Performance Evaluation of Device to Device Communications in Cellular Networks

KTHContact:Guowang Miao,

Homepage:

EricssonContact: Gábor Fodor,

Homepage:

Starting Time: Q1/2 2013

Background

Device-to-device (D2D) communications underlayedto a cellular infrastructure has recently been proposed to increase spectrum and energy efficiency. An example of D2D using downlink and uplink resources is illustrated in the following figure. D2D takes advantage of the physical proximity of communication devices and enables direct communications between devices. The proximity of devices may allow for extremely high bit rates, low delays, and low power consumption. Different from traditional cellular architecture, D2D requires new peer discovery methods, physical and MAC layer procedures, and radio resource managementalgorithms that help realize the potentialadvantages of D2D communications.

D2D Communications with Downlink and Uplink Resources [1]

The most recent trend in network assisted D2D communications is to allow for multiple hops and to form clusters of user equipments to cooperate in order to help another to communicate with each other and/or with the cellular infrastructure.

Thesis Work

The goal of the thesis is to investigate how D2D communications can be integrated with traditional cellular structures, with emphasize on cooperation techniques between the user equipments and infrastructure nodes from both algorithm and protocol perspectives. The final definition will be made together with the applicant. The following is what we currently expect in part of the work:

Literature survey of existing D2D communications and cooperative communication techniques and analyze the issues in integrating D2D in cellular networks;

Implement a simulation platform and observe the performance gain in spectral and energy efficiency.

Design or choose a centralized scheduling algorithm and implement it. Analyze the impact of different parameters on the performance.

Design or choose a distributed scheduling algorithm and implement it. Analyze the impact of different parameters on the performance.

Evaluate the impact of signaling overhead, user proximity, and traffic load on the network performance.

Qualifications

You should

  • be a bright Master of Science student in electrical engineering, applied physics or similar with excellent grades,
  • have taken courses and excel in wireless communications, wireless networks (signal processing a big plus),
  • master a certain programing language (Matlab or C/C++ preferred),
  • have a keen interest in telecom technologies, programming, and research,
  • have international experience (valuable, but not required),
  • have strong analytical skills,
  • speak and write outstanding English, and
  • be self-motivated, self-driven and communicative.

Practicalities

Scope: 30 points (20 weeks), 1 person
Target starting date: within Q1/2 2013, the exact dates will be set together with the selected candidate.
Location: Kista, Stockholm, Sweden

[1] Gábor Fodor, et al., Design Aspects of Network AssistedDevice-to-Device Communications, IEEE Communications Magazine,March 2012.

[2] Gábor Fodor, et al.,A Distributed Power Control Scheme for CellularNetwork Assisted D2D Communications, IEEE Globecom 2011.