Digital Signal Processing Reference
In-Depth Information
Table 1.1
EGPRS Modulation and Coding
Schemes and Peak Data Rates
Maximum Rate
per Slot (kb/s)
Scheme
Modulation
Code Rate
MCS-1
GMSK
8.8
0.53
MCS-2
GMSK
11.2
0.66
MCS-3
GMSK
14.8
0.80
MCS-4
GMSK
17.6
1.00
MCS-5
8-PSK
22.4
0.37
MCS-7
8-PSK
44.8
0.76
MCS-6
8-PSK
29.6
0.49
MCS-8
8-PSK
54.5
0.92
MCS-9
8-PSK
59.2
1.00
different SNR regions can be obtained through the use of different modulations depend-
ing on the SNR.
Link adaptation using adaptive coding and modulation is deployed in some of the
new-generation wireless communications systems. For example, EGPRS, which is the
evolution of the second-generation GSM, employs two different modulation options
(GMSK and 8-PSK) along with different coding rates, resulting in nine different modu-
lation/coding options, as shown in Table 1.1 [5, 43]. In addition, EGPRS introduces the
use of a type II hybrid ARQ system, commonly known within the specification as incre-
mental redundancy. In link adaptation, the link quality is measured regularly and the
most appropriate modulation and coding scheme is assigned for the next transmission
interval. On the other hand, in the incremental redundancy scheme, information is first
sent with low coding power (high coding rate). This results in a high bit rate if decoding
is successful with this rate. However, if decoding fails with such a high rate, additional
coded bits (redundancy) should be sent so that the transmitted bits can be decoded suc-
cessfully. However, sending extra coded bits incrementally reduces the resulting bit rate
and introduces undesired extra delay. Therefore, the initial code rate and modulation for
the incremental redundancy scheme should be based on measurements of the link qual-
ity, instead of starting with any arbitrary rate [5]. As a result, by combining incremental
redundancy with adaptive initial code rate, lower delays with lower memory require-
ments, and high data rates can be achieved. The different initial code rates are obtained
by puncturing a different number of bits from a common convolution code (rate 1/3).
Incremental redundancy operation is enabled by puncturing a different set of bits each
time a block is retransmitted, whereby the code rate is gradually decreased toward one-
third for every new transmission of the block.
Recent studies introduce new modulation and coding options together with other
capacity enhancement techniques to further increase the data rate and throughput of
EGPRS [64, 65]. Higher-order modulations like 16-QAM and 64-QAM are being pro-
posed along with some more coding options to optimize the performance.*
* 16-QAM and64-QAM stand for 16-level and 64-level quadrature amplitude modulation,
respectively.
Search WWH ::
Custom Search