August 2, 2018
Editor’s Note: This content is contributed By Ambrose Finnerty, Xilinx DSP Technical Marketing Management
Maintaining transmission reliability in both Wireless and Cable data communications is a fundamental requirement for a quality solution. A key building block in these systems is a high-performance soft-decision forward error correction (SD-FEC) function required in both the transmit (encoder) and receive (decoder) paths.
<Figure 1: Typical Data Communication System>
With ever increasing data bandwidths, seen for example with 5G New Radio (5G NR) and Data Over Cable Service Interface Specification 3.1 (DOCSIS 3.1), these systems have considerable data throughput needs that the SD-FEC blocks must be able to process efficiently.
These SD-FEC functions have commonly been implemented in the programmable logic of high performance FPGAs. As the system requirements are pushed further to support multi-gigabit rates, performance, power, and cost have become key design factors. Soft implementation of these functions can be suboptimal compared to an integrated solution.
With this in mind, Xilinx has introduced the industry’s first integrated and configurable SD-FEC IP block enabling:
<Figure 2: Power Comparison of Integrated SD-FEC versus Soft LDPC Decoder>
As already pointed out, the ASIC-like SD-FEC block achieves greater throughput, lower latency, and lower power than a soft implementation. The ability to support Turbo decoding for Long-Term Evolution (LTE) and LTE-A applications make it a low-power solution for 4G and pre-5G systems.
With low-density parity check (LDPC) support for both decode and encode, it is possible to support applications like 5G baseband and backhaul platforms in the wireless market. In addition to the RF-ADCs/DACs found in the Zynq UltraScale+ RFSoC (ZU28DR) family, the SD-FEC provides an attractive solution for remote PHYs in the DOCSIS 3.1 standard.
To learn more on the specifics of Xilinx’s integrated SD-FEC in the Zynq UltraScale+ RFSoC and how you can leverage the power and throughput advantages, please take 10 minutes to read the following white paper outlining the block-level capabilities and benefits.