We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Determination of Channel Capacity Using Shannon's Relation Through Cerebellar Model Arithmetic Computer Using Artificial Intelligence.
- Authors
Babu, M. Hareesh; Raju, D. S. S. N.; Sarvani, A.; Rao, M. Purnachandra
- Abstract
The paper is a novel and simple method to determine the channel capacity using software called MATLAB. The neural network called Cerebellar Model Arithmetic Computer (CMAC) is designed, which is applied through a technique called system identification. CMAC consists of hundreds of thousands of adjustable weights that can be trained to approximate nonlinearities which are not explicitly written out or even understood. CMAC has an architecture similar to that of cerebellum, a part of brain, which is built-in generalization and acts as associative memory. CMAC can learn nonlinear relationships using system identification technique. System identification is a technique of mapping between the inputs and outputs of the identification block, which takes the current input and output variables of the process as its inputs and gives the estimates of the system parameters. The system identification network is simulated using MATLAB which is implemented to estimate the parameters of the Shannon's channel capacity. The implementation and simulation of the CMAC network using the systems identification is mainly concentrated. The performance of the network is superior when the training data points are uniformly distributed over the entire input range rather than random points.
- Subjects
CHANNEL capacity (Telecommunications); CEREBELLUM; ARTIFICIAL neural networks; SYSTEM identification; MATLAB (Computer software); RANDOM noise theory; MATHEMATICAL models
- Publication
IUP Journal of Computer Sciences, 2017, Vol 11, Issue 3, p44
- ISSN
2583-441X
- Publication type
Article