[Paper Review] Artificial Intelligence for Wireless Connectivity and Security of Cellular-Connected UAVs.

Ursula Challita, Aidin Ferdowsi+2 more|arXiv (Cornell University)|Apr 15, 2018

UAV Applications and Optimization18 references21 citations

TL;DR

This paper proposes artificial neural network (ANN)-based solutions to address wireless connectivity and security challenges in cellular-connected UAVs, including interference, mobility management, and cyber-physical attacks. The approach enables real-time, adaptive resource allocation and secure operations across UAV applications like delivery, video streaming, and intelligent transportation, with simulation results demonstrating improved performance in key metrics.

ABSTRACT

Cellular-connected unmanned aerial vehicles (UAVs) will inevitably be integrated into future cellular networks as new aerial mobile users. Providing cellular connectivity to UAVs will enable a myriad of applications ranging from online video streaming to medical delivery. However, to enable a reliable wireless connectivity for the UAVs as well as a secure operation, various challenges need to be addressed such as interference management, mobility management and handover, cyber-physical attacks, and authentication. In this paper, the goal is to expose the wireless and security challenges that arise in the context of UAV-based delivery systems, UAV-based real-time multimedia streaming, and UAV-enabled intelligent transportation systems. To address such challenges, artificial neural network (ANN) based solution schemes are introduced. The introduced approaches enable the UAVs to adaptively exploit the wireless system resources while guaranteeing a secure operation, in real-time. Preliminary simulation results show the benefits of the introduced solutions for each of the aforementioned cellular-connected UAV application use case.

Motivation & Objective

• To identify and address wireless and security challenges in cellular-connected UAVs across key application domains.
• To develop real-time, adaptive resource allocation mechanisms for UAVs using artificial neural networks.
• To ensure secure operation against cyber-physical attacks and authentication threats in UAV-based systems.
• To evaluate the performance of ANN-based solutions in UAV delivery, multimedia streaming, and intelligent transportation systems.

Proposed method

• Designing ANN-based frameworks to dynamically manage wireless resource allocation in response to changing UAV mobility and channel conditions.
• Integrating neural networks into the physical and medium access control layers to optimize signal-to-interference-plus-noise ratio and reduce handover latency.
• Implementing machine learning models to detect and mitigate cyber-physical attacks such as spoofing and jamming in real time.
• Training ANNs using simulation data from UAV mobility traces and network conditions to enable fast, low-latency decision-making.
• Using end-to-end learning to jointly optimize connectivity and security objectives in UAV networks.
• Validating the approach through simulations across three UAV application scenarios: delivery, real-time streaming, and intelligent transportation.

Experimental results

Research questions

• RQ1 How can artificial neural networks be leveraged to enhance real-time wireless connectivity for cellular-connected UAVs?
• RQ2 What are the key security threats in UAV-based cellular networks, and how can ANNs mitigate them effectively?
• RQ3 How do ANN-based solutions compare to conventional methods in managing interference and handover in high-mobility UAV environments?
• RQ4 To what extent can ANNs improve system reliability and security in UAV-assisted multimedia streaming and intelligent transportation systems?

Key findings

• The proposed ANN-based solutions significantly improve resource utilization and reduce handover latency in high-mobility UAV scenarios.
• Neural network models effectively detect and respond to cyber-physical attacks, enhancing system resilience.
• Simulation results show improved throughput and reduced outage probability in UAV-based delivery and video streaming applications.
• The integration of ANNs enables adaptive, real-time decision-making that outperforms traditional static or heuristic approaches.
• The approach maintains secure authentication and mitigates spoofing and jamming attacks with high detection accuracy.
• The framework demonstrates scalability and robustness across diverse UAV application use cases, including intelligent transportation systems.