Research On The Rate Adaptive Algorithm At Receiver Based On Rateless Error Correcting Codes

In the wireless communication,it is necessary to adjust the transmission rate according to the channel quality.The traditional rate adaptive technology obtains the knowledge of the channel state information(CSI)from the receiver by means of a feedback path,which may bring a significant amount of overhead and a high control complexity.Adaptive demodulation is a rate adaptive technology realized at the receiver,which needs neither the change of modulation mode at transmitters nor the feedback of CSI when the channel quality has changed.Compared with the rate adaption schemes realized at transmitters,the feedback cost is reduced,the implementation complexity at transmitters is decreased obviously,and channel states can be traced more quickly.Adaptive demodulation is generally combined with rateless codes because of a random deletion of the bits in demodulator at the receiver.The analysis and design of schemes in existing adaptive demodulation algorithms just consider hard decision before decoding.Two rate adaptive schemes at the receiver based on rateless error correcting codes and high order quadrature amplitude modulation(QAM)are given,which consider soft decision and aim at the error performance after decoding in the analysis and design.The main work is as follows:1.The decoding complexity of rateless codes is proportional to the decoding codeword length.To control the decoding complexity,a rate adaptive scheme using rateless error correcting codes at the receiver is given,which chooses demodulated bits from one symbol and considers the error performance after decoding.It includes multiple demodulation modes and can be applied to any higher order QAM.The bits with smallest log-likelihood ratio(LLR)in one symbol are deleted and the number of deleted bits varies in different demodulation modes,so the reliability and the average mutual information(MI)per demodulated bit can be boosted.Through the analysis of MI,the theoretical required codeword length for decoding which meets the requirements of error performance after decoding is obtained.Moreover,the selection of the demodulation mode under a decoding complexity constraint is given.A sample scheme employing 256-QAM modulation constellations and Raptor codes is provided,and the simulation results of this sample are consistent well with those of the theoretical analysis,which confirms the effectiveness of the scheme.2.A rate adaptive scheme using rateless error correcting codes at the receiver based on log-likelihood ratio demodulation threshold is proposed.The bits in the received symbols with absolute LLR values higher than the demodulation threshold are demodulated,otherwise,deleted.The demodulation threshold can be adjusted continuously according to the signal-to-noise ratio(SNR),and the average MI of the bits after demodulation can be kept constant in order to meet the required error performance with a fixed decoding codeword length.Based on the analysis of the average MI of the demodulation bits,the calculation method of LLR demodulation threshold and demodulation ratio under different SNR is given.A sample design scheme employing 256-QAM constellations and Raptor codes is provided,and the simulation results are consistent well with the theoretical analysis,which confirms the effectiveness of the scheme.