Wireless distributed computing with cognitive radio Energy Harvesting Network

Abstract

Energy Harvesting Network (EHN) is a promising technology that offers a viable solution for prolonging the lifetime of energy-constrained radio nodes. However, the EHNs are subject to energy causality constraint. Therefore, processing bulky task by EHN is expected to burden the limited resources of energy harvesters by draining the stored energy and thereby reaching rapidly to energy causality constraint. In such scenario, energy harvesters flip into sleep mode and thereby the execution time of the next task will be delayed until the energy harvesters revert back into active mode. To tackle this problem, this research work proposes an Intelligent Computing Energy Harvesting Network (ICEHN) that employs Wireless Distributed Computing (WDC) network in order to maximize the energy savings and to overcome the challenge of processing intensive computationally complex tasks by energy harvesting radio nodes. The ICEHN algorithm automates the decision making process of energy harvesters to decide whether to execute the task locally or to distribute it among a set of collaborating nodes wirelessly. Therefore, the decision making process of ICEHN is formulated by using Constrained Partially Observable Markov Decision Process (CPOMDP) to act under uncertain conditions, e.g. imperfect CSI and uncertain data packets levels and energy arrival. Furthermore, an Intelligent and Adaptive Task Allocation (IATA) algorithm is proposed to distribute the workload adaptively among the collaborating nodes. In addition, a Myopic ICEHN (MICEHN) policy is proposed to reduce the computational complexity of the ICEHN algorithm. On the other hand, Intelligent Computing Cognitive Radio-Energy Harvesting Network (ICCR-EHN) algorithm is proposed by applying a Cognitive Radio (CR) application, i.e. FFT time smoothing algorithms, onto the ICEHN algorithm. Finally, the proposed algorithms are justified by mathematical demonstrations and simulation programs as well as compared against the conventional research works on EHN, CR-EHN, WDC network and cloud computing. Based on the simulation results, the ICEHN algorithm is found to perform better than the traditional EHN in terms of energy by 9.1% and 27.3% in terms of delay. Meanwhile, the ICCR-EHN algorithm is found to outperform the utilization of energy detector in the previous research works on CR-EHN by 13.9% and 83.6%, in terms of probability of detection and probability of false alarm, respectively.