Artificial Neural Networks are computing methods that remember
the human brain. Currently, they are being applied for solving
problems in several areas such as, robotics, image processing,
pattern recognition, etc.... The Multi Layer Perceptron is a
neural network model that is widely used due to its simple learning
algorithm and its very good results. This model shows a inherent
massive parallelism. To take advantages of this parallelism, a
parallel implementation should be performed. Hardware implementations
are very interesting ones due to its high performance and parallel
capabilities. This paper presents a reconfigurable hardware parallel
implementation for Multilayer Perceptron handled by embedded
processors, which can add more flexibility to the system.
