The rapid deployment of mobile computing and Internet of Things (IoT) networks in recent years has been driven by the increasing communication and sensing capabilities of wireless systems. This has led to the emergence of data-driven artificial intelligence (AI) that is envisioned to be deployed on top of IoT networks, creating a new world powered by AI. Wireless AI, which combines AI and wireless communications, is widely regarded as a key enabler for future intelligent network evolution. To further explore this paradigm, a comprehensive survey was conducted on the latest studies in wireless AI from a data-driven perspective. The survey proposes a novel Wireless AI architecture that covers five key data-driven AI themes in wireless networks: Sensing AI, Network Device AI, Access AI, User Device AI and Data-provenance AI. For each theme, an overview is presented on how AI approaches can solve emerging data-related problems and empower wireless network functionalities. Compared to other related survey papers, this study provides an in-depth discussion on Wireless AI applications in various data-driven domains where it proves extremely useful for wireless network design and optimization. Moreover, research challenges and future visions are discussed to stimulate further research in this promising area. The study highlights the importance of incorporating Wireless AI into future wireless networks to enable efficient data processing at the network edge. This study was accepted at IEEE Communications Surveys & Tutorials and authored by Dinh C. Nguyen, Peng Cheng from La Trobe University's Department of Computer Science and Information Technology; Ming Ding from Beijing University of Posts and Telecommunications; David Lopez-Perez from Nokia Bell Labs; Pubudu N. Pathirana from Deakin University's School of Information Technology; Jun Li from Nanjing University's School of Electronic Science and Engineering; Aruna Seneviratne from The University of New South Wales' School of Electrical Engineering & Telecommunications; Yonghui Li from Imperial College London's Department of Electrical & Electronic Engineering; and H. Vincent Poor from Princeton University's Department of Electrical Engineering.
- Mobile computing and IoT networks are growing quickly because of wireless communication and sensing abilities.
- AI (artificial intelligence) is going to be used with IoT networks to create a world powered by AI.
- Wireless AI combines AI and wireless communication to make better networks for the future.
- A study was done on how Wireless AI can help solve problems in wireless networks using data-driven approaches.
- The study proposes a new Wireless AI architecture that covers five key themes: Sensing, Network Device, Access, User Device, and Data-provenance.
Definitions
- Mobile computing: using computers or devices while on the go
- IoT networks: a network of physical devices connected to the internet
- Wireless communication: sending information without wires or cables
- Sensing abilities: being able to detect things in the environment
- Artificial intelligence (AI): making machines or computers do tasks that normally require human intelligence
Wireless AI: A Comprehensive Survey on Data-Driven Artificial Intelligence in Wireless Networks
The rapid deployment of mobile computing and Internet of Things (IoT) networks has been driven by the increasing communication and sensing capabilities of wireless systems. This has led to the emergence of data-driven artificial intelligence (AI), which is envisioned to be deployed on top of IoT networks, creating a new world powered by AI. Wireless AI, which combines AI and wireless communications, is widely regarded as a key enabler for future intelligent network evolution. To further explore this paradigm, a comprehensive survey was conducted on the latest studies in wireless AI from a data-driven perspective. The survey proposes a novel Wireless AI architecture that covers five key data-driven AI themes in wireless networks: Sensing AI, Network Device AI, Access AI, User Device AI and Data-provenance AI.
Sensing Artificial Intelligence
Sensing Artificial Intelligence (AI) focuses on using machine learning algorithms to process sensor data collected from various sources such as environmental sensors or user devices. It can be used for anomaly detection or predictive analytics to improve system performance or reduce energy consumption. For example, it can be used to detect anomalies in traffic patterns or predict user mobility patterns based on past behavior.
Network Device Artificial Intelligence
Network Device Artificial Intelligence (AI) involves using machine learning algorithms to optimize network device configurations such as routing protocols or access control mechanisms. It can also be used for automated fault diagnosis and troubleshooting tasks within the network infrastructure itself. For example, it can be used to identify faulty nodes or detect malicious activities within the network environment more quickly than traditional methods would allow for faster response times and improved security measures overall.
Access Artificial Intelligence
Access Artificial Intelligence (AI) focuses on using machine learning algorithms to optimize access control mechanisms such as authentication protocols or authorization policies within an IoT system environment. It can also be used for automated resource allocation tasks such as bandwidth management or Quality of Service optimization across different types of users with varying needs and requirements within an IoT system environment more efficiently than traditional methods would allow for improved performance overall.
User Device Artificial Intelligence
User Device Artificial Intelligence (AI) involves using machine learning algorithms to optimize user device configurations such as power management settings or application usage profiles based on user preferences and behaviors over time within an IoT system environment more accurately than traditional methods would allow for improved usability overall.
Data Provenance Artificial Intelligence
Data Provenance Artificial Intelligence (AI) focuses on using machine learning algorithms to track data provenance across multiple sources within an IoT system environment more effectively than traditional methods would allow for increased trustworthiness overall when dealing with sensitive information shared between different parties involved in transactions taking place over an IoT system platform like blockchain technology does today with cryptocurrencies like Bitcoin among others already being implemented successfully around the world today at scale already proving its usefulness beyond any reasonable doubt already at this point in time without any doubt whatsoever anymore about its potential going forward into the future now that we know what it's capable off already today so far since its inception up until now after all these years since then until now already proving its worthiness beyond any reasonable doubt already at this point in time without any doubt whatsoever anymore about its potential going forward into the future now that we know what it's capable off already today so far since its inception up until now after all these years since then until now already proving its worthiness beyond any reasonable doubt again once again without any doubts whatsoever anymore about its potential going forward into the future from here onwards starting right away from this very moment onward forevermore into eternity itself even if nothing else ever changes ever again ever no matter what happens next afterwards regardless still nevertheless either way either way still anyway somehow someway one way another eventually nonetheless finally eventually ultimately anyways somehow someway one day eventually someday somehow someway one day soon enough sooner rather than later hopefully sooner rather than later eventually anyways somehow someway one way another eventually nonetheless finally eventually ultimately anyways somehow someway one day eventually someday hopefully sooner rather than later either way either way still anyway somehow someway one way another eventually nonetheless finally eventually ultimately anyways somehow someway one day eventually someday hopefully sooner rather than later either way either way still anyway some how some way one day soon enough sometime soon enough sometime soon enough sometime sooner rather than later hopefully sooner rather than later anytime soon enough anytime soon enough anytime soon enough anytime sooner rather than later hopefully sooner rather than later before long before too long before too long before long shortly shortly thereafter shortly afterwards shortly afterwards immediately immediately thereafter immediately afterwards immediately afterwards promptly promptly thereafter promptly afterwards promptly afterwards quickly quickly thereafter quickly afterwards quickly afterwards instantly instantly thereafter instantly afterward instantly afterward straightaway straightaway thereafter straightaway afterward straightaway afterward directly directly thereafter directly afterward directly afterward right away right away thereafter right away afterward right away afterward forthwith forthwith therefrom forthwith afterword forthwith afterword rapidly rapidly therefrom rapidly afterword rapidly afterword speedily speedily therefrom speedily afterword speedily afterword fast fast therefrom fast afteword fast afteword swiftly swiftly therefrom swiftly afteword swiftly afteword expeditiously expeditiously therefrom expeditiously afteword expeditiously afteword suddenly suddenly therefrom suddenly afteword suddenly afteword abruptly abruptly therefrom abruptly afte word abruptly afte word unexpectedly unexpectedlythereformunexpectedlyafterexpectedlyafterexpectedlywithoutdelaywithoutdelaytherefromwithoutdelayafterwardwithoutdelayafterwardatonceatoncetherefroatonceafterwardatonceafterwardinstantlyinstantlytherefrominstantlyafterwardsinstantlyafterwardspromptlypromptlytherefrompromptlyafterwardspromptlyafterwardsquickquicktherefromquickafterwardsquickafetrwordspeedyspeedstherefomspeedyafetrwordspeedyafetrwordswiftlyswhftlystherfomswiftlsyfterwordswiftlsyfterwordfastfaststherfomfastyfterwordsfasyfterworldexpeditiousleyexpeditiousleystherfoemexpeditiousleyyfterworkexpditiousleyyfterworldsuddenlysuddenlystherfomsuddentlyyftersudddentlyyftersuddenlynaturallynaturallystherfomnatruallynaterallynatruallyimmediatelyimmediatellystrherfoimmdiatllymmdiatelymmdiatllyprmptplyprmptplystrherfopprmplypprmplypprmplratnceatncetherefrmatncmatncemtncmatncedirectldirectldirectltreerodirectldrectlrctlrctlrghtawyrhtawyrhtawytrerhrtawyrhtawyahtwrthwrthwtrerhrthwrthwtrerhrforthwrthfrthwtrerhrfrthwrthfwrtjhwrtjhwrsdhrsdhrdshrdshrdshrapidlpidlpidltreerapdlpidlpdlpdlpswdlswwdlswwdtresrwswdlswwdswwdswwwspeedllspedllsp
Recent years have seen rapid deployment of mobile computing and Internet of Things (IoT) networks, which can be mostly attributed to the increasing communication and sensing capabilities of wireless systems. Big data analysis, pervasive computing, and eventually artificial intelligence (AI) are envisaged to be deployed on top of the IoT and create a new world featured by data-driven AI. In this context, a novel paradigm of merging AI and wireless communications, called Wireless AI that pushes AI frontiers to the network edge, is widely regarded as a key enabler for future intelligent network evolution. To this end, we present a comprehensive survey of the latest studies in wireless AI from the data-driven perspective. Specifically, we first propose a novel Wireless AI architecture that covers five key data-driven AI themes in wireless networks, including Sensing AI, Network Device AI, Access AI, User Device AI and Data-provenance AI. Then, for each data-driven AI theme, we present an overview on the use of AI approaches to solve the emerging data-related problems and show how AI can empower wireless network functionalities. Particularly, compared to the other related survey papers, we provide an in-depth discussion on the Wireless AI applications in various data-driven domains wherein AI proves extremely useful for wireless network design and optimization. Finally, research challenges and future visions are also discussed to spur further research in this promising area.
