Multiobjective optimization of fractional frequency reuse for irregular OFDMA macrocellular deployments

• Published in: Telecommunication systems 2016-04, Vol.61 (4), p.659-673
• Main Authors: González G., David, García-Lozano, Mario, Ruiz, Silvia, Lema, María A., Lee, Dongseop
• Format: Article
• Language: English
• Subjects: Algorithms; Artificial Intelligence; Business and Management; Cellular; Computer Communication Networks; Comunicacions mòbils; Design analysis; Design engineering; Electric communication systems; Energy consumption; Enginyeria de la telecomunicació; FFR; Fractional frequency reuse; Frequency distribution; Frequency reuse; Interference; IT in Business; Layouts; Long term evolution; Long-Term Evolution (Telecommunications); Long-term evolution (Telecomunicacions); LTE; Multiobjective optimization; Multiplexers; Network management systems; Optimization; Optimization techniques; Probability Theory and Stochastic Processes; Quality of service; Radiocomunicació i exploració electromagnètica; Studies; Telecommunications systems; Wireless networks; Àrees temàtiques de la UPC
• ISSN: 1018-4864; 1572-9451
• DOI: 10.1007/s11235-015-0060-3

Interference mitigation has been identified as a key challenge for emerging cellular technologies based on Orthogonal Frequency Division Multiple Access, such as Long Term Evolution. In this context, static intercell interference coordination including Fractional Frequency Reuse (FFR) have been adopted by mobile operators as a good alternative to improve the quality of service at cell edges. Nevertheless, recent results made evident the need for additional research efforts as default FFR configurations only offer tradeoffs in which spectral efficiency is severely penalized. Moreover, the performance of such baseline designs has been showed to be poor in realistic cellular deployments featuring irregular cell patterns. This paper solves this problematic by introducing a novel multiobjective optimization framework based on evolutionary algorithms that jointly takes into account system capacity, cell edge performance, and energy consumption. With respect to important reference schemes, the proposed algorithm succeeds in finding FFR configurations achieving gains between 10 and 40 % in terms of system capacity while simultaneously improving cell edge performance up to 70 %.