BCJR Algorithm Symbol Detection Complexity Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems face high complexity in decoding symbols due to large channel memory, leading to inefficiencies in achieving optimal channel capacity, particularly in wireless communication systems like LTE and GSM, where reduced complexity receivers are necessary but struggle to match the performance of full complexity receivers.
Innovation Solution
The method involves computing soft symbol estimates and variance based on a priori Log Likelihood Ratios (LLRs), filtering and interference cancellation using modified Wiener filters, estimating the Mean Square Error (MSE) matrix, and optimizing the BCJR algorithm parameters to reduce memory and improve decoding performance, incorporating iterative processes with an outer ECC decoder.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the BCJR algorithm is used for optimal symbol detection, then detection accuracy is improved, but computational complexity increases exponentially with channel memory
Solution Approach 1:
The patent segments the channel memory into two parts: a short-term memory component handled by a modified BCJR algorithm and a long-term memory component handled by a separate filter. This segmentation allows the complex BCJR algorithm to operate only on the short-term memory portion, reducing computational complexity from exponential in K to exponential in L where L < K, while maintaining detection accuracy through the combined operation of both components
Solution Approach 2:
The patent introduces an intermediary filter that processes the received signal before it enters the BCJR algorithm. This filter captures and removes long-term memory effects, serving as a mediator that prevents these effects from overwhelming the BCJR algorithm. The filter acts as a preprocessing stage that simplifies the input to the BCJR algorithm, enabling it to focus on short-term dependencies without being burdened by long-term channel memory
2Device complexity
If reduced complexity receivers are used to handle large channel memory, then computational complexity is reduced, but detection performance deteriorates
Solution Approach 1:
The patent merges two separate detection components: a filter that handles long-term memory effects and a modified BCJR algorithm that handles short-term memory effects. By combining these two components, the system achieves detection performance close to the optimal full-complexity BCJR algorithm while maintaining reduced computational complexity. The merged system processes signals through both components in sequence, leveraging the strengths of each approach
Solution Approach 2:
The patent employs dynamic adaptation by iteratively updating the filter coefficients and BCJR algorithm parameters based on channel conditions and detected symbols. This dynamic adjustment allows the reduced complexity receiver to optimize its performance for different channel scenarios, maintaining high detection accuracy while adapting to varying memory conditions. The iterative process enables the system to learn from past detections and improve subsequent performance
Data Source
AI summary
The present invention relates to method for detection of symbols, comprising: receiving at least one communication signal y transmitted over a radio channel H, and receiving a priori LLRs of said channel inputs x; computing soft symbol estimates μ and variance D of said channel inputs x based on LLRs of said channel inputs x; filtering and interference cancelling said received signal y by using a first filter and filtered soft symbol estimates μ so as to obtain a filtered and interference cancelled received signal; estimating MSE matrix B of said filtered and interference cancelled received signal; computing an optimal matrix Gopt for a BCJR algorithm based on said MSE estimate B; further filtering said filtered and interference cancelled received signal by using a second filter so as to obtain an input signal r to the BCJR algorithm; and feeding said input signal r into the BCJR algorithm.