Introduction: The Challenge of Dynamic C-V2X Environments
In Cellular Vehicle-to-Everything (C-V2X) networks, efficient communication relies heavily on accurate Channel State Information (CSI). Vehicles must share limited radio frequency resources for both Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) links. However, the high mobility of vehicles creates a rapidly changing wireless environment. Doppler shifts and multipath effects often render CSI "stale" or imperfect by the time it is utilized for resource allocation.
Traditional approaches to this problem typically rely on rigid model-based assumptions (e.g., Gaussian error distributions) or data-driven methods that may neglect performance degradation during the learning phase. This research proposes a shift toward resilience: the capacity of a system to absorb the impact of imperfect CSI and actively adapt to recover optimal performance.
Methodology: A Two-Phase Resilient Framework
The core of the proposed solution is a two-phase framework that decouples the complex task of power allocation under uncertainty into manageable "Absorption" and "Adaptation" phases.
1. Absorption Phase
During this phase, the system does not assume a pre-defined error model. Instead, it utilizes Received Signal Strength (RSS) measurements to estimate the actual Probability Distribution Function (PDF) of the CSI error. By applying deconvolution estimation theory, the framework extracts the "true" error signal from noisy measurements.
A critical innovation here is the use of a Hazard Rate (HR) metric. Borrowed from survival analysis, the HR helps the network balance a fundamental tradeoff: setting transmit powers to gather high-quality estimation data versus maintaining a minimum Quality of Service (QoS) for active vehicular links.
2. Adaptation Phase
Once the error PDF is accurately characterized, the system enters the adaptation phase. Armed with a custom-fit error model, the Roadside Unit (RSU) performs real-time power allocation. This statistical knowledge allows the system to optimize transmit power such that V2V latency requirements are strictly met while simultaneously maximizing V2I throughput.
Key Insights: Resilience over Reliability
The research highlights that in highly dynamic environments, striving for absolute reliability (avoiding failure) is often less effective than building resilience (managing recovery). By treating CSI mismatch as a deconvolution problem, the system can "see through" interference without prior assumptions about the channel's behavior. This allows the network to maintain a stable delay profile, effectively absorbing transient spikes that typically cause benchmark models to fail.
Performance Evaluation and Results
The framework was evaluated using high-fidelity simulations, including the Manhattan Mobility Model and WINNER II channel models, across various complex error distributions. Key results include:
- Latency Recovery: The framework reduced V2V delay exceeding QoS requirements by 35% compared to model-based approaches and 56% compared to standard data-driven benchmarks.
- Throughput Improvement: Average V2I throughput saw gains of 14% to 16%.
- System Stability: The resilient approach successfully "absorbed" transient spikes in channel error, maintaining performance where traditional models failed.
Limitations and Future Directions
While the framework shows significant promise, several areas remain for future exploration. Currently, link matching (pairing V2V and V2I links) remains static during the absorption phase; dynamic re-matching could further enhance performance in fluid traffic. Additionally, while the model handles short-term fading well, incorporating more dynamic models for large-scale fading and testing scalability in much larger vehicle clusters are logical next steps.
Summary updated