| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 857.22 KB | JPEG | |||
| 2.65 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
The design of future broadband wireless systems presents a big challenge,
since these systems should be able to cope with severely time-dispersive
channels and are expected to provide a wide range of services (which may
involve data rates of several tens of Mbit/s) and to have high spectral and
power efficiencies. The use of equalization techniques to deal with the
channel time-dispersion effects associated to the multipath signal propagation
between transmitter and receiver becomes inevitable to compensate the
inherent signal distortion levels and ensure good performance. However,
implementation complexity and power consumption cannot be too high, especially
at the mobile terminals (MT), since low-cost and relatively long live
batteries are desirable at these terminals. The optimum receiver structure for
time-dispersive channels corresponds to the well-known Viterbi equalizer [1],
whose complexity grows exponentially with the channel impulse response
length, making it recommendable only for channels whose impulse response
spans over just a few symbols.
Time-domain equalization techniques are traditionally employed to mitigate
channel frequency selectivity effects, leading to a much lower implementation
complexity than Viterbi equalizers. Nonlinear equalizers such as
decisions feedback equalizers (DFE) [2] can significantly outperform linear
equalizers and have a good complexity/performance tradeoff. However, for
conventional single carrier (SC) modulations the signal processing complexity
(number of arithmetic operations per data symbol) required to mitigate
the strong intersymbol interference (ISI) levels inherent to digital transmission
over severely time-dispersive channels rapidly becomes prohibitive,
especially when conventional time-domain equalization is employed at the
receiver side.
Description
Keywords
Citation
Publisher
River Publishers