SLIC Receiver Multiple Survivor Threads MIMO Demodulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional SLIC receivers face limitations in exploiting useful knowledge about the residual signal, leading to decreased noise covariance as processing progresses, which affects demodulation performance in MIMO systems.
Innovation Solution
The SLIC receiver structure is generalized to allow multiple solutions to survive at each stage, forming multiple solution threads and using a decision processor to select the best overall demodulation solution, with pruning to control complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional SLIC receiver uses unique solution at each stage, then device complexity is reduced, but demodulation performance deteriorates due to propagation of incorrect decisions
Solution Approach 1:
The patent segments the demodulation process into multiple serial stages, where each stage processes a portion of the signal and generates candidate solutions. By dividing the complex demodulation task into manageable stages with multiple candidates each, the system maintains lower overall complexity while improving reliability through cumulative decision-making across stages.
Solution Approach 2:
The patent changes the parameter of solution multiplicity from single to multiple at each stage. This parameter change allows the receiver to explore multiple potential demodulation paths simultaneously, improving demodulation performance by selecting the best overall solution from combined stage outputs rather than propagating potentially incorrect unique decisions.
2Reliability
If multiple solutions survive at each stage, then demodulation performance is improved, but device complexity increases due to multiple solution threads
Solution Approach 1:
The patent applies partial action by maintaining multiple candidate solutions at each stage rather than processing all possible solutions. This partial approach balances performance improvement with complexity control, as it processes enough candidates to achieve better demodulation without the excessive complexity of exhaustive search across all possible solution threads.
Solution Approach 2:
The patent segments the solution space into multiple discrete candidate solutions at each stage, allowing independent processing and evaluation of each candidate. This segmentation enables manageable complexity by treating multiple solutions as separate, processable units rather than a continuous complex search space.
3Measurement precision
If full search over all candidate N-tuples is performed, then measurement precision is improved, but productivity decreases due to computational burden
Solution Approach 1:
The patent segments the full search process into multiple serial stages, where each stage performs a reduced search over a subset of candidates. This segmentation transforms one computationally intensive full search into multiple simpler sequential searches, improving processing speed while maintaining detection accuracy through cumulative refinement across stages.
Solution Approach 2:
The patent performs partial search at each stage by evaluating only a subset of candidate N-tuples rather than the complete search space. This partial action at multiple stages achieves comparable detection precision to full search while significantly improving productivity by reducing the computational burden at each individual stage.
Data Source
AI summary
In one aspect, the present disclosure provides an advantageous generalization of a SLIC receiver structure, by allowing multiple solutions to survive at any one or more of the serial stages included in the SLIC receiver. Where any given stage produces multiple solutions, the next stage of the SLIC receiver processes those multiple input solutions to produce multiple output solutions. Consequently, the contemplated SLIC receiver effectively builds a tree structure with as many levels as there are stages in the SLIC receiver. Such operations allow the SLIC receiver to form two or more solution threads spanning the stages, and to make an improved overall demodulation decision for a received symbol vector by selecting the best solution thread. Further, the additional complexity of allowing multiple survivor solutions at one or more SLIC stages is controlled in one or more embodiments, using survivor pruning for example.