Search

This paper mainly deals with the statistical processing methods in wireless filed strength prediction process after sampling data are collected. The conventional standard for data collection and statistical processing in the prediction of wireless filed strength is Lee's criteria. It utilizes linear averaging to process sampling data, as many other related papers discussed. However, the accuracy of such first order moment estimation is low, especially under serious Rayleigh fading channels. As regards to the defects of the method, we propose four novel statistical processing methods to process the collected data for wireless field strength. Computer simulations and corresponding analysis verify its much better prediction results with small estimation variances. [ABSTRACT FROM AUTHOR]

Abstract  Polynomial predistortion techniques for power amplifier (PA) non-linearities based on indirect learning architecture (IDLA) are widely used. The benefit of the IDLA leaves unnecessary the assumption of a model for PA, corresponding parameters estimation and inverse construction. In this paper, a novel scheme based on IDLA for PA predistortion is proposed. It has better power spectral density (PSD) and relative mean square error (RMSE) performances than the conventional IDLA-based methods especially when the PA non-liearity is more severe. Simulations and theoretical analysis verify the good performances of the proposed scheme. [ABSTRACT FROM AUTHOR]