Sparse ML Block Code Decoding Using Reduced TFCI Code Space
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless data communication systems, such as those in the 3GPP, face challenges in improving the error correction capability of block codes, particularly in decoding the transport format combination indicator (TFCI) without modifying the encoder, and existing decoders do not effectively utilize a priori knowledge of the code word space to enhance decoding performance.
Innovation Solution
A method using a Maximum Likelihood (ML) sparse decoder that iterates over known transport format combination indicators (TFCIs), correlates encoded TFCIs with soft outputs from a rake receiver, and compares correlation results to determine the most correlated hypothesis, thereby improving error correction capability without altering the encoder.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional decoders (e.g., Green Machine Decoder) search through all possible code words in the code space, then they can decode any potential code word, but the error correction capability is limited and power consumption is high due to exhaustive search
Solution Approach 1:
The patent extracts and utilizes a priori knowledge about the specific subset of code words that are actually transmitted in the communication system. By taking out only the relevant code words from the complete code space and forming a reduced code space, the decoder avoids searching through all possible code words. This extraction of necessary information reduces the search space significantly, lowering power consumption while maintaining or improving error correction capability for the actual transmitted signals.
Solution Approach 2:
The patent performs preliminary action by pre-identifying and storing the subset of code words that are actually used in the communication system before decoding operations begin. The reduced code space is prepared in advance based on the specific communication protocol and transmission constraints. This preliminary preparation allows the decoder to operate efficiently on a pre-filtered set of possibilities rather than searching the entire code space during actual decoding, improving both speed and energy efficiency.
2Reliability
If the decoder searches through all code words in the complete code space, then no code word is missed, but the decoding performance is suboptimal because the distance between code words is not maximized
Solution Approach 1:
The patent extracts the essential subset of code words that are actually transmitted in the communication system, removing unnecessary code words from the search space. This extraction creates a reduced code space that contains only the relevant code words, thereby increasing the effective distance between them and improving decoding performance without requiring the complexity of handling the complete code space.
Solution Approach 2:
The patent performs preliminary identification of the actual code words used in the communication system based on a priori knowledge of the transmission protocol. By preparing the reduced code space in advance, the system optimizes the geometric arrangement of code words in the search space, maximizing the minimum distance between adjacent code words and thereby improving error correction capability while reducing computational complexity.
3Reliability
If existing decoders do not utilize a priori knowledge of the code word space, then they are more general and applicable to any code, but they fail to achieve optimal decoding performance for specific communication systems
Solution Approach 1:
The patent extracts and utilizes a priori knowledge specific to the communication system, such as the actual subset of code words transmitted and their statistical properties. By taking out this system-specific information and incorporating it into the decoding process through a reduced code space, the patent achieves optimal error-free detection probability for the specific communication system while maintaining the ability to adapt to different encoders through the general framework of reduced code space decoding.
Data Source
AI summary
A method is provided to decode data encoded by any block code in a manner that substantially improves the error correction capability of the block codes, and that is independent of the encoder. The structure associated with the method desirably allows the testing of those hypotheses that are known to exist, such that one can use the a priori knowledge of the possible set of hypotheses to only search from among them. The method of decoding data is both advantageous and desirable since knowing the subset of the code word space that is being utilized in essence allows the distance between the code words to be increased yielding significant decoding benefits.