Research On Images Transmission Based On Systematic Raptor Codes Over Deep-space Channel

The information about images is very important in deep- space exploration, which contains the most intuitionistic demanding data in deep-space communications. But deep- space communicatio n has a long distance and large path- loss. It takes long time delay from data sending to receiving. The existing ARQ mechanism can not obtain high-effic iency performance in deep- space images transmission. In addit ion, digita l fountain codes is a new kind o f forward error correction codes which makes a good transmitting performance in deep-space environment with its characteristics of rateless and non- feedback. It is appropriate to be used in transmitting streaming media of deep-space exploration.This paper based on wavelet transformation of deep-space images processing. Protect the fundamental frequency component of images by introducing 3GPP(The 3rd Generation Partnership Project) systemic Raptor codes, after wavelet transform. From both choose weight sid e and choose number side respectively, we improve the performance of 3GPP systemic Raptor codes unequal error protect algorithm. Obtain a high- efficiency performance of images transmission when channel environment is poor. The main research work is divided into the following three parts:Firstly, basing on the view of selection weight, this paper proposed an UEP coding algorithm, which is suitable to image transmission in deep space channel. This scheme designs a kind of short intertwined LT coding, which is aimed at the fundamental frequency data after wavelet transform. Then, 3GPP systemic Raptor coding is cascaded to achieve a SISR(Short Intertwined Systematic Raptor) scheme, which has the property of UEP. The UEP ability of SISR coding is verified by analys is, coding complexity, system overhead and PSN R gains are comprehensive studied. By comparing w ith other relevant algorithms, this paper clarifies the applicable scope and performance of SISR algorithm. Optimization of parameters is also explored, whic h means how to choose optima l parameters according to original symbol.Secondly, basing on the view of the number of source data packets to partic ipating in encoding, this paper proposed an UEP coding scheme, which is suitable to image transmission in deep space channel. This scheme increases the probabilit y of important signals pa rticipating in encoding by using expanding window protection, which attains protection of important data and achieve expanding w indow systemic Raptor coding scheme. By us ing A nd-O r analys is, the theoretical upper bound of its performance is verified.. During the simulation experiment, the recovery probability of different origina l symbols and the restoration performance of image under different channe l conditions are observed. By comparing SISR with expanding w indow systemic Raptor coding scheme, this paper analyses characterist ics of these two algorithms and optimizes the whole deep space image transmission algorithm.At last, our proposed UEP coding algorithm are realized on the e xist ing deep space file transfer simulation platform, The simulation verifies the SISR algorithm and expanding window algorithm, and obtains the actual performance of the two algorithms in the existing deep space communication protocol system, which provides a useful reference for the further improvement of the deep file transfer protocol.