Probabilistic Signal Shaping With FEC for Higher Modulation Orders
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems face limitations in using high modulation orders due to the constraints imposed by the relationship between modulation order and FEC-code rate, which restricts the size of pulse-amplitude-modulation (PAM) constellations and affects the overall shaping gain.
Innovation Solution
The system employs probabilistic signal shaping and FEC coding by generating constellation symbol amplitudes using a shaping encoder, where some symbols carry parity-bit values and others carry unshaped information bits, allowing for increased modulation orders without a penalizing increase in FEC-code rate, specifically by using a low-density parity-check (LDPC) code.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If high modulation orders are used to increase transmission efficiency, then productivity is improved, but device complexity increases due to constraints on FEC-code rate and PAM constellation size
Solution Approach 1:
The patent segments the binary label bits of constellation symbols into two groups: information bits and parity bits. The parity bits are generated by an FEC encoder operating on a subset of the binary labels. This segmentation allows the system to use high modulation orders (m≥8) without proportionally increasing the FEC-code rate, as the parity bits provide error correction capability while the information bits carry the actual data, thus resolving the contradiction between transmission efficiency and device complexity
Solution Approach 2:
The patent implements nested coding by combining probabilistic signal shaping with FEC coding in a hierarchical structure. The shaping encoder generates amplitude probabilities based on binary labels, while the FEC encoder operates on a subset of these labels to generate parity bits. This nested arrangement allows both shaping gain and error correction to work together efficiently, enabling high modulation orders without the penalizing increase in FEC-code rate that would normally be required
2Productivity
If high modulation orders are used to increase transmission efficiency, then productivity is improved, but reliability decreases due to increased error susceptibility
Solution Approach 1:
The patent applies preliminary error correction by generating parity bits through FEC encoding before the constellation symbols are transmitted. The FEC encoder processes a subset of the binary labels of the shaped amplitudes to create redundancy bits that will be used for error detection and correction at the receiver. This preliminary action ensures that even when high modulation orders increase error susceptibility, the reliability is maintained through pre-prepared error correction capability
Solution Approach 2:
The system implements feedback through the FEC decoding process at the receiver end. The FEC decoder uses the received parity bits and information bits to detect and correct errors in the transmitted data. This feedback mechanism allows the system to maintain low bit-error rates even when using high modulation orders for improved transmission efficiency, as errors are identified and corrected based on the redundancy information
Data Source
AI summary
A communication system in which probabilistic signal shaping and FEC coding are jointly used in an efficient manner that enables, e.g., the use of relatively high modulation orders without a penalizing increase of the FEC-code rate. In an example embodiment, the amplitudes of transmitted constellation symbols are generated using a shaping encoder. Some transmitted constellation symbols carry, as sign bits, the parity-bit values generated by an FEC encoder configured to operate on (i) a subset of bits from the binary labels of the shaped amplitudes and (ii) an additional set of unshaped information bits. Some other transmitted constellation symbols carry, as sign bits, the bit values of the additional set of unshaped information bits. In some embodiments the used FEC code can be a low-density parity-check code. Some embodiments can be used in communication systems relying on discrete multi-tone modulation, such as the systems providing DSL access over subscriber lines.