Digital Signal Processing Reference
In-Depth Information
In voice-dominated cellular systems, the objective of the power control was mainly
to maintain the minimal (target) link quality at a constant level for individual users.
The data rates for all users are constant in this case, and each user experiences roughly
the same quality of service. While this was appropriate for voice, recently, with the
increased demand for multimedia services and high-speed data access, different objec-
tives and cost functions to optimize the use of power resources have been developed. In
mixed-traffic environments, the cost function for each service will be different, leading
to different power allocation strategies [62]. Use of constant power along with variable
coding, modulation, and spreading that adapts the data rate to the channel variations
is one objective that some new-generation wireless systems have been adopting (rate
control or rate adaptation) [43]. Also, water-filling types of power assignments, which
assign more power to the users that have favorable channels, are being studied exten-
sively [63].
In adaptive power control mechanisms, estimation of the link quality parameters is
the key factor. Typical parameters used for adaptation include SIR, FER, and RSS. Dop-
pler spread estimate can also be used to adjust the adaptation rate. Depending on the
adaptation rate, power control can be classified as
fast power control
and
slow power con-
trol
. Fast power control compensates the changes in power level due to Rayleigh fading
(small-scale fading), while slow power control is used for lognormal fading (shadowing)
and path loss. The parameters that are used for them can also be different. For example,
for fast power control, instantaneous SIR, SNR, SINR, and RSS can be more suitable
than FER and BER, which might better suit slow power control. As mentioned in the
parameter estimation section, parameter selection depends on the delay, complexity,
and accuracy requirements. The estimation errors and delays, between measurements
and adaptation of power, limit the efficient application of power control schemes. There-
fore, more accurate and practical algorithms that estimate and predict the parameters to
be used in adaptation are needed.
1.4.2.2 Adaptive Modulation and Channel Coding
Given the high price of spectrum and its scarcity, it is in the interest of operators to
continue evolving their networks toward higher capacity and quality. Adaptive modula-
tion and coding provide a framework to adjust modulation level and FEC coding rate
depending on the link quality. Higher-order modulations (HOMs) allow more bits to
be transmitted for a given symbol rate. On the other hand, HOM is less power efficient,
requiring higher energy per bit for a given BER. Therefore, HOMs should be used only
when the link quality is high, as they are less robust to channel impairments. Similarly,
strong FEC and interleaving provide robustness against channel impairments at the
expense of lower data rate and spectral efficiency, suggesting adaptation of coding rate
based on the link quality.
Figure 1.10
illustrates the capacity gain that can be achieved
by employing adaptive modulation only. First, the BER performances of different modu-
lations as a function of SNR are given in Figure 1.10(a). As can be seen, a desired BER
can be achieved with low-order modulations for lower SNRs. Higher-order modulations
need better link quality (higher SNRs) in order to obtain the same BER performance.
Figure 1.10(b) shows the spectral efficiencies of different uncoded modulations, where
an arbitrary packet size of 200 bits is used. Notice that the optimal spectral efficiency for
Search WWH ::
Custom Search