It is important to track the cognitive activity of service personnel in a Prognostics and Health Monitoring (PHM) related training or operation environment. EEG data is a good candidate for cognitive activity recognition. Analyzing EEG data in an unconstrained (natural) environment is a challenging task due to multiple reasons such as low signal-to-noise ratio, transient nature, lack of baseline and uncontrolled mixing of various tasks. This paper proposes a framework based on deep learning using both deep belief network (DBN) and deep convolutional neural network (DCNN) that monitors cognitive activity by fusing multiple non-collocated EEG probes and also selects a smaller sensor suite for a lean data collection system. Validation on realistic data along with comparison with benchmark machine learning techniques are performed. It is observed via sensor selection that a significantly smaller EEG sensor suite can perform at a comparable accuracy as the original sensor suite.