ADS-B and CPDLC fault modeling for safety assessment in a distributed environment

Transitioning from the current air transportation system to the Next Generation Air Transportation System (NextGen) heavily depends on distributed surveillance and communication technologies. These technologies operate as a part of a complex distributed System-of-Systems (SoS) of avionics, pilots, aircraft, ground stations and air traffic controllers. A comprehensive understanding of the failure modes of the avionics in the SoS is essential to ensure safe operations, especially while implementing automated separation assurance algorithms. Our study investigates the impact of faults in Automatic Dependent Surveillance - Broadcast (ADS-B), Controller Pilot Data Link Communication (CPDLC), and their networked devices on the safety in airspace for distributed self-separation. Existing literature on faults for these technologies are spread across two categories: (1) observed faults reported from experiments, simulation, or real-world operations of the devices and (2) potential faults based on failure modes that generally lack a rigorous representation suitable for modeling and simulation. These faults are compiled in the form of a generalized fault map that abstracts them into a minimalistic set of fault models suitable for an agent-based model (ABM). The ABM simulations further demonstrate the degradation in safety due to faults in ADS-B as quantified by a set of safety metrics. Our results indicate faults that corrupt the position and heading data of the aircraft result in the greatest threat to safety during self-separation.