Abstract

In the framework of error correction in space telecommand (TC) links, the Consultative Committee for Space Data Systems (CCSDS) currently recommends short block-length BCH and binary low-density parity-check (LDPC) codes. Other alternatives have been discarded due to their high decoding complexity, such as non-binary LDPC (NB-LDPC) codes. NB-LDPC codes perform better than their binary counterparts over AWGN and jamming channels, being great candidates for space communications. We show the feasibility of NB-LDPC coding for space TC applications by proposing a highly efficient decoding architecture. The proposed decoder is implemented for a (128,64) NB-LDPC code over GF(16) and the design is particularized for a space-certified Virtex-5QV FPGA. The results prove that NB-LDPC coding is an alternative that outperforms the standardized binary LDPC, with a coding gain of 0.7 dB at a reasonable implementation cost. Given that the maximum rate for TC recommended by the CCSDS is 2 Mbps, the proposed architecture achieves a throughput of 2.03 Mbps using only 9615 LUTs and 5637 FFs (no dedicated memories are used). In addition, this architecture is suitable for any regular (2,4) NB-LDPC (128,64) code over GF(16) independently of the H matrix, allowing flexibility in the choice of the code. This brief places NB-LDPC codes as the excellent candidates for future versions of the telecommand uplink standard.

Highlights

  • S HORT block-length error correcting codes have been largely discussed and developed during the last years regarding the transmission of telecommand (TC) messages in the uplink channel in space communications

  • This work presents an efficient architecture which proves the feasibility of NB-low-density parity-check (LDPC) coding for error correction in space telecommand communications

  • Despite non-binary LDPC (NB-LDPC) codes not being included in the Consultative Committee for Space Data Systems (CCSDS) recommendation for TC due to their high decoding complexity, the proposed decoder exceeds the maximum throughput of 2 Mbps recommended by the CCSDS for TC links while using only 9615 LUTs and 5637 FFs on a Virtex-5QV FPGA

Read more

Summary

INTRODUCTION

S HORT block-length error correcting codes have been largely discussed and developed during the last years regarding the transmission of telecommand (TC) messages in the uplink channel in space communications. LDPC under most reliable basis (MRB) decoding [2] and non-binary LDPC (NB-LDPC) codes were investigated by CCSDS These were discarded due to hardware complexity limitations, which put a significant constraint on the choice of the channel code. MRBdecoded binary LDPC schemes achieve even higher coding gain, the implementation cost turns out to be prohibitive for the required throughput and target device, as detailed in the results section. To the best of the authors’ knowledge, there are no comparable implementations targeting short block-length NB-LDPC codes for space applications at a reasonable cost. The rest of this brief is organized as follows.

NB-LDPC DECODING ALGORITHM
PROPOSED DECODER ARCHITECTURE
A B Mode Selection
Parity Check Module
CNUs 16 CNUs
Overall Architecture of the NB-LDPC Decoder
Findings
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call