
AD6636
Biphase Filtering Option
The second special function that can be performed by the
second subblock of the output data router is called the biphase
filtering option. With this option, the AD6636 can be used to
process data from ADCs that run faster than the input clock
frequency by using two channels or two streams to form a
biphase filter.
Rev. 0 | Page 39 of 72
For example, a 300 MHz ADC can be used with a clock rate of
150 MHz driving the ADC. The ADC data can be decimated by
2 to produce even and odd data streams of data. The even
stream can be clocked into ADC Input Port A, and the odd
stream can be clocked into ADC Input Port B. These input ports
drive separate channels or separate groups of channels. The
filters of the RCF can be designed to place a 300 MHz sample
time difference (1/300 MHz = 3.3 ns) between the even and odd
path filters.
After the channel-filter coefficients have appropriate delay, a
complex addition of the odd and even sample channels can be
performed to create a single filter. This equivalent filter looks
like a single channel with a 300 MHz input rate, even though
the clock rate of the chip runs at only 150 MHz.
A biphase filter summation is implemented by the following
equation:
Output
= (
Ie
×
Ce
+
Io
×
Co
) +
j
(
Qe
×
Ce
+
Qo
×
Co
)
where:
Ie
×
Ce
,
Qe
×
Ce
are even in-phase and quadrature-phase
samples from one stream.
Io
×
Co
and
Qo
×
Co
are odd in-phase and quadrature-phase
samples from the other stream.
Ce
and
Co
are the even and odd coefficients, which differ by
1 high speed sample time (300 MHz in the previous example).
Users can program certain streams to be summed using the
biphase filtering option. This option can be programmed using
the same 3-bit complex control word in the Parallel Output
Control 2 register. The values for the 3-bit control word and
their corresponding settings are listed in Table 23.
AUTOMATIC GAIN CONTROL
The AD6636 is equipped with six independent automatic gain
control (AGC) loops that directly follow the second data router
and immediately precede the parallel output ports. Each AGC
circuit has 96 dB of range. It is important that the decimating
filters of the AD6636 preceding the AGC reject unwanted
signals, so that each AGC loop is operating only on the carrier
of interest, and carriers at other frequencies do not affect the
ranging of the loop.
The AGC compresses the 24-bit complex output from the
second data router into a programmable word size of 4 to 8, 10,
12, or 16 bits. Because the small signals from the lower bits are
pushed in to higher bits by adding gain, the clipping of the
lower bits does not compromise the SNR of the signal of
interest.
The AGC maintains a constant mean power on the output
despite the level of the signal of interest, allowing operation in
environments where the dynamic range of the signal exceeds
the dynamic range of the output resolution. The output width of
the AGC is set by writing a 3-bit AGC word length word in the
AGC control register of the individual channel’s memory map.
The AGC can be bypassed, if needed, and, when bypassed, the
24-bit complex input word is still truncated to a 16-bit value
that is output through the parallel port output. The six AGCs
available on the AD6636 are programmable through the six
channel memory maps. AGCs corresponding to individual
channels can be bypassed by writing Logic 1 to AGC bypass bit
in the AGC control register.
Three sources of error can be introduced by the AGC function:
underflow, overflow, and modulation. Underflow is caused by
truncation of bits below the output range. Overflow is caused by
clipping errors when the output signal exceeds the output range.
Modulation error occurs when the output gain varies while
receiving data.
The desired signal level should be set based on the probability
density function of the signal, so that the errors due to under-
flow and overflow are balanced. The gain and damping values
of the loop filter should be set, so that the AGC is fast enough to
track long-term amplitude variations of the signal that might
cause excessive underflow or overflow, but slow enough to avoid
excessive loss of amplitude information due to the modulation
of the signal.
AGC Loop
The AGC loop is implemented using a log-linear architecture. It
contains four basic operations: power calculation, error calcula-
tion, loop filtering, and gain multiplication.
The AGC can be configured to operate in either desired signal
level mode or desired clipping level mode. The mode is set by
the AGC clipping error bit of the AGC control register. The
AGC adjusts the gain of the incoming data according to how far
it is from a given desired signal level or desired clipping level,
depending on the selected mode of operation.
Two datapaths to the AGC loop are provided: one before the
clipping circuitry and one after the clipping circuitry, as shown
in Figure 39. For the desired signal level mode, only the I/Q
path from before the clipping is used. For the desired clipping
level mode, the difference of the I/Q signals from before and
after the clipping circuitry is used.