Prediction in LMS-type adaptive algorithms for smoothly timevarying environments
IEEE Transactions on Signal Processing
| Hi-index | 0.00 |
Because of the time-varying characteristic of channel in HF communication, it is indispensable to track the channel for the receiver. In this paper, we give a joint iterative algorithm about LMS estimation and decoding based on sub-band adaptive filter. The detailed method is as follows: Firstly, by a shorter periodic training sequence to pre-estimate the channel, supposing that the channel parameter will not change when information sequences come, the information is decoded. Then, the decoded sequences will become the apriori input information of channel in order to track the channel in real time. The decoding process will continue until fulfilling the condition. The simulation result proves that the convergence rate of this optimized algorithm is raised apparently without increasing the complexity of algorithm.