This paper presents a method of the reverse recognition of the check matrix coded by low density parity check code (LDPC), which is based on the posterior probability log likelihood ration (LLR). The posterior probability LLR was obtained by estimating the channel amplification and the variance of the noise in the received code. A reverse recognition of LDPC code check matrix was achieved following the LLR mean value maximization principle. Simulation results show that the receiver can precisely retrieve the LDPC check code used by the sender under different channel circumstances through the LDPC reverse recognition method.

In this paper, in-situ EPR, TPSR and NMR were used to study the Rh-Mn/SiO₂ catalyst for ethanol synthesis. The results show that the reactive metals Rh and Mn are fixed on the catalyst surface by the interaction with Si-OH, which can enhance the stability of highly dispersed metals. Rh and Mn interact with each other to form the acitve site with high ethanol synthesis activity. The surface Si-OH was found to be of significance on the preparation of catalyst with high catalytic activity.