach. In the grid-based model, a grid organizer was selected based on the sailfish optimization algorithm. Subsequently, a fuzzy expert system is used to select the relay node to reach the shortest path for data transmission. The second stage involves designing a trust model for secure data transmission using the two-fish algorithm. A simulation study of the proposed framework was conducted to evaluate its performance. Some metrics, such as the packet delivery ratio, end-end delay, and average residual energy, were calculated for the proposed model. The average residual energy for the proposed framework was 96%, which demonstrates the effectiveness of the proposed routing design.

scholarly journals A Deep Learning Approach to Anomaly Detection in the Gaia Space Mission Data

2019 ◽  
pp. 390-401
Author(s):  
Alessandro Druetto ◽  
Marco Roberti ◽  
Rossella Cancelliere ◽  
Davide Cavagnino ◽  
Mario Gai
Keyword(s):  
Deep Learning ◽  
Anomaly Detection ◽  
Space Mission ◽  
Learning Approach

scholarly journals Smart energy monitoring based on IoT Lora-Wan on the campus buildings: Case study Indonesian Defence University (Unhan), Sentul, Bogor

2021 ◽  
Vol 878 (1) ◽  
pp. 012065
Author(s):  
S Ramadhan ◽  
L Lisapaly ◽  
D Boesrony
Keyword(s):  
Interconnection Networks ◽  
Electrical Energy ◽  
Physical Structure ◽  
Maximum Efficiency ◽  
Smart Devices ◽  
Energy Usage ◽  
Smart Building ◽  
Energy Meter ◽  
Campus Building

Abstract Smart building constructions such as Campus Buildings have been designed for use, where the physical structure and system components are interrelated and can maximize functionality for operation and maintenance. So that the Campus building can be used with a longer age. One of the sub-systems that can monitor and notify the range of energy usage on a campus building is a smart electrical energy meter (kWh meter), which is connected to all devices that consume electrical energy in campus buildings. These interconnected smart devices use IoT (Internet of Things) interconnection networks and low power wireless technology (Lora). In a case study of the use of this system at the Indonesian Defense University, Unhan, Sentul, Bogor, West Java, it can be seen how the maximum efficiency in the use of electrical energy can be obtained in the smart campus building construction, which runs automatically.

Sign in / Sign up

Export Citation Format

Share Document