Decision-Directed Channel Estimation with Reliability Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
OFDM systems in outdoor wireless communications face challenges with time-variant channels due to multi-path fading and mobility, leading to inter-symbol and inter-carrier interference, which complicates channel estimation and tracking, especially in vehicle-to-vehicle and vehicle-to-infrastructure applications.
Innovation Solution
A decision-directed channel estimation technique that uses both pilot and data symbols, with reliability coefficients to select the most reliable data symbols for channel estimation and tracking, and employs maximal-ratio-combining techniques with multiple receiving antennas to improve estimation accuracy and reduce error propagation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If decision-directed channel estimation using data symbols is employed, then channel estimation accuracy is improved, but error propagation occurs when data symbols are incorrectly detected
Solution Approach 1:
The patent changes the parameter of symbol selection by introducing reliability coefficients to filter and select reliable data symbols for channel estimation, thereby reducing error propagation while maintaining estimation accuracy
Solution Approach 2:
The patent implements feedback mechanisms where channel estimates are continuously updated using selected data symbols, and the reliability of these estimates is monitored to adjust future selections, creating a closed-loop system that mitigates error propagation
2Reliability
If multiple receiving antennas are used for maximal-ratio-combining, then channel estimation robustness is improved, but device complexity increases
Solution Approach 1:
The patent merges the signals from multiple receiving antennas using maximal-ratio-combining technique, where the signals are combined weighted by their signal-to-noise ratios, thereby improving robustness while managing complexity through efficient signal processing
3Reliability
If reliability coefficients are used to select data symbols, then error propagation is reduced, but computational complexity increases
Solution Approach 1:
The patent changes the parameter of symbol selection by introducing reliability coefficients to filter and select reliable data symbols for channel estimation, thereby reducing error propagation while managing computational complexity through efficient filtering criteria
Data Source
AI summary
An embodiment of the time-variant channel estimation in an OFDM transmission system envisages the operations of: a) determining the received signal vector Yj,k for one or more of R receiving antennas, j=1, . . . , R, and at the current time instant k together with the frequency-domain channel-estimate vector Ĥj,k-1 at a preceding time instant k−1 and for each antenna j of one or more of R receiving antennas; b) producing a first estimate Âk of the transmitted digital symbols at the current time instant k on the basis of the aforesaid frequency-domain channel-estimate vectors Ĥj,k-1 at the preceding time instant k−1; c) computing an updated version of the frequency-domain channel-estimate vectors Ĥj,k at the current time instant k and for each antenna j of one or more of R receiving antennas on the basis of the received signal vectors Yj,k at the current time instant k and for each antenna j of one or more of R receiving antennas, using the aforesaid estimate Âk of the transmitted digital symbols at the current time instant k as set of known transmitted digital symbols; d) using the aforesaid updated version of the frequency-domain channel-estimate vectors Ĥj,k at the current time instant k and for each antenna j of one or more of R receiving antennas, for equalizing the paths of the transmission channel and decoding the received signal vectors Yj,k at the current time instant k and for each antenna j of one or more of R receiving antennas; and e) storing the updated versions of the channel-estimate vectors for using them in steps a) and b) at the next time instant k+1.