JAJSEY1 April 2019 ADC12DJ5200RF
ADVANCE INFORMATION for pre-production products; subject to change without notice.
The decimation filters are arranged to provide a programmable overall decimation of 4 or 8. All decimation filters operate on complex data (from the complex digital mixer) and the outputs have a resolution of 15 bits. The decimation filters are implemented as linear phase finite impulse response (FIR) filters. Table 11 lists the effective output sample rates, available signal bandwidths, output formats, and stop-band attenuation for each decimation mode.
|DECIMATION SETTING||ƒ(DEVCLK)||OUTPUT FORMAT|
|OUTPUT RATE (MSPS)||MAX ALIAS PROTECTED SIGNAL BANDWIDTH (MHz)||STOP-BAND ATTENUATION||PASS-BAND RIPPLE|
|No decimation (DDC bypass)||ƒ(DEVCLK)||ƒ(DEVCLK) / 2||—||< ±0.001 dB||Real signal,
|Decimate-by-4||ƒ(DEVCLK) / 4||0.8 × ƒ(DEVCLK) / 4||> 90 dB||< ±0.001 dB||Complex signal,
|Decimate-by-8||ƒ(DEVCLK) / 8||0.8 × ƒ(DEVCLK) / 8||> 90 dB||< ±0.001 dB||Complex signal,
Figure 10 to Figure 13 provide the composite decimation filter responses. The black portion of the trace shows the pass-band region, or alias-protected region, of the response. The red portion of the trace shows the transition region of the response as well as any frequency regions that will alias into the transition region. The transition region is not alias protected and therefore desired signals should only be placed in the pass-band region of the filter response. The blue portion of the trace shows the frequency regions that will alias into the pass-band after decimation and therefore define the stop-band region of the frequency response. The stop-band attenuation is defined to sufficient filter any undesired images or signals to prevent them from aliasing into the desired pass-band. Use analog filtering before the analog inputs (INA± or INB±) for additional attenuation of signals that fall within this band or to sufficiently reduce signals at the ADC inputs that may produce harmonics, interleaving spurs or other undesired spurious signals that will alias into the desired signal band (before the complex mixing and decimation operations).
For maximum efficiency, a group of high-speed filter blocks are implemented with specific blocks used for each decimation setting to achieve the composite responses illustrated in Figure 10 to Figure 13. Table 12 describes the combination of filter blocks used for each decimation setting and Table 13 lists the coefficient details and decimation factor of each filter block. The coefficients are symmetric with the center tap indicated by bold text.
|DECIMATION SETTING||FILTER BLOCKS USED (Listed in Order of Operation)|
|8||CS20, CS40, CS80|
|FILTER COEFFICIENT SET (Decimation Factor of Filter, Scale factor)|
|CS20 (2, 2-14)||CS40 (2, 2-17)||CS80 (2, 2-19)|