Update default filter coefficients

README.SmurfProcessor.md

@@ -89,6 +89,13 @@ y(n) = gain / a(0) * [ b(0) * x(n) + b(1) * x(n -1) + ... + b(order) * x(n - ord
 
 This module can be disabled; the incoming frame will just pass through to the next block.
 
+The default coefficients were generated using this python code:
+
+```
+import scipy.signal as signal
+b, a = signal.butter(4, 2*63 / 4000.)
+```
+
 ### Downsampler
 
 Perform a downsampling of the data in the incoming frame, by letting pass only 1 data point each `Factor` number of point.

python/pysmurf/core/devices/_SmurfProcessor.py

@@ -157,11 +157,11 @@ def __init__(self, name, device, **kwargs):
             name='A',
             description='Filter a coefficients',
             mode='RW',
-            value= [ 1.0,
-                     -3.9999966334205683,
-                      5.9999899002673720,
-                     -3.9999899002730395,
-                      0.9999966334262355 ] + [0] * 11,
+            value= [  1.0,
+                     -3.74145562,
+                      5.25726624,
+                     -3.28776591,
+                      0.77203984 ] + [0] * 11,
             localSet=lambda value: self.device.setA(value),
             localGet=self.device.getA))
 
@@ -173,11 +173,11 @@ def __init__(self, name, device, **kwargs):
             name='B',
             description='Filter b coefficients',
             mode='RW',
-            value= [ 1.7218423734035440e-25,
-                     6.8873694936141760e-25,
-                     1.0331054240421264e-24,
-                     6.8873694936141760e-25,
-                     1.7218423734035440e-25 ] + [0] * 11,
+            value= [ 5.28396689e-06,
+                     2.11358676e-05,
+                     3.17038014e-05,
+                     2.11358676e-05,
+                     5.28396689e-06 ] + [0] * 11,
             localSet=lambda value: self.device.setB(value),
             localGet=self.device.getB))