Application of bio-inspired data processing in intelligent transportation systems

There has been a growing interest in using the soft computing techniques for distributed data processing in monitoring systems during last years; wireless sensor network is an example which can be utilized to record and process the environmental conditions in transportation systems. In this paper, by using knowledge based soft computing techniques, the reliability of the sensory records is evaluated autonomously in an intelligent transportation system; for this purpose, a new combinational architecture is applied including a bio-inspired local/global data approximation as well as a knowledge based data classification mechanism. The approximation mechanism is established on a least squares mechanism for global data prediction and a radial basis function network including immunological center selection to perform the local data approximations. To classify the records, a new probabilistic neural network including immune system features is developed. The proposed data processing technique is applicable for various intelligent data processing purposes.