Load-balancing algorithms for multifrequency data communications

This paper describes an approach for enhancing the network performance of the air-ground (A/G) portion of the National Airspace System (NAS) data communications (Data Comm) network when multiple frequencies are used per ground station (GS), and the expected aircraft and data traffic for each location is known. The main point made in this paper is that, in a scenario with multiple frequencies per GS, one can achieve better system performance by carefully allocating aircraft to each frequency, as opposed to having each frequency have an equal share of the traffic or by using one frequency to its maximum capacity before offloading to the next. The algorithms for assigning aircraft to GSs were developed initially with a focus for implementation on simulated scenarios, but could potentially be applied in actual field scenarios, a topic to be explored in future studies. The work described in this paper builds on the evolving Spectrum Dimensioning and Optimization framework developed at The MITRE Corporation [1]. The framework was designed to address the dimensioning and optimization of a radio frequency (RF) network for A/G Data Comm. The Data Comm program is a key part of the Federal Aviation Administration (FAA)'s Next Generation Air Transportation (NextGen) initiative to implement and modernize A/G communications and air traffic management (ATM) capabilities in the NAS.