Design And Implementation Of Digital Equalizer Algorithm In Gigabit Ethernet PHY

With the continuous innovation of network information technology,Ethernet has entered the Gigabit era.Gigabit Ethernet PHY is a very important chip in Gigabit Ethernet system,in which digital equalizer plays an important role,responsible for equalizing channel attenuation and noise interference.Due to the complex channel environment,there are three digital equalizers that need to be designed in the Gigabit Ethernet PHY,all of which use the same algorithm to adjust the tap coefficients,the channel attenuation equalizer is the most complex and important one.This thesis studies the channel attenuation equalizer in 1000BASE-T system,establishes model of UTP-Cat.5 using Wideband model,and analyzes transmission characteristics of the channel.The characteristics of various equalizer structures are compared,and the equalizer combining feedforward equalizer and decision feedback equalizer is finally selected.At present,the digital equalizer in 1000BASE-T system is realized by DDLMS algorithm,but the performance of DDLMS algorithm is relatively poor compared to other blind equalization algorithms.DDCMA algorithm suitable for Gigabit Ethernet signal characteristics is proposed in order to improve the performance of equalizer.This algorithm retains the characteristics of CMA algorithm and has better performance than DDLMS algorithm,but its residual error is larger.Therefore,combination blind equalization algorithm based on residual error is proposed,which combines the advantages of DDLMS algorithm and DDCMA algorithm,and its performance better than DDLMS algorithm.The performance of DDLMS algorithm,DDCMA algorithm and combination blind equalization algorithm are compared through MATLAB.The result shows that the performance of two algorithms proposed in paper is better than DDLMS algorithm.Three algorithms of digital equalizers are designed through Vivado,and the reliability of MATLAB simulation results is verified when strict timing control conditions,and the use of resources by the equalizers of different algorithms is analyzed.Finally,The simulation results show that the proposed algorithms can replace DDLMS algorithm.