Abstract
This paper is concerned with unmanned aerial vehicle (UAV) video coding and transmission in scenarios such as emergency rescue and environmental monitoring. Unlike existing methods of modeling UAV video source coding and channel transmission separately, we investigate the joint source-channel optimization issue for video coding and transmission. Particularly, we design eight-dimensional delay-power-rate-distortion models in terms of source coding and channel transmission and characterize the correlation between video coding and transmission, with which a joint source-channel optimization problem is formulated. Its objective is to minimize end-to-end distortion and UAV power consumption by optimizing fine-grained parameters related to UAV video coding and transmission. This problem is confirmed to be a challenging sequential-decision and non-convex optimization problem. We therefore decompose it into a family of repeated optimization problems by Lyapunov optimization and design an approximate convex optimization scheme with provable performance guarantees to tackle these problems. Based on the theoretical transformation, we propose a Lyapunov repeated iteration (LyaRI) algorithm. Both objective and subjective experiments are conducted to comprehensively evaluate the performance of LyaRI. The results indicate that, compared to its counterparts, LyaRI achieves better video quality and stability performance, with a 47.74% reduction in the variance of the obtained encoding bitrate.
Problem Context: UAV Video Transmission
Transmitting high-quality video from a UAV in real-time—whether during a forest fire or a search-and-rescue mission—requires a delicate balancing act. UAVs are inherently constrained by limited battery life and volatile wireless channel conditions.
Traditionally, "source coding" (video compression) and "channel transmission" (data delivery) are treated as separate problems. However, these processes are deeply coupled. For instance, a complex compression algorithm might reduce bandwidth requirements but drain the UAV's battery faster due to high CPU usage. Conversely, simpler compression preserves battery but produces a massive bitrate that can clog a weak wireless link, leading to lag or stalling. This paper addresses the lack of a unified mathematical model that captures these trade-offs, specifically for the High Efficiency Video Coding (HEVC) standard.
Methodology: Joint Source-Channel Optimization
The researchers developed a comprehensive eight-dimensional delay-power-rate-distortion (d-P-R-D) model. This model maps how fine-grained parameters—specifically the quantization step ($Q$) and motion estimation search range ($\lambda$)—impact end-to-end distortion, power consumption, and latency.
To solve the resulting non-convex optimization problem, the authors proposed the Lyapunov Repeated Iteration (LyaRI) algorithm, which utilizes three core techniques:
- Lyapunov Optimization: This breaks down long-term, time-averaged goals (like maintaining a power budget over a flight) into a series of real-time decisions for each time slot.
- Virtual Queues: These track how much the current bitrate deviates from channel capacity, ensuring system stability.
- Successive Convex Approximation (SCA): Because video distortion math is non-convex, SCA approximates the problem into a simpler, convex form that can be solved efficiently in real-time using standard optimization tools.
Experimental Evaluation
The approach was evaluated using standard HEVC test sequences, including "City" (simulating a moving camera over urban targets) and "Coastguard" (simulating a fixed camera tracking moving targets). These were processed using the HM-16.20 (HEVC Test Model) encoder.
Key Results
- Bitrate Stability: LyaRI achieved a 47.74% reduction in the variance of the encoding bitrate compared to standard benchmarks. This leads to significantly more predictable network behavior and prevents "buffer bloat."
- Subjective Quality: Visual tests demonstrated that LyaRI avoids the artifacts and stalling common in the standard HEVC Rate Control (HM RC) algorithm. As UAV-to-ground distance increases and signals weaken, LyaRI gracefully adjusts parameters to maintain image clarity.
- Power-Coding Trade-off: The research confirms that efficient coding isn't always optimal; high-efficiency coding can be as power-hungry as increasing radio transmit power. LyaRI successfully navigates this Pareto frontier to optimize for both battery life and video quality.
Conclusion and Future Work
LyaRI provides a robust framework for stable, high-quality UAV video transmission under strict resource constraints. While the current model focuses on a single UAV-to-Ground link, future research may extend this to multi-UAV swarms. Additionally, the authors suggest that future iterations could integrate AI to better predict channel conditions and video content complexity.
Summary updated