GMSVM-Based Prediction for Temporal Data Aggregation in Sensor Networks
Jian Kang, Liwei Tang, Xianzhang Zuo, Xihong Zhang, Hao Li · 2009
Data aggregation is a current hot research area in sensor networks. Aiming at the time series data in sensor networks, we present GMSVM (Grey Model Support Vector Machines), a novel prediction model data aggregation of sensor networks. In this model, grey model (GM) prediction theory is introduced into support vector machines (SVM). And the RBF kernel function is improved by Riemannian geometry analysis and the experimental data series, which can raise the arithmetic speed. The model is validated with fuel pressure data of injector. The results show that the model can execute dynamic multi-step prediction, and it has high precision prediction and flexibility. Thus, it can observably reduce the number of transmissions in sensor networks and save energy. Besides, it also has better performance in latency and computation. Comparing with other prediction algorithms, GMSVM is more effective for senor networks, so it has a good foreground to improve the prediction performance of data aggregation.