Digital Signal Processing Reference
In-Depth Information
Signal processing flow in a narrowband receiver architecture.
Passband signal processing in analog domain
IF/baseband processing in digital domain
frequency
frequency
D
Sparse spectral density of RF
inefficient for the digital domain
A
N
A
L
O
G
I
G
I
T
A
L
ADC
band-select
channel-select filter
DC
LO
Bandwidth compression in a pulse-based receiver.
RF-to-baseband very early in receive chain
Digital domain receives partially cleaned signal
frequency
frequency
D
In-band interference forms problem
for fully di
Weak signals in correlator are
vulnerable to noise injection.
A
N
A
L
O
G
I
G
I
T
A
L
matched-filter
correlator
∫
ADC
time-window
low-pass filter
DC
template pulse
Figure 5.5.
Analogously to the approach used in a narrowband system, the wide
spectral footprint of pulse-based radio must be compressed to a limited-
bandwidth baseband signal as soon as possible in the receive chain.
back-end, then the entire receive chain must support the large bandwidth of the
rf-signal. A significant amount of power can be saved by moving (part of) the
demodulation process to the front of the receive chain. In a pulse-based radio
system, the demodulation or detection of the short pulses can be seen as the
equivalent of downconversion in a carrier-based transmission (Figure 5.5). In
the downconversion process, the rf-signal is translated from a passband signal
- adjusted for transmission over the wireless medium - to a baseband signal. In
virtually all cases, downconversion is performed by the analog domain. Unlike
analog circuits, it is much more energy-efficient for a digital system to process
a baseband signal than to boldly deal with the rf-signal itself. For the same
reasons, a pulse-based radio receiver could employ some sort of bandwidth
compression technique to convert the rf-pulses into a baseband signal.
