摘要
泄露变步长最小均方算法是一种改进型LMS算法,克服了LMS算法无法同时兼顾收敛速度和稳态误差的的固有缺陷。提出一种基于泄露变步长LMS算法的汽车车内噪声主动控制方法,并将基于LMS算法和泄露变步长LMS算法的汽车车内噪声主动控制结果进行比较,结果表明:与LMS算法相比,泄露变步长LMS算法具有更快的算法收敛速度和较小的稳态误差,可有效进行汽车车内噪声主动控制。
Leakage variable step least Mean Square （LMS） algorithm is an improved LMS algorithm and overcomes the inherent defect of LMS algorithm, the inability to balance between convergence speed and steady-state error. Leakage variable step LMS algorithm-based active noise control method for vehicle interior noise is presented. Comparison of active noise control results which are obtained by using the LMS algorithm and leakage variable step LMS algorithm shows that leakage variable step LMS algorithm has a faster convergence rate and a smaller steady state error than LMS algorithm. It is an effective method in the active noise control for vehicle interior noise.
作者
王开轩
张心光
WANG Kaixuan;ZHANG Xinguang(Automotive Engineering College, Shanghai University of Engineering Science, Shanghai 201620, China)
出处
《机电设备》
2018年第1期22-25,共4页
Mechanical and Electrical Equipment
基金
国家自然科学基金项目(51609132)
国家自然科学基金项目(51675324)
上海工程技术大学启动基金(2015-66)
上海高校青年教师培养资助计划(ZZGCD15044)
关键词
泄露变步长算法
算法收敛速度
稳态误差
噪声主动控制
leakage variable step LMS algorithm
algorithm convergence speed
steady-state error
noise active control
