PDSP16510AA0AC Mitel Networks Corporation, PDSP16510AA0AC Datasheet - Page 17

• Manufacturer Part Number: PDSP16510AA0AC
• Description: Stand Alone FFT Processor
• Manufacturer: Mitel Networks Corporation

window it is six bins.
PDSP16510. These are constructed on the fly as needed, and
take the general form:
For Hamming, A = 0.54, B = 0.46, C = 0
For Blackman-Harris, A = 0.42323, B = 0.49755,C=0.07922
size options, except the 16 x 16 complex variant. When the
latter is specified the rectangular window option MUST be
selected, or the device will be configured in an internal test
mode.
externally. This can be conveniently achieved with either a
PDSP16112 or a PDSP16116, both of which are complex
multipliers but with different accuracies. Fig. 11 shows how
either one can be configured to perform two separate multipli-
cations with one input common to both. This arrangement is
necessary to perform the window function on complex inputs.
PDSP16510, and other commonly used windows, are illus-
Table 7. Window Performance ( from The use of Windows for Harmonic Analysis. F J Harris. Proc IEEE Vol 66. Jan 1978 )
WINDOW
The latter two windows are actually supported by the
These windows can be applied to any of the transform
Important features of the windows generated by
A - Bcosx + Ccos2x where x = (2 n)/N, n = 0 to N-1
If other operators are required these must be applied
DATA
PROM
REAL
XR
YR
PDSP16116
MULTIPLIER
COMPLEX
Window
Operator
Rectangular
Hamming
Dolph-Chebyshev
[C = 3.5]
Kaiser-Bessel
[C = 3]
Blackman
Blackman-Harris
[3 term]
ZERO
IMAG'
DATA
XI
YI
Fig. 11. External Window Generator
CLK
PARAMETERS
COUNTER
AUX
D
POWER
ON RESET
CLR
PDSP16510
Highest
Side Lobe
-13
-43
-70
-69
-58
-67
R
SAMPLE
CLOCK
SYSTEM
CLOCK
FIRST
SAMPLE
Mid-Point
Loss dB
3.92
1.78
1.25
1.02
1.1
1.13
trated in Table 7. The results are obtained from the reference
quoted, which should be consulted for a full mathematical
treatment. The significance of each parameter is outlined
below :
Highest Side Lobe Level
are only 13dB down from the mainlobe. These severely limit
the dynamic range. The object of the window is to improve this
situation with better side load attenuation.
Mid-Point Loss
an additional processing loss for a tone of frequency mid-way
between two bins. This is defined as the ratio of the coherent
gains of two tones, one at the mid-point and one at the sample
point. It is expressed in dB in Table 8.
Overall loss
can be obtained by adding the mid-point loss to the reciprocal
of the equivalent noise power bandwidth in dB. It is a measure
of the ability of the window to detect single tones in broadband
noise. The variance between windows is less than 1dB.
6.0dB Bandwidth
of the window to resolve two tones and should be as close to
unity as possible. As the highest sidelobe level is reduced, this
parameter tends to get worse, and a compromise must be
used when choosing a window.
Overlap Correlation
successive transforms are averaged to reduce the variance of
the measurements. If, however, a windowed FFT is applied to
non overlapping partitions of the sequence, data near the
boundaries will be ignored since the window exhibits small
values at those points. To avoid this loss partitions are usually
overlapped by 50% or 75%, which might, at first sight, remove
the need to average successive transforms. If non-windowed
Overall
Loss dB
3.92
3.1
3.35
3.55
3.47
3.45
The inherent rectangular window has sidelobes which
In line with the filter concept it is possible to conceive of
An overall figure for the reduction in signal to noise ratio
This figure, expressed in bin widths, represents the ability
In many practical systems the squared magnitudes of
6dB
Bandwidth
1.21
1.81
2.17
2.39
2.35
1.81
Overlap Correlation
75%
75
70.7
60.2
53.9
56.7
57.2
50%
50
23.5
11.9
PDSP16510
7.4
9
9.6
17