
Fuzzy Reinforcement Learning Model for Resource Adaptation in IoT
Author(s) -
Daneshwari I. Hatti,
Ashok V. Sutagundar
Publication year - 2021
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.a5824.0510121
Subject(s) - computer science , adaptation (eye) , reinforcement learning , quality of service , fuzzy logic , distributed computing , process (computing) , task (project management) , automation , artificial intelligence , machine learning , computer network , systems engineering , engineering , mechanical engineering , physics , optics , operating system
The automation of several applications is creating engrossment in Internet of Things (IoT). The prerequisite for employing IoT in daily life is the ability to interact with devices technologies and process the sensed data. The difficulty to process the sensed data with scarce resources for diverse requests is challenging. Learning the behaviour of changing demand and processing with available resources has resulted in adaptation policy. Adaptation policy by employing Reinforcement learning and Fuzzy logic is proposed to adapt the resources for executing the tasks. Prioritization of task, allocating of resources to the tasks by adapting with available resources with assured Quality of Service (QoS) is performed. Fuzzy Q learning Adaptation Algorithm (FQAA) is designed for evaluating resource adaptation mechanisms to execute the heterogenous tasks. The algorithm with different configuration is simulated using Ifogsim and python. It is compared with traditional method that is without adaptation mechanism, performs better compared to other algorithms in terms of Cost, energy consumption and latency.