Owing to the no free lunch theorem, no single optimisation algorithm can solve all optimisation problems accurately, so new optimisation techniques are required. In this paper, a novel metaheuristic called the deep sleep optimiser (DSO) is proposed. The deep sleep optimiser mimics the sleeping patterns of humans to solve optimisation problems. The DSO is modelled on the rise and fall of homeostatic pressure during the human sleep process. Human sleep is often modelled on the four sleep stages and the deep sleep stage is employed in this work. The mathematical model of sleep homeostatic pressure is employed to simulate and determine the deep sleep state. The performance of DSO is demonstrated by employing 23 traditional functions (i.e., unimodal, multimodal, and fixed multi-modal functions), six composite functions, three engineering design problems, two knapsack problems, and six widely known travelling salesman’s problems. Additionally, the performance is evaluated in terms of accuracy, computational running time, the Wilcoxon rank sum, and the Friedman test. Lastly, the DSO is compared with 11 other metaheuristics, including GA, PSO, TLBO, and GWO. The DSO fares comparably well and, in most instances, it outperforms other metaheuristics.
The Software-Defined Networking technology promises to enhance network performance and cost reduction for service providers by providing scalability, flexibility, and programmability through the separation of the control plane from the data plane. However, the separation between the control plane and data plane in the implementation of SDN presents scalability issues, as the controller has limited computational resources. To address the SDN scalability issues identified, we create a scalable hybrid switching solution using machine learning algorithms. We propose an SDN OpenFlow model switch which collaborates with the traditional switch to represent a scalable framework of Hybrid Routing with Reinforcement Learning (sHRRL). We implement a reinforcement algorithm to randomly explore new routes and discover the most optimal path through the Q-learning algorithm. This primitive and model-free form of reinforcement learning utilizes the Markov Decision Process and Bellman’s equation to reiteratively update Q-values in the Q-table for every transition in the network environment state until Q-function has converged to the best Q-values. The proposed hybrid switching model is benchmarked against the standard SDN OpenFlow switch in terms of network performance metrics, including throughput, packet exchange transmission rates, CPU load, and delay. When statistically comparing simulation results, it is evident that the proposed switching model, incorporating machine learning algorithms, can effectively tackle scalability challenges in the design of SDN controller networks, especially in Data Centre environments where rapid switching speeds are crucial.
In this study, the specific absorption rate (SAR) and exposure index (EI) of access points (APs) and user equipment (UEs) in fourth-generation (4G) and fifth-generation (5G) wireless technologies are examined with regard to the effects of exposure to radiofrequency (RF) electromagnetic fields (EMF) radiation and the implications of their reduction. We characterize the EI using a classical mathematical method while considering the power density, the SAR, the electric field strength, and the tissue's density and conductivity. As such, a novel exposure-index open-loop power control algorithm is proposed to evaluate the realistic RF-EMF radiation exposure on human users from both the downlink (DL) and uplink (UL) communication devices. To solve an EI minimization problem using the open-loop power control algorithm, we formulate it in the form of a mixed-integer nonlinear programming (MINLP) problem. As the energy capacity (i.e., power density) in wireless networks determines the radiation exposure (SAR and EI), it minimizes the EI by controlling and managing the transmitted and received powers under the restrictions of Quality of Service (QoS), interference, and power, while ensuring the users' QoS requirements are met. Our proposed scheme is numerically compared to other heuristic algorithms and exposure limits established by the International Commission on Non-Ionizing Radiation Protection (ICNIRP) and other similar organizations. Lastly, we compare the emissions from 4G and 5G networks to the emissions from UL and DL transmissions. Our simulation findings indicate that our proposed technique is a good alternative. Our assessment, in terms of numerical results and evaluation, also verifies that the exposures are bearable, fall within the recommended limits, and are minimized without impairing the users' QoS.
This paper presents the impact of radiofrequency (RF) electromagnetic fields (EMF) radiation exposure induced by wireless networks, most importantly 5G cellular networks for both the uplink and downlink radio emissions using exposure index-based power control algorithm (EIPCA), a novel simulation method that quantifies the realistic electromagnetic exposure of the human user. The exposure index (EI) is used to characterize the EMF exposure taking in account the power density, specific absorption rate (SAR); using both the electric field strength and magnetic field strength as well as considering the variability of other factors such as environment, the conductivity and the mass density of tissue. This work aims at simulating the radiations emitted from access points (APs) and mobile devices, analyzing and comparing them with the threshold set by the International Commission on Non-Ionizing Radiation Protection (ICNIRP) for the understanding of radiation impact. The numerical results reveal that the maximum radiation exposure emitted is far lower than the ICNIRP standard. It is shown that the exposure from mobile devices (uplink) has more EI compared with the ones due with the transmitting stations (downlink) radio emissions, and both can be minimized when there is an optimal power control scheme in the network as revealed in the power density received from the APs. Moreover, it compares the exposure level in both the fourth-generation (4G) and the fifth-generation (5G) networks, this will increase the health awareness concerning radiation exposure of 5G networks to the general public.
This paper addresses the dynamic resource allocation in a multi-slice multi-tier multi-domain based network with different network players. The dynamic resource allocation problem is formulated as a maximum utility optimisation problem from multi-player multi-domain perspective. Furthermore, a 3-level hierarchical business model comprising infrastructure providers (InPs), mobile virtual network operators (MVNO), service providers (SP) and slice users is investigated. We propose a multi-tier multi-domain slice user matching game and a distributed backtracking multi-player multi-domain game schemes in solving the transformed maximum utility optimisation problem. We compare the multi-tier multi-tenant multi-domain game scheme with a genetic algorithm (GA) intelligent latency- aware resource allocation scheme (GI-LARE) and, a static slicing resource allocation scheme via Monte Carlo simulation. Our findings reveal that the proposed scheme outperforms these other schemes.