We develop a robust data fusion algorithmfor field reconstruction of multiple physical phenomena.The contribution of this paper is twofold: First, wedemonstrate how multi-spatial fields which can have anymarginal distributions and exhibit complex dependencestructures can be constructed. To this end we developa model where a latent process of these physicalphenomena is modelled as Multiple Gaussian Process(MGP), and the dependence structure between these phenomena is captured through a Copula process. This model has the advantage of allowing one to choose any marginal distributions for the physical phenomenon.Second, we develop an efficient and robust linear estimation algorithm to predict the mean behaviour of the physical phenomena using rank correlation instead of the conventional linear Pearson correlation. Our approach has the advantage of avoiding the need to derive intractable predictive posterior distribution and also has a tractable solution for the rank correlation values. We show that our model outperforms the model which uses the conventional linear Pearson correlation metric in terms of the prediction mean-squared-errors (MSE). This provides the motivation for using our models for multimodal data fusion.
|Title of host publication||2019 IEEE International Conference on Communications (ICC): Ad Hoc and Sensor Networks Symposium|
|Publication status||Accepted/In press - 3 Feb 2019|
Zhang, P., Peters, G. W., & Nevat, I. (Accepted/In press). Multimodal Data Fusion of Non-Gaussian Spatial Fields in Sensor Networks. In 2019 IEEE International Conference on Communications (ICC): Ad Hoc and Sensor Networks Symposium IEEE.