Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.559706
Title: Interference characterisation and mitigation in mobile broadband wireless networks
Author: Halls, David Edwin
Awarding Body: University of Bristol
Current Institution: University of Bristol
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
Abstract:
Multiple-Input Multiple-Output (MIMO) techniques are widely touted as the technology that will enable the high wireless cellular network capacities demanded by the huge growth in demand for the 'triple play' of voice, data and media. Without suitable interference management, however, multi-billion dollar MIMO Mobile Wireless Broadband Networks (MWBNs) can collapse under the strain of heavy traffic loads. Fully loaded interference studies cannot be performed on the network until it has been fully deployed. As such, interference characterisation and mitigation must be accurately performed pre-deployment using detailed network simulators. The development of a detailed MWBN simulator is investigated in this thesis. The model includes an extremely Uplink (UL) and Downlink (DL) Mobile WiMAX system-level simulator with full support for a wide range of MIMO technologies. The impact of inter-cell interference is characterised and results show that it is the fluctuation in interference power, rather than signal power, that dominates the inaccuracies seen in the link adaptation algorithm. These error lead to reduced system throughput. It is found that the Modulation and Coding Scheme (MCS) errors increase with Channel State Information (CSI) delay and that the effect is exacerbated by high mobile velocity. With a 3-frame delay at vehicular speeds the CSI delay causes an average 31 % loss in DL throughput. It is found that 2 x 2 closed-loop MIMO systems can double the system capacity in high interference conditions but the MIMO techniques in high interference are used to exploit diversity, and not multiplexing, gain. The gains provided by closed-loop MIMO are only available to slow moving or stationary users. Interference management techniques can be divided into: interference randomisation, interference cancellation and interference mitigation. It is shown that the use of interference randomisation reduces MCS error and improves user throughput. Significant performance gains are achieved in this work using higher order MIMO configurations and interference cancellation schemes. The gains are particularly significant at the cell edge on the DL. They are also effective on the UL, where unlike the DL they are robust to delayed CSI at the transmitter. A 5-fold increase over the single antenna case is obtained using a 2 x 8 MIMO system with interference cancellation. The use of interference coordination combined with interference cancellation further enhances performance, particularly on the DL. Adding the " combined mitigation scheme to the 4 x 2 MIMO case improves the average cell throughput by 70% and the cell edge throughput by 370%. This rigorous study has shown that to perform effectively in interference-limited scenarios, future MWBNs should employ 8x 2 MIMO with interference cancellation and interference coordination. This can provide Base Station (BS) throughput gains of up to 4-fold on the DL and 5-fold on the UL over a single antenna implementation.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.559706  DOI: Not available
Share: