Yuanwei Liu
 

Research Interests

  • Non-Orthogonal Multiple access (NOMA)
  • Millimeter Wave Communications
  • Internet of Things (IoT)
  • Resource Allocation in 5G networks and Beyond
  • Stochastic Geometry and Matching Theory
  • Machine Learning and Big Data

  • Running Research Projects

    If you are interested on the following projects, just drop me an email. I am happy to work with you together.

    1) Non-Orthogonal Multiple access (NOMA) for 5G and IoT networks

    Non-orthogonal multiple access (NOMA), which has been recently proposed for the 3rd generation partnership projects long-term evolution advanced (3GPP-LTE-A), constitutes a promising technology of addressing the above-mentioned challenges in 5G networks by accommodating several users within the same orthogonal resource block. By doing so, significant bandwidth efficiency enhancement can be attained over conventional orthogonal multiple access (OMA) techniques. This motivated numerous researchers to dedicate substantial research contributions to this field. We advocate on power-domain multiplexing aided NOMA, with a focus on the theoretical NOMA principles, multiple antenna aided NOMA design, on the interplay between NOMA and cooperative transmission, on the resource control of NOMA, on the co-existence of NOMA with other emerging potential 5G techniques and on the comparison with other NOMA variants. We highlight the main advantages of power-domain multiplexing NOMA compared to other existing NOMA techniques. We will continue to work on NOMA with tackling some implementation issues as well as identifying promising research opportunities for the future.
    The Pulpit Rock
    This project is jointly worked with [Princeton University], [Southampton University], [Lancaster University] and [China Mobile]. Details for this project can be found in [Paper] and [Magazine]. An interesting demonstration for this project is available on [Video] [Researchgate] and [Slides]. A full tutorial version is available here [Tutorial],

    2) When Big Data Meets Machine Learning in Wireless Networks

    Recent several decades have witnessed the exponential growth in commercial data services, which lead to step in the so-called big data era. The pervasive increasing data traffic present both the imminent challenges and new opportunities to all aspects of wireless system design, such as efficient wireless caching, base station deployment and adaptive multiple access design. Machine learning, as one of the most promising artificial intelligence tools, has been invoked in many areas both in the academia and industry. Nevertheless, the application of machine learning in wireless communication scenarios is still in its infancy, which motivates to develop this project. The aim of this project is to use social media data to predict the requirements of mobile users for improving the performance of wireless networks. More particularly, a unified machine learning framework with the aid of the social media data is proposed in this project. Four stages are included in the proposed framework, which consists syntax processing, semantics analysis, data modelling and online prediction/refinement. The main benefits of the proposed framework is by utilizing the social media data which reflect the real requirements of users, to assist refining the motivation, problem formulation, and methodology of powerful machine learning algorithms in the context of wireless networks. The proposed framework is as follows.
    The Pulpit Rock
    This project is jointly worked with an industry company [Onfido].

    3) Low-Power Wide-Area Networks (LPWANs) for Internet-of-Things (IoT)

    Internet-of-Things (IoT) is envisioned as a means to connect billions of small computing devices embedded in the environment (e.g., walls and soil) and implanted in human bodies. Aiming to provide the required connectivity involving such numbers of devices, two possible networking approaches have been proposed. One is the evolution from existing wireless networks with the purpose of supporting machine-to-machine (M2M) communications in the 5G system. Another approach is to design M2M-dedicated networks from scratch, such as the emerging low-power wide-area networks (LPWANs). Two technical options have been proposed to increase transmission range and power efficiency of such systems yet reduces the data rate. One solution is the narrowband approach which slices the bandwidth for data transmission to reduce noise at the receiver. One of the classic technologies with the narrowband approach in LPWANs is Sigfox. The second solution is to add coding gain to a higher rate signal to combat the high receiver noise in a wideband receiver. LoRa is an example of the latter LPWAN approach. We focus on investigating the mathmatical modelling, resource allocation, and real testing for LPWANs.
    The Pulpit Rock
    Details for this project are available on [Paper] and [Paper]. This project is jointly worked with [Imperial College London], [Georgia Institute of Technology] and [Lancaster University].

    4) Matching Theory Based Resource Allocation in 5G Networks and Beyond

    Matching theory is a powerful tool to study the formation of dynamic and mutually beneficial relations among different types of rational and selfish agents. It has been widely used to develop high performance, low complexity, and decentralized protocols. Recently, there has been significant progress in intensive research work that uses matching theory to handle resource allocation problems in wireless networks. We are working on invoke matching theory to solve the resource allocation problems in 5G networks and beyond, such as in NOMA, D2D communications, heterogeneous cellular networks, millimeter wave communications, etc. A simple matching algorithm is listed as follows.
    The Pulpit Rock
    Details for this project are available on [Paper] and [Slides]. More related works can be found in my publications.

    5) Stochastic Geometry on Wireless Communications

    Stochastic geometry is a powerful mathematical and statistical tool for modeling and analyzing wireless networks. Unlike the traditional topology approaches which always ignore the density and mobility of nodes, stochastic geometry is capable of capturing the topological randomness of the networks and hence can provide tractable analytical results for the average network behaviors according to some distributions. This is particularly essential in large-scale networks which consist of a large number of randomly deployed nodes (e.g., BSs, mobile users, etc.) whose channels and locations are with high uncertainty. We are invoking stochastic geometry to model different wireless network scenarios, such as physical layer security, UAV communications, millimeter wave communications, NOMA, IoT networks, etc. Below figure is an example of stochastic geometry model, which shows the spatial distribution of a wireless powered secure D2D communication scenario.
    The Pulpit Rock
    Details for this project are available on [Paper] and [Slides]. More detials for stochastic geometry can be found in my [Thesis].

    6) Exploiting QoE Awareness in Wireless Networks: A Cross-Layer Perspective

    Wireless networks are conventionally evaluated using PHY connection-centric metrics such as throughput or sum-rate capacity. However, conventional connection-centric designs have become a barrier to meeting the diverse application requirements and the quality expectation of the end users, especially for the rapid expanding visual-experience-oriented services, such as VR, AR, and video streaming. Even though the data rate and throughput are increasing, current mobile networks are still facing poor user experience and low service quality. The gap between the PHY system performance and higher-level user’s perceived experience drives a paradigm shift from connection-centric designs to experience-centric designs. QoE is the perceptual quality of service (QoS) from the user’s perspective. Given the context that the growth of future traffic demands is largely driven by the visual-experience-oriented services, such as VR, AR, and video streaming, there is broad consensus among leading industry and academic initiatives that improving user’s QoE is a key pillar to sustaining the revenue of service providers in future networks.
    The Pulpit Rock
    Details for this project are available on [Paper] and [Poster].