Convolutional Decoding for Flexible Non-Integer Code Rates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems using convolutional coding for error correction in M-value FSK systems face restrictions in code rate, limiting the flexibility in setting transmission speed due to the requirement for a trellis diagram with specific branch bit lengths, which is not feasible for non-natural reciprocal code rates.
Innovation Solution
A wireless communication system that includes a transmitter with a coder for convolutional coding, a bit eraser to generate a second bit sequence with a higher code rate, and a modulator, and a receiver with a demodulator, likelihood extender, and decoder that constructs a trellis diagram using extended bit sequences and reliabilities to perform decoding, allowing flexible transmission speed settings even with non-natural reciprocal code rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a trellis diagram is constructed with the number of coded bits per branch equal to the common multiple C of log2 M and 1/R, then decoding can be performed using likelihood per symbol, but the code rate R is restricted and transmission speed cannot be set flexibly
Solution Approach 1:
The invention segments the bit sequence processing into two stages: first, convolutional coding generates an initial bit sequence; second, a bit eraser selectively removes bits to create a second bit sequence with a higher effective code rate. This segmentation allows the system to use a simplified trellis diagram with fewer states while maintaining decoding accuracy, thereby enabling flexible transmission speed settings without being constrained by the common multiple requirement of traditional approaches.
Solution Approach 2:
The invention changes the code rate parameter dynamically by using a bit eraser to adjust the number of bits transmitted. By erasing specific bits from the convolutionally coded sequence, the system can adapt the effective code rate to match different transmission speed requirements while still using a fixed, simplified trellis diagram structure for decoding, thus resolving the contradiction between decoding accuracy and transmission flexibility.
2Ease of manufacture
If the reciprocal 1/R of the code rate R is not a natural number, then the decoder cannot be used according to the conventional method, but this limits the code rate options
Solution Approach 1:
The invention performs preliminary bit erasure before modulation and transmission. By pre-processing the convolutionally coded bit sequence to remove specific bits, the system prepares the data in a format that can be decoded using a simplified trellis diagram with a fixed number of states. This preliminary action eliminates the need to handle complex non-natural reciprocal code rates during decoding, maintaining implementation simplicity while expanding code rate options.
3Productivity
If bits are erased to increase the effective code rate, then transmission speed flexibility improves, but the system complexity increases due to additional processing
Solution Approach 1:
The invention introduces dynamic bit erasure that can be configured based on transmission requirements. The bit eraser can be programmed to remove different numbers and patterns of bits depending on the desired transmission speed and channel conditions. This dynamic approach allows the system to adjust its effective code rate and transmission speed flexibly without requiring multiple fixed hardware configurations, thereby managing complexity while maintaining productivity.
Data Source
AI summary
A system includes: a transmitter including: a coding unit generating a first bit sequence by convolutional coding on information bits based on a code rate; a bit erasing unit generating a second bit sequence by erasing one or more bits from the first bit sequence for every predetermined first number of bits; and a modulation unit generating a symbol by modulation using the second bit sequence; and a receiver including: a demodulation unit calculating first reliabilities that can be derived from the symbol; a likelihood extension unit generating extended bit sequences each composed of bits for the first number of bits, and generating a plurality of second reliabilities by assigning first reliabilities duplicated, as the reliabilities of the extended bit sequences; and a decoding unit creating a trellis diagram using the code rate and the extended bit sequences, and assigning the second reliabilities to branches of the trellis diagram.