Smart packet prioritization in cognitive radio networks for smart agriculture

Abstract

The quick evolution in the field of Smart Agriculture, along with the Internet of Things (IoT), has created the need for developing wireless communication paradigms with the ability to handle heterogeneous data for different applications with different degrees of priority. This paper proposes an autonomous packet priority management framework for IEEE 802.11af-based CRNs to meet the heterogeneous data demands of Smart Agriculture and IoT. By employing a Dueling Double Deep Q-Network (D3QN) with a new dynamic aging threshold, the model avoids data starvation while ensuring reliable transmission of high-priority agricultural data over TVWS. Simulation results demonstrate that the D3QN framework achieves significantly better performance compared to standard and DQN-based models, especially in highly congested conditions (λ > 8). The proposed scheme reduces average delay for high-priority packets by 16.1% compared to the Baseline and by 10.5% compared to standard DQN under high-load conditions, and achieves approximately 14.8% reduction in relative energy consumption per high-priority packet, while maintaining comparable throughput. These results demonstrate a more robust and energy-efficient solution for real-time intelligent farming environments.