Download Now Free registration required
Parameter reduction from experimental data is an important issue arising in many frequently encountered problems with different types of applications in communications engineering. However, the computational effort grows drastically with the number of parameters in such types of applications. This paper proposes a technique that reduces the performance parameters of a communication system based on eigenvalues of covariance matrix as well as providing a weighted rank of parameters by an approach called Non-negative Matrix Factorization (NMF).
- Format: PDF
- Size: 343.6 KB