Optimal Resource Allocation for Heterogeneous Transmission in Multi-Cell OFDMA Systems

Visiting researcher: Prof. Qinghai Yang (XDU)
Host researcher: Prof. Lajos Hanzo (University of Southampton)
Duration: 3 months
Status: Ongoing

Biography of the visiting researcher

Prof. Qianghai Yang received his B.S. degree in Communication Engineering from Shandong University of Technology, China in 1998, M.S. degree in Information and Communication Systems from Xidian University, China in 2001, and Ph. D. in Communication Engineering from Inha University, Korea in 2007 with university-president award. From 2007 to 2008, he was a research fellow at UWB-ITRC, Korea. Since 2008, he is with Xidian University as an associate professor. His current research interest lies in the fields distributed communication and control, resource sharing and fast scheduling for smart communications, specifically on the points of guessing game based resource sharing for smart green radios, quick scheduling for vehicle-to-vehicle communications, low complexity algorithms of resource allocation in multi-cell OFDMA systems with minimal information exchange among BSs, secure multi-PHY based WPAN/WBANs.
He has published 26 international journal papers, 38 international conference papers, and several patent/draft proposals related to LTE-A standardizations. He is the member of IEEE, IEICE and KICS, and has served the session chair of IEEE IWCLD 2007 & ICACT2008 & Wicom2008 & ISCIT2009, and the TPC member of IEEE ICCS2008, CCNC2009, ISCIT2009, PIMRC2009, CCNC2010, ISCIT2010, WCNIS2010 and ICAIT2009&ITS World Congress, etc.

Background and motivation

Spectrum and energy costs, regarded as the essential resource consumption, account for the majority of a mobile operator’s operating expenses. Efforts to gain more efficient resource usage have never been ended. In multi-cell OFDMA systems (downlink), inter-cell interference (ICI) dramatically erodes the achievable performance, especially in the case of unity frequency-reuse. Efficient resource allocation approaches have to be used to mitigate the ICI problem by minimizing the interference level and/or by maximizing the system-wised throughput subject to power constraints, inter-cell signaling limitations, fairness or minimum data rate requirement. For the cellular scenario, the users in the cell-centre and cell-edge region have dissimilar performance characteristics and/or QoS requirements. In contrast to cell-centre users, cell-edge users suffers from high ICI and thus it recommended to assign the spectrum resources as orthogonal as possible in the cell-edge region among multiple cells. A possible way to improve the throughput of the cell-edge users is to assign them more power. But this would increase the ICI to other cells. Contrarily, the effective way to increase the rate of cell-centre users is to allocate more spectral resource instead of more power. On the other hand, while promising for green communications, the power saving in cellular networks is not only beneficial for the global environment but also makes commercial sense for the operators. Thus, from the social and system-wised points of view, how to improve the spectral and power efficiency constitutes a crucial issue.
Furthermore, the network is heterogeneous, consisting of diverse network modes and kinds of user services, e.g., macro-cell, relaying, pic-cell and femto-cell along with best-effort (non-real time) and high priority users (real time or rate constraint), etc. In a heterogeneous multi-user multi-cell scenario, the QoS demand of high priority users may vary from user to user depending on applications. A user may not use the same application all the time; as a result, the required data rate will vary with time. In order to maximize the revenue, the service provider will improve the resource efficiency to satisfy the QoS requirements for as many users as possible. By contrast, for the best effort users, it requires dynamic resource allocation to further improve the resource efficiency and then to maximize the overall throughput. For the heterogeneity of the cellular network structure, the deployment of relaying and/or femto-cell generally benefits the QoS improvement (e.g., coverage extension) and energy saving, but it requires much more advanced resource/interference management algorithms. Additionally, by modifying the priority level of high priority and best effort users and by also considering the grouping of cell-centre and cell-edge users, the fairness issue of resource allocation arises as a new problem.
Currently, the progresses in this area lie in the following.
  1. Cooperative Multiple Points (CoMP) Transmission: it involves the coordination among multiple cells and has been proposed as a contribution to the standardization of LTE-A. As a key focus, interference coordination schemes have been studied to improve the performance of cell-edge users. While extended to heterogeneous networks with relay and femto-cells, few results have ever been proposed in terms of global interference/resource management, especially containing dissimilar types of users/traffic.
  2. Power Saving Solutions: One direction lies in the usage of relay and femto-cells to reduce the transmitting distance, hence saving power. This further complicates the interference management and thus requires new investigations. The other one exploits the network resource to perform the load-balanceing among multiple sites via BS switching on/off, namely self-organizing networks (SON) in cellular scenarios. Current results always assume the SON knows the perfect global information. How the SON works with the fuzzy information and the constrained signaling remains open, and when a BS in SON wakes up from sleeping under imperfect information leaves unclear as well.
In summary, in heterogeneous multi-cell OFDMA scenarios, the relay and femto-cells may be deployed; the traffic load may not spatially homogeneous, and the users may have different QoS constraints, which also varies with time. While considering more practical factors, to adapt the network dynamics and then to maximize the system-wised throughput along with power saving in cellular access networks, optimal resource allocation approaches require deep investigations.

Research activities

  1. To propose a distributed and dynamic subchannel and power allocation scheme for heterogeneous multi-cell OFDMA systems.
  2. To propose innovative power saving schemes for practical purpose



To be updated.


Powered by Drupal, an open source content management system