Name
Abstract | ||
|---|---|---|
Most current adaptive filters fix the filter order at some compromise value resulting in too short and too long filters with issues like undermodelling and adaptation noise in time varying scenarios. The tap length learning algorithm dynamically adapt the filter order to the optimum value makes the variable order adaptive filter more efficient including smaller computational complexity, higher output SNR and lower power consumption. The optimum order best balance the complexity and steady state performance of the adaptive filter. Choice of parameter, noise level and convergence issues affect the performance up to a great extent in the existing dynamic order estimation algorithm. In this paper a variable step LMS (VLMS) based pseudo-fractional optimum order estimation algorithm has been proposed that improves the overall performance of the adaptive filter searching the optimum order dynamically with fast convergence. Simulations and results are provided to observe the analysis of the proposed algorithm. |
Year | DOI | Venue |
|---|---|---|
2011 | 10.1145/1947940.1947966 | ICCCS |
Keywords | Field | DocType |
pseudo-fractional optimum order estimation,filter order,optimum order dynamically,adaptive filter,current adaptive filter,existing dynamic order estimation,optimum order,variable order,long filter,optimum value,computational complexity,steady state | Convergence (routing),Control theory,Computer science,Noise level,Algorithm,Kernel adaptive filter,Adaptive filter,Steady state,Power consumption,Filter design,Computational complexity theory | Conference |
Citations | PageRank | References |
4 | 0.53 | 10 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Asutosh Kar | 1 | 15 | 6.33 |
| Ravinder Nath | 2 | 192 | 23.43 |
| Alaka Barik | 3 | 4 | 0.53 |