Polynomial Mixture for Frequency Domain Multiuser Channel Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing channel estimation methods in frequency-multiplexed communication systems are limited by the static assumption, which restricts the accuracy and number of signal sources that can be jointly estimated, especially in scenarios with multiple simultaneous data streams and varying channel coefficients across subcarriers.
Innovation Solution
The method employs fitting functions, such as polynomials, to approximate variations in channel coefficients as a function of subcarrier frequency, allowing for more accurate joint channel estimation by selecting groups of subcarriers and using algorithms like the Kalman filter to determine channel estimates, thereby overcoming the limitations of the static assumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the static assumption is used for channel estimation, then the estimation process is simple, but the accuracy of channel estimation deteriorates when channel coefficients vary across subcarriers
Solution Approach 1:
The patent transitions from a static channel estimation model to a dynamic one by using polynomial fitting functions that adapt to channel variations across frequency. The fitting functions (e.g., polynomials of order M) dynamically adjust to capture the frequency-dependent behavior of channel coefficients, allowing the system to handle time-varying and frequency-selective channels while maintaining computational tractability.
Solution Approach 2:
The patent changes the parameter representation from constant coefficients (static assumption) to polynomial coefficients that vary with frequency. By representing channel coefficients as evaluations of polynomial fitting functions with frequency-dependent parameters, the system achieves higher accuracy in modeling channel variations while keeping the number of parameters manageable through the polynomial order M.
2Quantity of substance
If more signal sources are jointly estimated, then the system capacity increases, but the dimensionality of the estimator increases requiring more reference symbol samples
Solution Approach 1:
The patent segments the channel estimation problem by applying polynomial fitting functions to groups of subcarriers. Instead of estimating all subcarriers independently for all users, the method divides the frequency spectrum into subcarrier groups and uses low-order polynomial fits within each group. This segmentation reduces the effective dimensionality of the estimator while maintaining accuracy across a larger number of signal sources.
Solution Approach 2:
The patent introduces a polynomial order dimension M to reduce the overall system dimensionality. By fitting polynomials of order M < K (where K is the number of subcarriers per group), the system effectively projects the high-dimensional channel estimation problem onto a lower-dimensional manifold, enabling joint estimation of more signal sources with the same number of reference symbols.
3Measurement precision
If the number of adjacent frequency subcarriers for smoothing is increased, then the error reduction improves, but the static assumption becomes invalid due to channel coefficient variations
Solution Approach 1:
The patent replaces the static smoothing assumption with dynamic polynomial fitting that explicitly models channel variations across frequency. The polynomial fitting functions adapt to local channel conditions in each subcarrier group, allowing the system to use larger groups of subcarriers without violating the underlying assumption of channel behavior, thereby reducing error without relying on the invalid static assumption.
4Measurement precision
If polynomial fitting functions are used instead of static assumption, then channel estimation accuracy improves, but computational complexity increases
Solution Approach 1:
The patent applies partial polynomial fitting by using low-order polynomials (order M) that capture the essential channel variations without over-fitting. By selecting M < K (where K is the number of subcarriers per group), the system achieves sufficient accuracy with reduced computational complexity, applying just enough polynomial complexity to model the channel behavior without the full burden of high-order or exact fitting.
Data Source
AI summary
Methods and systems for obtaining improved channel estimates for frequency-multiplexed data transmissions such as OFDM, OFDMA, or SC-FDMA transmissions overcome the limitations of the static assumption by using a polynomial or other fitting function to fit and model the frequency dependence of the channel coefficients, so that estimates can be applied to larger subcarrier groups. Some embodiments provide channel estimates for a singular signal source, while other embodiments provide joint channel estimates for a plurality of signal sources. In embodiments, selection of the fitting functions is influenced by all previously determined channel estimates. In some embodiments, a tracking algorithm allows use of the lowest possible order of polynomial or other fitting function to model the frequency dependence of the channel coefficients, whereby the channel estimate is continually shifted in frequency while the order of the polynomial or other fitting function remains low.