Rate Fairness Based QoS Provisioning for Operators in 5G Shared Networks

In this work, we focus on two problems related to the rate fairness based SLAs (Service Level Agreements) in an environment where multiple operators are sharing the spectrum. At first we show analytically that achieving rate fairness among arbitrary number of operators on the smallest time-frequency radio resource (e.g. one OFDM resource element) is feasible. Then we transform the problem of rate fairness to power allocation problem and provide an analytical proof that a unique solution is possible for multiple operators. In addition to the analytical method, we also provide an algorithm which is capable of guaranteeing rate fairness for any arbitrary number of operators. In our algorithm, we use Newton-Raphson based numerical approximation which is well known for its fast convergence and simplicity. We also present simulation results for the assessment of our proposed approach and algorithm. The results show that rate fairness based SLAs can also achieve system spectral efficiencies as high as the resource fairness based SLAs. This provides an extra degree of freedom to the 5G service providers in terms of rate based QoS provisioning.