Research On Channel Modeling And Channel Equalization Algorithm For High Speed Serial Communication

With the rapid development of information technology, the demand of raising informationtransmission speed is higher, so high speed serial communication technology is widely applied inthe field of communication. It is a point to point communication technology, with high transmissionrate, strong anti-jamming capability advantages. But in the process of transmission, the signal maybe affected by noise, skin effect, dielectric loss and so on. These affects usually cause intersymbolinterference and influence the quality of signal. Therefore, channel equalization is a key technologyof high speed serial communication.Firstly, this paper outlines the high speed serial communication technology, establishes channelmodel based on characteristics of signal transmission, and sets up a simple communication systemwith introducing adaptive algorithms and updating coefficients of using decision feedbackequalizers. DFE is using linear combinations of data which have been recovered to eliminate intersymbol interference. In the research of the stability of adaptive algorithms, this paper mainly studiesthe features of FSS-LMS algorithm which using fixed step, SVS-LMS algorithm, G-SVSLMSalgorithm, LMS algorithm and VSS-LMS algorithm, then it proposes a new variable step algorithmbased on LMS which is called HT-VSS. The simulation results show that, HT-VSS algorithm haswell anti noise performance, good tracking performance, high speed of convergence, and it’s betterthan the other four algorithms in the comprehensive performances. In the experiment, HT-VSSalgorithm is embedded into the decision feedback equalizer, it reflects the better quality of signalswith the comparison of send and received signals and observation of eye signal. Finally, the papercompares the error bit rate of different decision feedback equalizers of different variable step LMSalgorithms, and the simulation results show that, the DFE equalizer based on HT-VSS algorithm hasgood performance and low error bit rate.