Digital Gain Control And Interface Gain (Slicer) - Analog Devices ADRV9005 Reference Manual

Reference Manual
RECEIVER GAIN CONTROL
Table 90. A List of Rx Gain Control APIs
Rx Gain API Function Name
adi_adrv9001_Rx_GainControl_Mode_Set
adi_adrv9001_Rx_GainControl_Mode_Get
adi_adrv9001_Rx_Gain_Get
adi_adrv9001_Rx_Gain_Set
adi_adrv9001_Rx_GainTable_Write
adi_adrv9001_Rx_GainTable_Read
adi_adrv9001_Rx_DecimatedPower_Get
adi_adrv9001_Rx_GainControl_Configure
adi_adrv9001_Rx_GainControl_Inspect
adi_adrv9001_Rx_GainControl_MinMaxGainIndex_Set
adi_adrv9001_Rx_GainControl_MinMaxGainIndex_Get
adi_adrv9001_Rx_GainControl_Reset
adi_adrv9001_Rx_GainControl_PinMode_Configure
adi_adrv9001_Rx_GainControl_PinMode_Inspect

DIGITAL GAIN CONTROL AND INTERFACE GAIN (SLICER)

The digital gain control has two major purposes: gain correction to correct the small step size inaccuracy in the analog front-end attenuation,
and gain compensation to compensate for the entire analog front-end attenuation. In the gain compensation mode, for example, if 5 dB analog
attenuation is applied at the front end of the device, then 5 dB of digital gain is applied. This ensures the digital data is representative of the
RMS power of the signal at the receiver input port (plus the nominal receiver analog gain) so that any internal front-end attenuation changes
in the device to prevent ADC overloading are transparent to the baseband processor. In this way, the device's AGC is used to react quickly to
incoming blockers without the baseband processor needing to track the current gain index for the level of the received signal at the input to the
device for signal strength measurements.
The receiver gain table controls the digital gain block, as mentioned earlier. Note that different digital gain is applied when configured in the gain
correction or gain compensation mode. The receiver gain table has a unique front-end attenuator setting with a corresponding amount of digital
gain, programmed at each index of the table, as shown in
For the gain compensation mode, it is used in either the AGC or MGC mode. The digital gain compensates for both the internal analog
attenuator and an external gain component (such as a DSA or LNA). After the digital gain compensation, the signal power should only depend
on the input signal power.
Around the end of the receiver datapath, the receiver interface gain is further applied using a "Slicer" block for two major purposes: to avoid
digital saturation due to the bit-width limitation of the data port in the gain compensation mode, and to ensure the overall SNR is limited only
by analog noise and is unaffected by quantization noise. When the gain compensation mode is used, any analog attenuation is compensated
by a corresponding digital gain, such that the sum of the analog and digital gain is always equal to the nominal receiver analog gain of 20 dB.
At the ADC input, the full scale input signal is approximately 8.6 dBm. This value translates to 0 dBFS in the digital datapath for either the I or
Q channel. As an example, assuming a 5 dBm signal is applied at the receiver input port, at the receiver output, the signal power is 5 + 20 =
25 dBm or 25 − 8.6 = 16.4 dBFS. This causes the clipping at the 16-bit output signal. Therefore, the interface gain (less than 0 in this case) is
applied to attenuate the signal to avoid clipping. On the other hand, for a very low signal level, at the receiver input, within the RF bandwidth
of interest, it must be assured that the analog noise dominates the quantization noise. In the receiver datapath, there are different modes of
receiver data interface , which are classified into two major categories, one using a final 15-bit or 16-bit quantizer to round the 22-bit receiver
data to 15-bit or 16-bit, and the other passing the 22-bit receiver data entirely without using any quantizer. In the first mode, the quantizer
becomes the dominant noise source as a result of the final interface quantization. This quantization noise, as a result of the final 15-bit or 16-bit
quantizer, is spread over a bandwidth equivalent to its output sampling frequency. For NB applications where the output sampling frequency
is low, the total quantization noise per Hz is larger than the analog noise per Hz. By applying interface gain (greater than 0 in this case), prior
to the final quantizer, the signal level and analog noise level are both increased. Therefore, the analog noise dominates over the quantization
noise so that SNR is dominated by analog front-end noise in the RF bandwidth of interest. For WB applications, as the sampling frequency is
higher, the total quantization noise becomes much smaller. In such a case, the analog noise is way above the quantization noise. Therefore,
interface gain is not required. In the second mode, it passes the full 22-bit I/Q data from the receiver data path to the interface without rounding.
Therefore, it lowers the quantization noise without the need for additional interface gain. It uses 32 bit I data and 32 bit Q data on the interface
for CMOS 1-lane (64-bit) and LVDS 2-lane (32-bit I data and 32-bit Q data). Besides including the 22-bit I/Q data, the 32-bit data also includes
analog.com
Description
Configures the Rx gain control mode for a specific channel.
Retrieves the currently configured Rx gain control mode.
Reads the Rx gain index for the requested Rx channel.
Sets the current AGC gain index for the requested Rx channel.
Programs the gain table settings for Rx channels.
Reads the gain table entries for requested Rx channels.
Gets the decimated power for the specified channel.
Sets up the device Rx Gain Control for a specified channel
Inspects the device Rx gain control for a specified channel.
Sets the min/max gain indexes for gain control operation for the specified channel.
Gets the min/max gain indexes for gain control for the specified channel.
Resets all state machines within the gain control block.
Configures gain control for the MGC PIN mode.
Inspects gain control configurations for the MGC PIN mode.
Table 71.
ADRV9001
Rev. A | 202 of 377