Parameter Estimation For Broadband Channel System Based On SAGE Algorithm

Due to the dynamic characteristics of the radio channel, the radio propagation environment parameters change quickly. The emergence of many advanced technologies need to be pre-effective prediction of channel parameters, such as multi-user detection technology, balancing technology. The forecast channel parameters dependent on efficient and accurate parameter estimation method. Order to improve radio channels of communication channel can guarantee the effective transmission of the signal, and need effective algorithm for fast and accurate estimates of the channel parameters of channel parameter estimation on the the effective transmission signal of the wireless network has a very important significance.This paper studies the various parameters of the transmission channel (complex amplitude, delay, incidence angle, left angle, Doppler shift) estimation algorithm. These algorithms can be roughly divided into three categories:spectral estimation, the parameter subspace estimates and determine parameter estimation. More common in the first category is the MUSIC (Multiple Signal Classification) algorithm, the second category is the typical ESPRIT (Estimate of Signal Parameter via Rotational Invariance Techniques) algorithm. The third category is the typical expectation maximum (EM-Expectation Maximization) algorithm scalability and EM algorithm SAGE (Space-Alternating Generalized EM) algorithm. EM algorithm, SAGE algorithm compared to reduce the amount of computation and speed up the convergence, allowing more precise parameter estimates, improve system signal-to-noise ratio. This paper focuses on the principle of the SAGE algorithm, using the SAGE algorithm to simulate analog channel parameters, and verify the feasibility and correctness of the algorithm.