Multi-AAV-enabled Distributed Beamforming in Low-Altitude Wireless Networking for AoI-Sensitive IoT Data Forwarding

2025-09-01

AI Summary

To address high Age of Information (AoI) and low service reliability in Aerial Access Vehicle (AAV)-assisted IoT data transmission within low-altitude wireless networks, this paper proposes a multi-AAV collaborative distributed beamforming framework that jointly optimizes trajectory planning and communication scheduling. We design a novel deep reinforcement learning algorithm—SAC-TLS—integrating temporal normalization via GRU and a Squeeze-and-Excitation module to enhance long-horizon decision stability and accuracy. Distributed beamforming is employed to boost channel gain, thereby mitigating return-delay caused by coverage limitations. Experimental results demonstrate that the proposed method significantly outperforms baseline algorithms in average AoI reduction, energy efficiency, and convergence speed, effectively balancing data freshness and energy consumption.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Multiagent Systems: Agent Communication
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

With the rapid development of low-altitude wireless networking, autonomous aerial vehicles (AAVs) have emerged as critical enablers for timely and reliable data delivery, particularly in remote or underserved areas. In this context, the age of information (AoI) has emerged as a critical performance metric for evaluating the freshness and timeliness of transmitted information in Internet of things (IoT) networks. However, conventional AAV-assisted data transmission is fundamentally limited by finite communication coverage ranges, which requires periodic return flights for data relay operations. This propulsion-repositioning cycle inevitably introduces latency spikes that raise the AoI while degrading service reliability. To address these challenges, this paper proposes a AAV-assisted forwarding system based on distributed beamforming to enhance the AoI in IoT. Specifically, AAVs collaborate via distributed beamforming to collect and relay data between the sensor nodes and remote base station. Then, we formulate an optimization problem to minimize the AoI and AAV energy consumption, by jointly optimizing the AAV trajectories and communication schedules. Due to the non-convex nature of the problem and its pronounced temporal variability, we introduce a deep reinforcement learning solution that incorporates temporal sequence input, layer normalization gated recurrent unit, and a squeeze-and-excitation block to capture long-term dependencies, thereby improving decision-making stability and accuracy, and reducing computational complexity. Simulation results demonstrate that the proposed SAC-TLS algorithm outperforms baseline algorithms in terms of convergence, time average AoI, and energy consumption of AAVs.

Problem

• Overcoming limited communication range in AAV-assisted IoT data transmission
• Minimizing Age of Information and energy consumption in aerial networks
• Addressing latency spikes from AAV repositioning cycles through beamforming

Innovation

• Distributed beamforming for AAV collaboration
• Deep reinforcement learning with temporal sequence
• Joint optimization of trajectories and schedules