A deconvolution method to remove distortion caused by antenna radiation pattern from measurement

The influence of vegetation on radiowave signals has become an important aspect of the design of wireless communication links. In recent years the theory of Radiative Energy Transfer has been adopted as a reliable tool to predict the radiowave propagation through and near vegetation. However one major factor influencing accuracy of the measurement is the radiation pattern of the receiver antenna. The measured pattern will be the convolution product of the antenna radiation pattern and the phase function of the vegetation medium. The measured pattern therefore needs to undergo a deconvolution process before enable to provide reliable information. This paper presents a deconvolution method developed using optimum compensation filtering to remove the distortion caused by the receiver antenna radiation pattern. A pre-filtering technique using auto/cross-correlation is utilised to improve the deconvolution results, as well as an error function is deployed to determine the optimal parameter in the iterative filter.