Network dimensioning with carrier aggregation

A recent policy ruling by the Federal Communications Commission (FCC) set aside a fixed amount of cleared spectrum for smaller network providers. Thanks to this ruling, smaller providers can improve their quality of service using carrier aggregation. In this paper, we determine the optimal (minimum) level of carrier aggregation that a smaller provider needs in order to bring its service in line with a larger provider in the same market. Toward this end, we provide an asymptotically exact formula for the loss (blocking) probability of flows under a quality-driven (QD) regime. Using this formula, we establish an efficient way of numerically calculating the optimal level of carrier aggregation and derive scaling laws. Specifically, we show that the optimal level of carrier aggregation scales sub-linearly with respect to the scaling factor, i.e., the ratio between the network capacities of the two providers, and decreases with the initial traffic load of the providers. We derive a closed-form linear upper bound on the optimal level of carrier aggregation and prove that it is the tightest possible. We provide numerical results, showing the accuracy of our methods and illustrating their use. We also discuss the extension of our results to delay-related metrics as well as their application to profitable pricing in secondary spectrum markets.