Kalman Smoother Channel Estimation With Extended State Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-ideal channel responses in wireless communications systems degrade performance, and existing Kalman smoothers employing state augmentation impose high computational burdens and power consumption.
Innovation Solution
Implementing a Kalman smoother using an extended state vector instead of state augmentation to reduce computational burden and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a Kalman smoother employing state augmentation is used for non-causal channel estimation, then accurate channel response estimation is achieved, but computational burden and power consumption increase significantly
Solution Approach 1:
The patent changes the parameter of state vector representation from augmented state (containing both channel response and channel history) to extended state (containing only channel response with historical information embedded in transition model). This parameter change reduces the state vector dimension from N×(D+1) to N, directly lowering computational complexity while preserving estimation accuracy through the modified state transition model that incorporates channel history via powers of the transition matrix.
2Measurement precision
If a Kalman smoother employing state augmentation is used for non-causal channel estimation, then accurate channel response estimation is achieved, but power consumption increases significantly
Solution Approach 1:
The patent reduces power consumption by changing the state vector parameter from augmented to extended form. The extended state vector approach requires fewer mathematical operations (matrix multiplications and additions) per estimation cycle, directly reducing the energy consumed by the processor. The trade-off is acceptable because the extended state model compensates by embedding historical channel information in the transition model, maintaining estimation accuracy with lower computational energy expenditure.
3Reliability
If state augmentation is used in Kalman smoother, then channel history information is captured, but the number of state vector elements increases to kp(D+1)
Solution Approach 1:
The patent extracts the channel history information from the state vector and relocates it to the state transition model. Instead of storing explicit historical channel values in the state vector (which would require kp(D+1) elements), the extended state approach uses the transition matrix raised to power D to represent the relationship between current and past channel states. This extraction reduces the state vector to only N elements while preserving historical information in the model structure.
Data Source
Figure 1A~1B
Figure 2
Figure 3
AI summary
A system and a method are disclosed for channel estimation. In some embodiments, a method includes: calculating (600), by a receiver, a first channel estimate, at an nth point in time, the calculating including calculating an (n+D)th state vector, the (n+D)th state vector corresponding to a channel history at an (n+D)th point in time, D being a positive integer; performing (605) signal processing of a received signal, based on the first channel estimate, to generate processed data; and transmitting (610) the processed data to a data consumer, the (n+D)th state vector including elements from p update intervals, the first channel estimate at the nth point in time including an element for each of k subcarriers, k being a positive integer, and the (n+D)th state vector having fewer than kp(D+1) elements.