Learning Machine using Heterogeneous Neurons with High Dimensional Parameters

System of systems is inherently multidisciplinary while specific problems will require specific expertise but principle premise here is that common characteristics of all these large complex problems can lead to general tools and methodologies to support them In recent years there has been a growing need of significant methodologies in telecommunication systems, multimedia systems and vision systems. In SOS perspective the integration of above systems in intelligent infrastructure system is in early stage of development. Neural computing has been emerged as an important paradigm for prediction of complex dynamic behavior of intelligent systems. In this article we propose architectural diversity of neuron units and design learning machine using these neurons with high dimensional parameters which directly process high dimensional information in natural and efficient way hence significantly reduces time, space and connection complexities of the network.