AI-DRIVEN MULTI-OBJECTIVE OPTIMIZATION FOR ENERGY-EFFICIENT AMBIENT ASSISTED LIVING IN SMART HEALTHCARE IOT

Abstract

Ambient Assisted Living (AAL) systems make use of internet connected devices, wearable sensors, smart home technologies, and smart computing services to help elderly patients living safely and independently. However continuous monitoring and transferring of data greatly raises the levels of power consumption, communication overhead, and delay time for response, especially when many devices with different characteristics are working together. Our proposed method is an AI-based framework for energy-efficient AAL in smart healthcare IoT. Our solution makes use of artificial intelligence to recognize elderly person's activities and health-related information and offers an Adaptive Multi-Objective Honeybee Algorithm to choose the methods of monitoring, communication, and processing automatically. The method works in terms of energy consumption, delay, communication costs, monitoring effects simultaneously, providing the coverage of the most important health events. Based on the identified context, the framework can adapt the number of active sensors and computing resources without keeping all devices operating at their maximum capacities. The simulation-based experiment shows how well our method performs in comparison to the already known multi-objective optimization methodologies like NSGA-II, MOPSO, and standard Honeybee Algorithm.