Dear Eduard Babulak,

I am of the opinion that future universal and adaptive system solutions to guarantee the security, safety and energy management of Smart Homes /Buildings will be based on the use of Home /Building Digital Twins (H/ B DTs). When using H /B DTs, various criteria/guidelines for guaranteeing safety and security can be defined and used. See my literature sources.

Best regards

Anatol Badach

Zhansheng Liu , Anshan Zhang, Wensi Wang: A Framework for an Indoor Safety Management System Based on Digital Twin; Sensors, Vol. 20, Issue 20, Oct 2020 https://doi.org/10.3390/s20205771

V Gopinath, A Srija, C Neethu Sravanthi: Re-design of smart homes with digital twins; International conference on computer vision and machine learning; IOP Conf. Series; Journal of Physics: Conf. Series 1228, 2019; DOI:10.1088/1742-6596/1228/1/012031 https://iopscience.iop.org/article/10.1088/1742-6596/1228/1/012031/pdf

Abhilash Gopalakrishnan: „Chapter 1: Demystifying Digital Twin“; Dec 2020 https://www.researchgate.net/publication/348051656

Good question

Hi

I consider in smart home confidentiality, integrity and availability are the main topics to take in consideration. But, I think resilience is other key factor because if the smart home system is not capable to re-action to attacks, basic operations such as smart locks, turn on/off of lights or hvac systems don`t operate. If automatation of smart home is high the impact of cyber attacks is relevant. Additionality, I think smart home now has sensibility information like medical information, so the privacy is other relevant factor. Risk analysis (magerit, Nist) and security verification (OWASP ASVS) are steps for design and enhacing of security on smart home.

Conference Paper On the Road to Secure and Privacy-Preserving IoT Ecosystems

https://ieeexplore.ieee.org/abstract/document/8860028

There are new AI application in Cyber Security that will bring it to next level of safety and resilience.

This work has presented an intelligent home automation system based on IoT technologies, cloud computing, and a machine learning algorithm. The home automation system allows remote and local control of the home via an Android-based mobile application. The system controls electrical home appliances, monitors environmental conditions through temperature, humidity, and light sensors, and ensures home security through a motion sensor and an IoT camera. The system makes intelligent decisions to automatically turn ON or OFF lights and allow the user to view a picture of the person captured by the camera and decide whether to save it. The system is scalable, and the application allows additional appliances with an extra configured point, simply named outlet in our work. Also, the mobile application uses a real-time cloud database to record data gathered. It also displays the trend of the readings graphically on the screen of the mobile application. These data can be analyzed and used for future prediction. The next phase of the work will be to use a dataset of several real-life images captured by the camera from the system or sourced from relevant datasets in the field of home automation and train a machine learning algorithm to classify them so that the system could differentiate between an intruder and the home occupants. Also, we plan to enhance the system such that the user will only receive captured images of an intruder, and other captured images will be saved in real-time into the cloud-based database of the system. This will remove the limitation on the encumbrance of a load of images on the mobile application. In addition, machine learning will also be used to enhance the PIR sensor’s performance to differentiate between an animal, the occupant, and an intruder. This will reduce the notification from the PIR sensor to the user. The machine learning algorithm will also be applied to classify detected images to work with several occupants and a larger data set of intruders.