Performance Evaluation of SpectrumSharing between LTE Femtocell andGSM Networks

Abstract: In recent years, as the data traffic demand grows rapidly, GSM spectrum refarming is emerging as one cost effective solution for operators to improve its efficiency on spectrum utilization. Among several refarming strategies, a coexistence network where LTE femtocells share the radio spectrum with the existing GSM users is regarded as one with great potential in the future practice. One focus of the coexistence network research is to study the mutual impacts between the legacy GSM network and the newly deployed LTE femtocells on each other’s performance. Among all the aspects of the study, one major concern is about how the LTE femtocells’ performances are affected by the GSM deployment conditions. Targeting this issue, a coexistence network simulator is created and performance metrics are defined in this thesis. Having decent GSM service quality as a precondition, the impacts of GSM network conditions (cell size, frequency reuse factor, and traffic load) on the LTE femtocells’ performance are investigated by simulation. In order to increase the LTE’s utilization of GSM spectrum without degrading the legacy GSM service, the interference between GSM and LTE users should be properly restrained, which attaches great importance to the research on interference management methods. In this thesis, an approach based on LTE femtocell exclusion region is studied and presented. Aiming at reducing the LTE’s interference to GSM, we delimit a region on the edge of each GSM cell, where no LTE femtocell is allowed to be deployed. Simulation results show that the approach lowers the LTE’s interference to GSM network and can greatly improve the LTE femtocells’ overall performance. Another approach to controlling the interference is to devise advanced channel allocation schemes. In this thesis, we propose a novel 6-sector frequency planning scheme which divides a GSM macrocell into six sectors and allocates the available GSM spectrum to each LTE femtocell according to the information on which sector it is located inside. Simulation results show that our scheme substantially outperforms the previously proposed one in terms of supporting higher LTE’s overall utilization of GSM spectrum. Besides, this channel allocation scheme enables the operator to make a trade-off between overall spectrum utilization and individual femtocell capacity based on the real data traffic demands by adjusting the frequency utilization factor (FUF) adopted in the scheme.