Wireless communication has revolutionised the pace at which information is exchanged. However, today’s communication networks will soon be overwhelmed by the demand for new services such as HD mobile streaming, device-to-device communications or the advent of the IoT. These new services require that the total capacity of wireless networks be increased.
However, because of the nature of radio waves and radio wave propagation, having more devices in an already cluttered environment will only generate more interference that will ultimately degrade the performance achievable in future networks. One way to improve the performance of wireless networks and to cope with the increased demand in throughput is then to find a better technique than what is currently in use to mitigate the interference generated by all the devices that will populate the network.
This talk will be about such a technique namely Interference Alignment (IA). This impressive technique has been proposed by Cadambe and Jafar in 2008 and was shown to achieve the Degree-of-Freedom of the Interference Channel (IC) meaning that it achieves the best possible performance of the IC at high signal-to-noise ratio. This technique came as a surprise to the entire research community as it showed that instead of splitting the total available ressource (bandwidth) evenly between all the users in the network as for the current techniques, it was actually possible to give half the total ressource to everyone regardless of their number. Because of that this new approach was called the “Half the cake” approach.
During the talk the emphasis will be on the performance of IA given the limited precision of practical channel estimation methods. We will consider different case for the Channel State Information (CSI) error, in the case where the CSI is bounded, we’ll derive a lower bound for the maximum achievable rate using IA. In the case of unbounded but Gaussian error we’ll derive a metric called outage probability that will be applied to (1) Assess the performance of the network, (2) optimise the network, (3) Derive the coverage probability in cellular network with some help from stochastic geometry.
The duration of this talk will be of approximately 45min
Raoul Guiazon graduated in 2013 with an MSc degree in Communications and Signal Processing from Imperial College London, UK and a Diplôme d’Ingénieur from ENSEA Cergy, France. Currently, he is studying towards a PhD degree under the supervision of Pr. Kai-Kit Wong in the Communications and Information Systems group, Department of Electronic and Electrical Engineering, University College London (UCL). His research is sponsored by the Engineering and Physical Sciences Research Council (EPSRC) and British Telecom (BT). His research interests are Interference Alignment, Cognitive Radios, Massive MIMO, IoT, Information Theory.
