This paper mainly studies and realizes the spatio-temporal multidimensional visualization under the big data environment. By analyzing the spatiotemporal correlation of wireless sensor networks under the random large-scale dense deployment of sensor nodes, ensure that there is no monitoring blind spot in the area, the accuracy of data sampling and the reliability of communication. By studying the Markov chain’s spatial correlation data prediction algorithm, it solves the problem that nodes need to store a large amount of historical data during prediction, and the requirements for node storage space are relatively high; At the same time, the spatial correlation of the data in the network is analyzed for data prediction, which solves the problem of low prediction accuracy using the time-based correlation prediction method in the case of irregular data fluctuations.