@article{YANG2026, 
author = {Jinggang YANG and Chengbo HU and Xueqiong ZHU and Zhen WANG and Hong LIU and Hui LI},
title = {User satisfaction optimization of power wireless sensor networks based on the D3QN algorithm},
year = {2026},
journal = {Electric Power Engineering Technology},
volume = {45},
number = {3},
pages = {57-62},
keywords = {power wireless sensor network (PWSN), orthogonal frequency division multiplexing (OFDM), user satisfaction, resource allocation, dueling double deep Q-network (D3QN), reinforcement learning},
url = {https://www.sciopen.com/article/10.12158/j.2096-3203.2026.03.007},
doi = {10.12158/j.2096-3203.2026.03.007},
abstract = {In power wireless sensor networks (PWSNs), concurrent uplink access by multiple users is constrained by limited spectrum and power resources, while heterogeneous monitoring services exhibit markedly different requirements in terms of reliability and latency. These factors make it challenging for resource scheduling to simultaneously satisfy overall system efficiency and user-perceived quality. In this work, a joint resource allocation mechanism capable of providing differentiated quality-of-service guarantees under heterogeneous service demands is formulated within an uplink orthogonal frequency division multiplexing (OFDM) framework. A quantifiable user-satisfaction function is designed, and the joint optimization of subcarrier and power allocation is modeled as a Markov decision process (MDP). A dueling double deep Q network (D3QN) algorithm is further introduced to dynamically adjust the allocation strategy. In addition, an action-space down-sampling mechanism is proposed to reduce computational complexity and enhance training efficiency. Simulation results demonstrate that the proposed algorithm achieves fast convergence under various node densities and subcarrier configurations, and yields significant improvements in user satisfaction compared with conventional DQN, random allocation, and uniform allocation methods.}
}