Entropy‐Greedy Node Selection Algorithm in Spectrum Map Construction

Spectrum maps are visualization tools that reflect the underlying spectral environment, enabling advanced functions such as spectrum decision‐making and emitter identification. To enhance mapping accuracy and optimize resource utilization, this study addresses the sensor node selection problem in ground‐based sensing scenarios. We propose an entropy‐greedy node selection (EGNS) framework that employs a two‐stage scheduling strategy: the first stage performs coarse sensing via spatial sector partitioning to obtain an initial estimate of emitter locations, and the second stage executes an enhanced greedy selection algorithm to iteratively minimize the signal reconstruction error. Simulation results on real‐world spectrum datasets show that the proposed method achieves superior reconstruction accuracy and lower sensing costs compared to conventional sampling approaches, making it well‐suited for dynamic electromagnetic monitoring applications under constrained budgets.