中文摘要：

最小二乘格型（Least Squares Lattice,LSL）自适应均衡算法具有收敛速度快、算法复杂度低、对舍入误差不敏感等特点,适于应用在水下实时通信中。在LSL均衡算法的基础上,为增强对多普勒频移的跟踪能力,在均衡器中加入等效二阶锁相环（PLL）,提出了两种算法：后置PLL的LSL算法（LSL-PLL-post）和预置PLL的LSL算法（LSL-PLL-pre）。对RLS、LSL、LSL-PLL-post和LSL-PLL-pre四种算法进行了仿真,结果表明：PLL显著增强了LSL均衡器对多普勒频移的跟踪能力;在多径和多普勒环境下,LSL-PLL-pre是四种算法中性能最好的。湖试数据分析表明,LSL-PLL-pre算法性能略优于LSL-PLL-post算法,更具实用价值。

英文摘要：

Least squares lattice（LSL） adaptive equalizer brings some advantages such as fast rate of convergence,lower computational complexity,and insensitiveness to round off errors,etc.Consequently,it is suitable for real-time underwater communications.In order to achieve better tracking to Doppler frequency shift,an equivalent second order phase-locked loop（PLL） is inserted into LSL equalizer,and two kinds of algorithms are proposed,which are named as LSL with post-deposited PLL（LSL-PLL-post） and LSL with pre-deposited PLL（LSL-PLL-pre）,respectively.The RLS,LSL,LSL-PLL-post and LSL-PLL-pre algorithm are simulated,showing that the PLL brings distinct improvement of Doppler tracking ability of LSL adaptive equalizer.And LSL-PLL-pre algorithm is the best of four in circumstance of multi-path and Doppler shift.Furthermore,LSL-PLL-pre algorithm is proved practical and slightly better than LSL-PLL-post algorithm with analysis of experimental data from lake trial.