AS 2670.1 Filter Coeffs
Budget: $30 – $250 AUD
I need a precise set of digital filter coefficients that replicate the W.k and W.d weightings defined in Figure 2 of Australian Standard AS 2670.1-2001. These coefficients must be provided in Auto-Regressive and Moving-Average form (one MA term per pole) so they can be entered directly into SoMat Ease v3 for post-processing of vibration data.
I’m only somewhat familiar with the standard, but I do have the tabulated weightings from Table 3 and will supply them as soon as we start. What I’m missing is the mathematical translation of those weightings into stable, causal filters whose magnitude responses stay within the tolerance bands of the standard across the whole frequency range.
You may derive the coefficients in MATLAB, Python/SciPy, or any other tool you prefer, as long as the final numbers work flawlessly in SoMat Ease.
Deliverables:
• A set of AR and corresponding MA coefficients for both W.k and W.d weightings, formatted for direct entry into SoMat Ease.
• A brief method note (frequency response plots and verification figures) showing the filters match the standard curves within the specified tolerances.
If you’ve already tackled AS 2670.1 filters before, let me know—accuracy is everything on this job.
I’m only somewhat familiar with the standard, but I do have the tabulated weightings from Table 3 and will supply them as soon as we start. What I’m missing is the mathematical translation of those weightings into stable, causal filters whose magnitude responses stay within the tolerance bands of the standard across the whole frequency range.
You may derive the coefficients in MATLAB, Python/SciPy, or any other tool you prefer, as long as the final numbers work flawlessly in SoMat Ease.
Deliverables:
• A set of AR and corresponding MA coefficients for both W.k and W.d weightings, formatted for direct entry into SoMat Ease.
• A brief method note (frequency response plots and verification figures) showing the filters match the standard curves within the specified tolerances.
If you’ve already tackled AS 2670.1 filters before, let me know—accuracy is everything on this job.