Learning to Communicate: Deep Learning based solutions for the Physical Layer of Machine Type Communications [LeanCom]

Lead Research Organisation: University College London
Department Name: Electronic and Electrical Engineering


With the advent of the Internet of Things (IoT), machine type communications (MTC), cloud computing and many other applications, the wireless network will become far more complex, while at the same time far more essential than ever before.
Given the above exponential growth in both connectivity and complexity of the wireless systems and the unprecedented demands on latency, capacity, ultra-reliability and security, the network is becoming analytically intractable. Naturally, human-driven physical layer (PHY) design approaches rooted on mathematical models of communications systems and networks which drive today's network architectures are being surmounted by the sheer complexity of the emerging network paradigms. Hardware imperfections, that are inevitable with the employment of low-cost MTC sensors and transmitters, will drastically increase the volatility of the network, and theoretically driven solutions typically relying on generic and highly inaccurate models cannot address this as they are highly sub-optimal in practice. The above challenges necessitate new data-driven approaches to the design of communications systems, as opposed to traditional system-model driven designs that are becoming obsolete.

Towards the diverse communication paradigms of MTC of the future, there is an urgent need to address reliable and adaptive links detached from mathematical models, and instead based on data-driven approaches. This visionary project will address these fundamental challenges by developing new Neural Netowrk architectures tailored for wireless communications, and new transceiver architectures based on data-driven training. Our research will address the development of a) a communications specific DL framework, b) DL-inspired PHY solutions and, c) proof-of-concept verification of the proposed solutions.

LeanCom will be performed with Huawei, NEC Europe, Duke University, The Digital Catapult and CommNet and aspires to kick-start an innovative ecosystem for high-impact players among the infrastructure and service providers of ICT to develop and commercialize a new generation of learning-based networks. The implementation, experimentation and testing (within WP3) of the proposed solutions serves as a platform towards commercialisation of the results of LeanCom, aiming towards an impact of a foundational nature for the UK's digital economy.

Planned Impact

The explosive growth of industrial control processes and the industrial IoT applications, provides unique opportunities for the creation of impact through MTC research. This project promotes a fundamental paradigm shift to communications system design, and therefore holds the potential for high impact research of a foundational nature for the UK's digital economy. Impact will be measured in the number of products and commercialisation activities, consultancies and patents filed, by any follow-up industry funded collaborations, and through the success of public engagement activities.
To foster the economic competitiveness of the UK, this project will develop novel, low-cost solutions suitable for the MTC ecosystem. Direct beneficiaries are: (1) researchers in wireless communications, machine-type communications, supported by the Internet-of-Things (IoT), and researchers in the general areas of machine learning and in other engineering fields, (2) users of MTC applications, as well as, government departments and private sector companies (from small firms to big enterprises) with interest in emerging MTC (e.g. industry 4.0 and IoT), (3) policy making regulators, such as Ofcom and.
Commercial and societal impact: The growing reliance of society on wireless technologies, along with the emergence of industrial applications, prominently Industry 4.0, that rely on MTC solutions, and others such as smart cities and connected cars, provides a unique platform for achieving commercial impact through the proposed work. By focusing on practical implementation and hardware efficiency, along with proof-of-concept testing, we will ensure the promised massive connectivity targets are brought to practice. By enabling highly complex and adaptive communication scenarios, the proposed technological shift will be able to tackle the data deluge in the upcoming decades. It is anticipated that the outputs of this research project will be used by telecom/electronics manufacturers (e.g., Huawei, National Instruments, Nokia, Samsung), telecom business operators (e.g., Vodafone-UK, BT) and data stakeholders (e.g. Facebook, Google) within the UK and abroad to build machine-type networks with massive connectivity. The strategic placement of our Partners Huawei, NEC, Digital Catapult at the centre of the 5G ecosystem provides direct means for commercialization through IP exploitation of the project's solutions and promotion of the developed solutions to their product lines.
Overall the project lies at the centre of EPSRC's artificial intelligence strategy, and aligns with the following EPSRC's cross-ICT priorities (2017-2020) : Future Intelligent Technologies - learning-based transmission is at the centre of this portfolio and our MTC solutions open new horizons for high-reliability communication between different objects, machines and humans, fully aligned with the EPSRC vision to '' promote research which aims to develop intelligent, adaptive or autonomous systems that can learn, adapt and make decisions without the need for human control''. Information and Communications Technology (ICT) networks and distributed systems - by proposing disruptive solutions for future dense MTC, we will facilitate their rollout after 2025.

New experts: The research training within the project, involving visits for hands-on training to our industrial partners Huawei and NEC will develop the research profile, expertise and man power of the world class research group in UCL and foster a team with excellence in DL-based communications and MTC, by producing new experts in the field of communications-tailored learning architectures, learning-based transmission, machine-type communications, and wireless communications in general.

The commercialization, exploitation, outreach and dissemination activities to realise the above impact are detailed in the Pathways to Impact section.


10 25 50
publication icon
Liu F (2020) Radar-assisted Predictive Beamforming for Vehicular Links: Communication Served by Sensing in IEEE Transactions on Wireless Communications

publication icon
Mohammad A (2020) Complexity-Scalable Neural-Network-Based MIMO Detection With Learnable Weight Scaling in IEEE Transactions on Communications