Sliding-Window SISO Decoding with Shifted Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Soft input/soft output (SISO) algorithms, such as MAP processing for turbo codes, face reliability issues due to high memory requirements and poor output soft value reliability at window boundaries, especially in sliding window traversal, which affects communication link performance, particularly at higher data rates.
Innovation Solution
Varying window placements between iterations of the SISO algorithm shifts window boundaries relative to the input sequence, improving reliability attributes of output soft values by changing window sizes and offset values, thereby addressing the reliability concerns at the boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of stationary object
If sliding window traversal is used to reduce memory requirements, then memory usage is reduced, but output soft value reliability at window boundaries deteriorates
Solution Approach 1:
The patent divides the trellis into multiple processing windows to reduce memory requirements. Each window processes a subset of the trellis stages, allowing state metrics to be accumulated only over the window length rather than the full trellis length. This segmentation enables decoding of long sequences with limited memory resources.
Solution Approach 2:
The patent applies preliminary actions by using overlapping windows where the overlap regions are processed in advance to warm up the recursions. By processing overlapping regions first, the algorithm ensures that when the main decoding proceeds, the recursions have already accumulated sufficient state metrics to produce reliable output soft values at window boundaries.
2Reliability
If window overlap is increased to improve reliability at boundaries, then output soft value reliability improves, but decoder throughput deteriorates
Solution Approach 1:
The patent applies partial action by using a moderate amount of window overlap rather than extensive overlap. The overlap is sufficient to warm up the recursions and improve boundary reliability, but not so large as to significantly reduce throughput. This finds an optimal balance between reliability improvement and maintaining decoder productivity.
3Device complexity
If fixed window placements are used, then processing is simplified, but reliability at fixed boundary positions deteriorates
Solution Approach 1:
The patent introduces dynamic window placements that shift between iterations rather than remaining fixed. The window positions are adjusted based on iteration number, ensuring that different portions of the sequence are at window boundaries at different iterations. This dynamic approach prevents consistent reliability degradation at the same positions while maintaining relatively simple processing.
Data Source
AI summary
In one or more embodiments, a method of processing a soft value sequence according to an iterative soft-input-soft-output (SISO) algorithm comprises carrying out sliding-window processing of the soft value sequence in a first iteration using first window placements and in a second iteration using second window placements, and varying the window placements between the first and second iterations. In at least one embodiment, a communication receiver circuit is configured to carry out iterative SISO processing, wherein it processes a soft value sequence using sliding windows, and wherein it varies window placements between one or more iterations. The communication receiver circuit comprises, for example, all or part of a turbo decoding circuit or other type of iterative block decoding circuit, an equalization and decoding circuit, a soft demodulation and decoding circuit, a multi-user detection and decoding circuit, or a multiple-input-multiple-output detection and decoding circuit.