Multi-Node Efficient Activation Scheduling and Global Estimation Method Based on Spatio-Temporal Scope Joint Information Model

For Internet of Things (IoT) applications involving the deployment of multiple sensing nodes, this paper establishes a multi-node Spatio-temporal Scope Joint Information Model (SSJIM) based on spatio-temporal correlations between nodes to quantify the regional value of information. An efficient node activation and global data estimation algorithm is proposed. Specifically, the mathematical expression of multi-node joint scope effective information is established from the perspective of information entropy and conditional entropy. Subsequently, a locally optimal iterative algorithm is designed to make decisions on the selection of optimized node sets. In the design of the global estimation model, the approach primarily relies on the Masked Autoencoder (MAE) architecture with modifications. Furthermore, a tandem queuing model is developed to model the bidirectional communication process involved in acquiring sensor data. Two communication time models are considered, and the mean steady-state increment of the Age of Information (AoI) is derived. Simulation results are found to be consistent with the theoretical derivations. Experimental results demonstrate that the proposed approach exhibits significant performance advantages, which are especially pronounced when only a small number of nodes are activated in partial data scenarios. Additionally, experiments indicate that the SSJIM can further enhance the accuracy of the global estimation model.