Biomedical Engineering Reference
In-Depth Information
overlooks the precision in pulse generation, hence the achievable controllability in
the pulse spectrum over the hardware complexity. The main drawback of this
approach is that the DAC has to operate at very high sampling rates (in the order of
10 Gsps) in order to generate the UWB pulses. This is not only challenging for the
implementation of the DAC, but also the input data stream has to operate at very
high rates; hence it demands the use of high speed logic circuits. In general, the
waveform synthesis UWB pulse generation method is suitable for on-chip
implementations using advanced technologies such as CMOS due to the require-
ment of the high precision in circuit implementation.
4.3 UWB Receiver Design Techniques
Due to the short pulse width and low power of the signal, front-end circuitry for
the UWB receiver is complex in design and has high power consumption. An
Analog to Digital Converter (ADC) in a UWB receiver requires a large input
bandwidth and a high sampling rate. For example ADC12D1800 [
17
] by National
Semiconductors has 3.5 Giga samples per second sampling rate and an input
bandwidth of 1.75 GHz, but it consumes 4.4 W of power which is not suitable for
battery powered UWB sensor design. Although the ADC has been brought close to
the antenna with the evolution of the front-end circuitry for narrow band systems,
it is not considered as a suitable technique for UWB systems. The fully digital
implementations of the UWB receivers require precise synchronization of nano-
second scale narrow UWB pulses and resolving numerous multipath components
of the received UWB signals [
18
].
UWB receivers are of two types: non-coherent receivers and coherent receivers.
These two receiver architectures are discussed in following sections.
4.3.1 Non-Coherent UWB Receivers
Non-coherent UWB receivers can be further sub-divided into two categories:
Energy Detection (ED) receivers and Autocorrelation (AcR) receivers. ED UWB
receiver architectures are discussed in [
19
,
20
]. In this receiver type, a squaring
device is used to correlate the received UWB signal with itself. This can be
achieved by operating a MOSFET in the saturation region. Block diagram of the
receiver described in [
20
] is shown in Fig.
4.5
. The ED UWB receivers do not
require channel estimation; hence hardware complexity is greatly reduced. This
leads to superior performance in terms of power consumption. However the Sig-
nal-to-Noise Ratio (SNR) of this type of receivers is inferior to other types of
UWB receivers mainly due to use of the noisy received signal as the template
signal. Also, the receiver performance degrades rapidly in an environment with a
large number of interferers.
