#	Pagina
attuale pagina	/open-h2020/projects/194535/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 3 - WILLOW (WIreLess LOWband communications: massive and ultra-reliable access)

Teaser

Summary

Cellular wireless systems from 2G to todayâ€™s 4G have been continuously evolving towards offering broadband connectivity to the users. While the trend of reaching even higher rates will continue in the fifth generation of wireless systems, there is a common consensus that 5G will not only be â€œ4G, but fasterâ€œ, but it will offer new modes of connectivity to a massive number of simple devices and/or support extremely reliable connections. WILLOW has the ambitious objective to create the fundamental wireless transmission schemes and protocols for massive and ultra-reliable lowband connectivity. Here lowband indicates that the target applications and services do not require high rates, but rather low rate represented by short messages from a large number of machines/sensors (e.g. as in the smart grid) and/or ultra-reliable delivery of short critical messages (e.g. among interconnected cars at a crossing). Designing massive and ultra-reliable lowband communication cannot be done by incremental changes to the current protocols, but it requires a fundamental leap in the wireless system architecture, by introducing the capability to handle a massive number of short packets and redefining the relationship between the data and the metadata (control data). In addition to the obvious advantages of broadband communication, massive and ultra-reliable lowband communication is bringing a ground-breaking novelty to the wireless 5G systems, as it will make wireless a true commodity, just like electricity or water, and enable a plethora of new applications and services.

Massive and ultra-reliable lowband connections are essential enablers of the Internet of Things (IoT). The avaliability of IoT connectivity for massive number of devices, as well as support for IoT connections with ultra-high reliability, will introduce disruptive changes in the vertical sectors: energy, transportation, industrial production, health, etc. For example, once ultra-reliable wireless connections are available, the notion of a robot changes from a single physical entity to a distributed group of modules that are wirelessly collaborating towards performing the robotic function. This paradigm fits in the emerging concept of Industry 4.0 taht consists of smart interconnected production facilities.

The two fundamental issues that WILLOW investigates towards creating lowband communication systems are:
1- Efficient communication with short packets, in which the data size is comparable to the size of the metadata, i.e. control information. This is not an issue in broadband communication, where the large data makes the metadata negligible. The considerations of control information are essentially changing the traditional communication model introduced by Shannon by the one on Figure 1, which takes into account protocol information.
2- System architecture in which graceful rate degradation, low latency and massive access can exist simultaneously with the usual, broadband data services.

Considering these two fundamental issues, the overall objective of WILLOW is to investigate the architectural principles of lowband communication and create communication schemes and protocols that are optimized for massive and reliable transmission of short packets.

Work performed

The overall objective of WILLOW is to investigate the fundamental wireless transmission schemes and protocols for massive and ultra-reliable lowband connectivity. The term â€œlowbandâ€ was coined to indicate that the target applications and services do not require high rates, but rather low rate represented by short messages from a large number of machines/sensors (e.g. as in the smart grid) and/or ultra-reliable delivery of short critical messages (e.g. to support an industrial production plant).

The research plan for the WILLOW project was made in 2014 and it is very rewarding to see that this plan has become one of the central elements on the global research agenda for wireless communications. Namely, since 2014 there has been a growing consensus that 5G wireless systems will support three generic services, which, according to the ITU-R nomenclature, are classified as: Enhanced mobile broadband (eMBB), Massive machine-type communications (mMTC) and Ultra-reliable and low-latency communications (URLLC). It is noted that two of the he three services envisioned in 5G are exactly the ones that represent the main subject of WILLOW, such that the impact of the project is already very significant.
The research and technological achievements will be presented in the framework of the workpackages, through which the project is structured.

WP1: Fundamental concepts.

This WP deals with the investigation of the basic trade-offs between the resources used for metadata and data versus system performance in terms of connection reliability, scaling of the rate/delay with the number of connected devices, energy consumption.
The first major contribution of WP1 is described in [1], which surveys recent results from information theory for short packets (finite blocklength) and uses these results to identify three new fundamental trade-offs that arise in communication systems where the packets are short and where the data and metadata are comparable in size. The impact of this article has been very significant, as it is the first comprehensive work that puts the final blocklength theory in the system design perspective. The paper is published in the Proceedings of the IEEE, arguably one of the most visible publications in electrical engineering, as covers all the societies related to IEEE.

The second example from [1], concerning downlink multi-user communications, is investigated in detail in the two workshop papers [2] and [3] and in a journal publication [4]. The key idea is to generalize and systematically analyze the observations made in [1], but also to include the need for communicating metadata in the analysis. The paper [4] has been seen as one of the first theoretically-supported instances that show the impact of the control information on wireless communication systems.

An important elements towards supporting reliable low latency communication is a HARQ protocol. The work [5] considers a generalization of the hybrid ARQ (HARQ) protocol which achieves significantly improved throughputs under an average latency constraints. The work takes an information-theoretic approach and proves optimality of the protocol under certain conditions. Information-theoretic aspects of HARQ have also been the subject of [31], where we have investigated HARQ buffer management by leveraging information-theoretic achievability arguments based on random coding. The analysis sheds light on the impact of different compression strategies. The optimization of coding blocklength was investigated, highlighting the benefits of HARQ buffer-aware transmission scheme.

Another series of works considers the problem of communicating a common-message and the role of feedback. Feedback already plays an important role in current wireless communication systems in form of acknowledgements and handshakes, but the more stringent requirements to reliability makes it relevant to consider the role of feedback in the upcoming ultra-reliable communication systems. The co

Final results

The technology aspects investigated in WILLOW are central to the upcoming revolution of the Internet of Things. Indeed, in traditional broadband services the transmitted data volume is very large, which makes the size of metadata not significantly large in order to be a target for highly optimal transmission methods. The results achieved so far have clearly advanced the understanding of communication protocols for sending short data packets and have revealed new aspects of encoding control information. The papers published within the WILLOW project are already cited and very much in the focus of many ongoing 5G projects. Furthermore, the standardization of ultra-reliable communications will be carried out within the coming years and it is expected that it will proliferate in 2025, which makes the results of WILLOW perfectly timed in order to have a very visible impact on the upcoming wireless communication technology.

As the project has evolved, it became apparent that various methods of machine learning and data analytics need to be applied in order to carry out design and performance assessment of ultra-reliable systems. Furthermore, we have started to use machine learning to address the problem of integration of the wireless communication in Internet-of-Things (IoT) systems with the data mining. These activities will be pursued in the second half of the project. Finally, the communication paradigm in IoT systems has recently started to change significantly due to the emergence of the blockchain-based systems, that are poised to have a significant role in IoT as enablers of smart contracts. Our research group is, to the best of our knowledge, one of the first to have started analysis of the communication aspects of blockchain-based IoT systems and this topic will be pursued within the second half of the WILLOW project.