Spiking Neural Networks -- Part III: Neuromorphic Communications
Show affiliations
Abstract
Synergies between wireless communications and artificial intelligence are increasingly motivating research at the intersection of the two fields. On the one hand, the presence of more and more wirelessly connected devices, each with its own data, is driving efforts to export advances in machine learning (ML) from high performance computing facilities, where information is stored and processed in a single location, to distributed, privacy-minded, processing at the end user. On the other hand, ML can address algorithm and model deficits in the optimization of communication protocols. However, implementing ML models for learning and inference on battery-powered devices that are connected via bandwidth-constrained channels remains challenging. This paper explores two ways in which Spiking Neural Networks (SNNs) can help address these open problems. First, we discuss federated learning for the distributed training of SNNs, and then describe the integration of neuromorphic sensing, SNNs, and impulse radio technologies for low-power remote inference.
Next-Generation Wireless Networks (NGWN) for Autonomous Intelligent Communications
View this Special Issue
Research Article | Open Access
Volume 2021 |Article ID 9614520 |
https://doi.org/10.1155/2021/9614520
Show citation
New Trends and Advancement in Next Generation Mobile Wireless Communication (6G): A Survey
Sher Ali,1 Muhammad Sohail,2 Syed Bilal Hussain Shah,3 Deepika Koundal,4Muhammad Abul Hassan,5
Asrin Abdollahi,6 and Inam Ullah Khan7
Show more
Academic Editor: Abdul Basit
Published11 Dec 2021
Abstract
As the commercial implementations of 5G networks have been initiated in different regions of the world, the focus of the researchers is bending towards the next generation of wireless communication. This research study intends to investigate the requisites of the fast establishment of the theoretical and practical measures for sixth generation (6G) wireless communication. To this end, this paper first outlined the existing research works that have considered different aspects of 6G, and then based on this existing works, the future vision is established. Then, the 6G vision is based on four types of connectivity and is summarized as “Wherever you think, everything follows your heart.” To fill the gap between the market requirements after one decade and the limited capabilities of 5G, different specifications of 6G that make it an appropriate replacement are discussed. Furthermore, different candidate technologies that can potentially realize the 6G communication are studied, followed by discussion on different challenges in the realization and possible research directions to cope with these challenges. By exploring the vision of future, its specification, and key candidate technologies, this paper attempts to summarize the general 6G framework. In addition, with mentioned challenges in realization of 6G, the aim of this paper is to guide the researcher and attract their interest to consider them.
1. Introduction
Though the deployment of fifth-generation (5G) wireless communication is in the early phase, i.e., the associated features should be more enhanced, it is also essential to uniformly set the communication requirements of the upcoming information society and initiate the theoretical and practical projects on future wireless system generation (henceforth mentioned as 6G). To this end, we intend to investigate the requisites for the establishment of the theoretical and practical works on 6G from three perspectives: (i) 10-year gap between successive generations
3.1. Intelligent Connectivity
Nowadays, artificial intelligence (AI) has gained the interest of researchers and become the most active research topic. Therefore, almost every field is moving towards the utilization of AI technology. The AI enabled wireless communication network has become a foreseeable trend [32–48]. In recent times, people have started to attempt the utilization of AI in 5G system [49, 50]; however, the enabling of AI in 5G systems is possible only as optimization of conventional networks based on AI, instead of a new one. AI-enabled 5G will change the dynamics of the networks. Even though the initial 5G architecture is more flexible (software definable), but AI has not been considered in it, therefore, it is still considered to be conventional network architecture. Secondly, even though the development of AI technology is speedy enough, and in many areas, it has proven its performance efficiencies; still it is in the experimental phase in most of the other fields. The enabling of AI in wireless communication technology is in its early research phase, and a long-term research planning is desirable before the actual technology grows up to maturity.
The growing tendency towards AI indicates the high probability of technology growth in the upcoming decade. In addition, believing that the upcoming 6G networks will be growingly massive and heterogeneous, and different application scenarios will become growingly dynamic; the active use of AI is expected to cope with these challenges. It is envisioned that the upcoming 6G will bring a revolution in the applications scope of conventional cellular systems and turn into the main Internet supporting all the operations in every industry and society. If the upcoming generation networks still utilize the current unified communication models to handle the very complex and diverse applications in 6G era, it will be faced with many severe challenges [51]. Therefore, we consider that building the AI-based 6G network will be an unavoidable move, and “Intelligent” will be its essential feature, resulting in “Intelligent connectivity.”
The “Intelligent Connectivity” characteristic can be considered as an essential intellectualization of communication networks: intellectualization of connecting units (terminal devices) and intellectualization of network architecture and elements. In future, lots of issue will be faced in the realization of 6G networks: growingly massive and complex networks, massive densification of terminal devices, and complex business models. “Intelligent Connectivity” will simultaneously fulfil two demands: (1) all the connected network devices need to be intelligent and (2) the need of intelligent management for the growingly massive network.
7. Conclusion
In this research paper, we have provided a detail and systematic survey on 6G wireless communication. This survey is carried out in such a way that initially, the literature on 6G and different practical research initiatives taken by different organizations are presented followed by the 6G vision, specifications, challenges, different candidate technologies, and future research directions. The vision of 6G is based on requirements that 5G is unable to fulfil and those services that 5G is unable to provide. To this end, the vision of 6G is categorized into four fundamental services: “Intelligent Connectivity,” “Deep Connectivity,” “Holographic Connectivity,” and “Ubiquitous Connectivity.” Based on discussed 6 visions, the achievable 6G goals are specified which are (a) latency minimization, (b) global connectivity, (c) massive connectivity, (d) enormously high data rates, (e) energy efficiency of network devices, (f) connection reliability, and (g) machine learning-based connected intelligence. To achieve these specified 6G goals, many potential technologies are proposed which include AI, FSO backhaul network, blockchain, UAVs, 3D networking, DNS, sensing-based communication, big data analytics, and some new spectrum-based technologies, e.g., terahertz spectrum and Optical Wireless Communications. To realize the vision of 6G and implementation of the potential candidate technologies to achieve the specified 6G goals, a lot of challenges will be faced which require intense research. These challenges include Peak Rate-Terabit, higher energy efficiency, connection everywhere and anytime, self-aggregating, high propagation and atmospheric absorption of THz, complexity in resource management for 3D networking, heterogeneous hardware constraints, autonomous wireless systems, modelling of sub-mmWave (THz) frequencies, and spectrum and interference management.
As the commercial implementations of 5G networks have been initiated in different regions of the world, the focus of the researchers is bending towards the next generation of wireless communication. This research study intends to investigate the requisites of the fast establishment of the...
www.hindawi.com