An Improved Variable Step-Size LMS Algorithm
15
Citation
7
Reference
10
Related Paper
Citation Trend
Abstract:
To improve the performance of LMS adaptive algorithm, in this paper a new variable step-size LMS algorithm is proposed based on the analysis of some variable step-size algorithms. Through establishing a new nonlinear relationship between the step size and the error, the algorithm eliminates the irrelevant noise and improves the convergence rate to obtain a better stability. And the computer simulation results are consistent with the theoretical analysis, which confirmed that the algorithm is superior to other algorithms on convergence rate, tracking speed and steady-state error.Keywords:
Algorithm design
For existing LMS algorithm could not simultaneously improve the convergence speed and lower steady-state error of contradictions.By building a nonlinear functional relationship between μ(n) and e(n),this paper proposed an improved variable step size LMS adaptive filtering algorithm.Compared with existing algorithms,while the introduction of memory factor λ and control functions values of the parameters β(n),so that the current iteration step were related to the previous step and the former M the square of error.Theoretical analysis and computer simulations show that several common with existing LMS algorithm,the improved convergence rate and steady-state error performance is improved.
Adaptive algorithm
Cite
Citations (2)
Abstract A new least-mean-squares-(LMS-)type algorithm that employs a time-varying variable step size in the standard LMS weight update recursion is introduced in this paper. The work is aimed at improving the directional estimate, in searching for the global minimum of the mean-square-error surface, in an effort to increase the algorithm's speed of convergence. First-order convergence analysis of the algorithm is developed. Additionally, expressions for the ith time constant and algorithm misadjustment are introduced. Several simulation examples are presented to compare the new algorithm with the LMS and other existing variable step size algorithms. Comparisons illustrate the new algorithm's possession of better convergence properties, under stationary and non-stationary signal conditions, when compared with the other algorithms. Notes Fax: +(962)(2) 295 123
Constant (computer programming)
Degree (music)
Cite
Citations (14)
This paper presents a simple and robust variable step-size normalized LMS (VSS-NLAIS) adaptive algorithm. The fixed step-size NLMS algorithm (FSS- NLMS) usually results in a trade-off between the residual error and the convergence speed of the algorithm. The variable step-size NLMS algorithm presented here relaxes such trade-off. Both analysis and simulation results show that the proposed VSS-NLMS algorithm outperforms the FSS-NLMS algorithm.
Adaptive algorithm
Algorithm design
Cite
Citations (14)
Motivated by the real time applications of adaptive signal processing algorithms a new Approximate RLS algorithm is developed. It is shown that the computational complexity of this algorithm is comparable to that of the LMS algorithm. Convergence analysis for this algorithm is presented showing the unconditional convergence of the algorithm in the mean and the mean square sense for stationary data. It is shown that the rate of convergence of this algorithm is n/sup -1/. The convergence characteristics of this algorithm shows that the algorithm is much faster than the LMS algorithm but somewhat slower than the RLS algorithm. Modifications to this algorithm are suggested for use in nonstationary data environment. Simulation results for this algorithm are compared with those for the LMS and the RLS algorithms.< >
Algorithm design
Cite
Citations (3)
In this paper, a new fast convergence adaptive algorithm with variable step size is proposed for FIR adaptive filter. This new proposed algorithm is derived based on the quasi-Newton family. Simulation results are presented to compare the convergence of the proposed algorithm with least mean square (LMS) algorithm and RLS algorithm. It shows that the proposed new algorithm has comparable convergence speed to the other known adaptive algorithms
Adaptive algorithm
Algorithm design
Cite
Citations (2)
This paper discusses about a type of algorithms of variable step-size LMS adaptive filtering and presents another new non-linear functional relationship between the step-size and the error signal,which improves the algorithm performance.On the condition of the same convergence properties or the same excess MSE, the new algorithm has less excess MSE or has better convergence properties than the former algorithms.Computer simulation results confirm the theoretical analysis and show the algorithm is superior to the former algorithms in performance.
Adaptive algorithm
Cite
Citations (5)
Cite
Citations (0)
The conventional fixed step size LMS algorithm is difficult to result in fast convergence speed and low steady state error simultaneously.On the basis of the problem,an improved LMS adaptive algorithm is proposed in this paper,which combine variable step size algorithm and transformdomain algorithm,it can get good convergence and misadjustment property.The results of simulating manifest the performance of the new algorithm is better than others,especially in low SNR situation.
Adaptive algorithm
Cite
Citations (4)
LMS (Least Mean Square) algorithm is widely used due to its simple and stable performance. As is well known, there is an inherent conflict between the convergence rate and stead-state misadjustment, which can be overcome through the adjustment of size factor. The paper has analyzed some LMS algorithms that already existed and a new improved variable step-size LMS algorithm is presented. The computer simulation results are consistent with the theoretic analysis, ?which show that the algorithm not only has a faster convergence rate, but also has a smaller steady-state error.
Cite
Citations (0)
The block-based LMS algorithm (BLMS) is an efficient adaptive filtering algorithm aimed at increasing the convergence speed and reducing the computational complexity. The basic principle of the BLMS algorithm is that the filter coefficients remain unchanged during the processing of each data block, and are updated only once per block. The convergence properties of the unconstrained frequency-domain block LMS adaptive algorithm are analyzed. The learning characteristics of the unconstrained case are compared with the constrained case via computer simulation. It is shown that the unconstrained algorithm has a slower convergence rate and smaller stable range of step size than that of the constrained algorithm.< >
Algorithm design
Cite
Citations (5)