FORWARD ERROR CORRECTION
The tight requirement of finishing the overall detection within
3 ms poses a significant challenge for executing FEC decoding
within the cloud-platform due to its high complexity. Usually,
FEC decoders are implemented in specialized hardware, such as
application-specific integrated circuit (ASIC) designs or field-programmable
gate array (FPGA) implementations [22]. However,
the introduction of many-core architectures opens new perspectives
for massively parallel implementations. To meet stringent
requirements on data rates, cloud-based FEC decoders will need
to fully exploit the available parallelism of a cloud-computing
platform. In this context, low-density parity check (LDPC) [23]
and turbo codes [24] are two promising candidates because both
allow for accommodating various degrees of parallelization.