Making the smart grid a reality will entail communication capabilities that do not exist in current power grids and their legacy control systems. New grid elements such as distributed energy generation and PHEVs will necessitate changes in these existing communication networks. These changes will be dictated by the need for new control operations that will allow efficient management of these new elements and that cannot be supported by the existing communication infra structure. Moreover, the smart grid will let households man age energy usage more efficiently through AMI and HEMSs. The realization of these systems will require pushing communication networks far out from the network core and deep into homes and buildings. We have presented the communication requirements of the smart grid, and we have identified the main technical challenges that need to be tackled. The only viable way to fulfill these requirements is to design a new communication architecture that can support smart grid services and control operations. This future communication architecture will need to take advantage of the recent progress made in communication technologies and protocols. It will require the introduction of communications technologies that were not considered in the past. In addition, this communication architecture will need to be reliable, scalable, and extendable to future smart grid services and applications.

Cognitive radio is considered as a new paradigm of designing wireless communication systems that aims at improving the spectrum utilization. Spectrum sensing is one of the most important elements in cognitive radio networks. Many spectrum sensing algorithms have been proposed in the literature. Among these algorithms, energy detection has been widely used in practice because of its computational and implementation simplicities. However, the performance of conventional energy detection is greatly deteriorating due to the impact of noise uncertainty, especially in low signal to noise ratio SNR environments. On the other hand, the inappropriate setting of the detection threshold could leads to a significant decline in the performance of detection. In this paper, we explore these situations and we propose an energy detection based optimal threshold scheme. We derived an optimal threshold level based on the tradeoff between misdetection probability and false alarm probability. We also developed an adaptive threshold factor with optimal algorithm in order to combat the noise uncertainty which is compatible with the real world communications and that allows gaining better spectrum sensing performance especially in low SNR.

The popularity of Wi-Fi-enabled devices alongside the growing demand for non-licensed spectrum, has made the Wi-Fi networks exceedingly congested. This endangers the efficiency of Wi-Fi and negatively affect the users' experience. The problem is especially pressing in dense areas (e.g. shopping centers) where Wi-Fi channel assignment is more likely to be uncoordinated and the working environment of Wi-Fi Access Points (APs) has become increasingly time-variant. As a result, the availability of Software-Defined Networking (SDN) and network virtualization technologies has motivated the use of centralized resource management as a solution. This paper provides an algorithm for channel assignment functionality in the context of SDN-based centralized resource management, which captures the live status of a Wi-Fi network and is capable of optimising the Radio Frequency (RF) channel assignment process. The APs' network arrangement, the current assignment of the channels and the characteristics of the RF channels in IEEE 802.11 have all been taken into account in the proposed model. The performance of the proposed model in terms of the level of the interference, the spectral efficiency at each AP and the Signal to Interference plus Noise Ratio (SINR) at the user-side is evaluated through simulation and compared against state of the art solutions.

The advent of the Internet of Everything and new Ultra-Reliable Low-Latency Communication (URLLC) services has resulted in an exponential growth in data demands at the network’s edge. To meet the stringent performance requirements of evolving 5G (and beyond) applications, deploying dedicated resources closer to mobile users is essential. Multi-Access Edge Computing (MEC) is a promising technology for bringing computational resources closer to users. However, the distributed and limited MEC resources must be effectively optimized to maximize the number of mobile users benefiting from low-latency MEC services at each time slot in highly congested, large-scale, and dynamic wireless network scenarios. In this research, we propose and evaluate a novel Artificial Intelligence-Defined Wireless Networking (AIDWN) approach that builds on conventional Software-Defined Networking (SDN), implementing a new AI-defined application plane for computational offloading and resource allocation in MEC-enabled wireless networks. The AIDWN approach implements a deep reinforcement learning framework and deep neural networks that dynamically adapt optimal computational offloading and wireless resource allocation decisions while considering the handover, mobility, and coordinated resource allocation challenges in highly dynamic and mobile multi-MEC server environments. Compared to recent state-of-the-art proposals, the proposed AIDWN demonstrates a substantial performance improvement, utilizing more than 90% of MEC resources per time slot across all MEC servers. It also accommodates significantly more mobile users in highly congested wireless network scenarios. We identified various future research directions highlighting the potential of the AIDWN approach in simplifying the management of next-generation wireless networks.