Download Now Free registration required
There has emerged a lot of focuses on sparse signal regressions among the researchers in the fields of signal processing and information theory. Although those works contain theoretical fundamentals, most of the algorithms are not tailored to time-varying environments with real - time requirements. Recently, Bajwa et al. used the Dantzig selector and Least - Squares (LS) for sparse channel sensing. Choi et al. investigated an Expectation-Maximization (EM) method for doubly-selective Multi-Input Multi-Output (MIMO) sparse channels. Although those methods offer good estimates with improved Mean-Square Error (MSE), the underlying sparsity is not fully exploited to reduce the complexity.
- Format: PDF
- Size: 299.6 KB