Channel adaptive estimation method based on multi-model weighted soft handover

Abstract

The invention belongs to the field of radio communication, and discloses a channel self-adaptive estimation method based on multi-model weighted soft handover. Firstly, the channel system model is established, the time-frequency dual channel is selected, and the channel parameters are determined. Then based on the channel model and estimation method, the channel estimation sub-model and multi-model channel estimation library are established, and the model error and estimation error of the channel estimation model are analyzed and calculated. Finally, according to the switching index combining model error and estimation error, the model switching is completed through the weighted multi-model adaptive estimation algorithm of LUMV. The present invention proposes that under the condition of uncertain transmission channel model, combined with channel model and estimation method, the multi-model idea in multi-model adaptive control theory is introduced into channel estimation, and a weighted multi-model with minimum variance of linear error is adopted An adaptive estimation method is used to complete the switching, so that the channel estimation method has high robustness and accuracy in the complex channel range.

Description

Technical field

[0001] The invention belongs to the field of radio communication and relates to a channel adaptive estimation method based on multi-model weighted soft handover. Background technique

[0002] The channel estimation algorithm designs the estimation algorithm according to different criteria for the established channel model to obtain model parameter values. Commonly used channel estimation algorithms include maximum likelihood ML estimation, EM estimation (EM algorithm is an iterative algorithm to achieve progressive ML estimation when the observation data is incomplete), LS estimation, RLS estimation, LMMSE estimation, Kalman filtering, etc. ML / EM estimation is based on the maximum likelihood criterion, which uses several known observations (complete or incomplete) to estimate the parameter without any prior knowledge of the estimated parameter, which is highly effective and complex The degree is also high; LS aims to minimize the square error between the estimate...

Images