Specifications


SLWS142J − JANUARY 2003 − REVISED AUGUST 2007
www.ti.com
24
ROUND
5
20 Data
Plus
8 Overflow
20
19
Data
In
28
7 Integer
Plus
12 Fractional
MAGNITUDE
agc_rnd
DELAY
DELAY
Valid Valid
20
Data Out
(Upper 20 − agc_rnd bits are valid,
lower bits are cleared)
8
COMPARE
8
agc_thresh
UNDER/OVER
DETECT
4
agc_sat_cnt
4
agc_zero_cnt
SHIFT SELECT
4
agc_Dblw
2 2
4
agc_Dadv
4
agc_Dzro
4
agc_Dsat
5
LIMIT
*agc_min < A(t) <
agc_max
16
agc_max
16
agc_min
MS16
7 Integer
Plus
9 Fractiona
l
ACCUMULATE
Sync
Sign Plus
7 Integer
Plus
16 Fractional
A(t+1)=A(t)+S 2
−(D+3)
G(t)
24
SHIFT
7 Integer
Plus
12 Fractional
19
S=±1, D=4-Bit Shift
Valid
agc_freeze
G(t)=Gain+A(t)
19
Gain
7 Integer
Plus
12 Fractional
19
A(t)=Gain Adjust
Under Limit
Over Limit
CLR
LS8
LS20
MS8
OVERFLOW
MS16
Figure 12. GC5016 AGC Circuit
The AGC portion of the circuit is used to change the adaptive gain so that the median magnitude of the output data
matches a target value. The magnitude of the gain-adjusted (manual + adaptive) output data is compared to a target
threshold. If the magnitude is greater than the threshold, the gain is decreased. If not, it is increased. The gain is
adjusted as:
G(t) = G + A(t)
A(t) = A(t) + G(t) x S x 2
−(D+3)
where G is the default, user supplied gain value, and A(t) is the time varying adjustment, where S=1 if the magnitude
is less than the threshold and is −1 if the magnitude exceeds the threshold, and where D sets the adjustment step
size. Note that the adjustment is a fraction of the current gain. This is designed to set the AGC noise level to a known
and acceptable level, while keeping the AGC convergence and tracking rate constant, independent of the gain level.
Because the adjustment is a fraction of the current gain, one can show that the AGC noise is an amplitude jitter in
the data output equal to ±(data output) x 2
−(D+3)
. This means that the AGC noise is always 6 x (D+3) dB below the
output signal’s power level. The AGC attack and decay rate is exponential with a time constant equal to 2
(D+1.75)
complex samples. This means the AGC covers to within 63% of the required gain change in one time constant and
to within 98% of the change in the four time constants.
If one assumes the data is random with a Gaussian distribution, which is valid for UMTS if more than 12 users with
different codes have been overlaid, then the relationship between the RMS level and the median is MEDIAN = 0.6745
x RMS.
Hence the threshold should be set to 0.6745 times the desired RMS level.