Digital TV Receiver Channel Estimation Using Adaptive Noise Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing adaptive processing methods for multi-carrier transmission systems, such as those used in digital terrestrial broadcasting, fail to accurately estimate channel characteristics when desired channel characteristics are unknown, leading to noise-induced estimation errors and reduced receiver performance.
Innovation Solution
A receiving apparatus and method that converts received signals from the time domain to the frequency domain, estimates noise power, calculates channel characteristics using a reference signal, computes autocorrelation values, and determines tap coefficients for a noise removal filter, enabling effective adaptive processing even without known desired channel characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If adaptive processing using Wiener filter principle is performed, then channel estimation accuracy is improved, but device complexity increases due to additional processing units
Solution Approach 1:
The system uses the estimated channel characteristics themselves to generate the desired signal for adaptive processing, making the system self-sufficient without requiring external reference signals or additional complex processing units. The channel estimation unit generates its own training data by utilizing its own output.
Solution Approach 2:
The desired signal acts as an intermediary between the estimated channel characteristics and the adaptive filter, enabling the Wiener filter processing to proceed without requiring direct access to the true channel characteristics, thus simplifying the overall system architecture.
2Reliability
If noise component is removed from channel characteristics estimation, then receiver performance is improved, but measurement precision deteriorates due to difficulty in accurately separating noise from signal
Solution Approach 1:
The system uses feedback by taking the estimated channel characteristics and using them to generate the desired signal, which then feeds back into the adaptive filter to produce refined channel estimates. This iterative feedback process progressively improves noise removal while maintaining measurement accuracy.
Solution Approach 2:
The patent replaces traditional mechanical or direct noise filtering methods with statistical signal processing using autocorrelation and Wiener filter principles, substituting physical noise separation with mathematical transformation and optimization.
3Measurement precision
If SP signal is inserted frequently in time and frequency directions, then channel estimation coverage is improved, but loss of information increases due to reduced data transmission capacity
Solution Approach 1:
The system uses partial action by inserting SP signals at reduced density (once every 12 carriers in frequency direction and once every 4 symbols in time direction) compared to what would be needed for complete channel coverage, relying on autocorrelation processing to compensate for the sparser sampling.
Data Source
AI summary
In a receiving apparatus (100), a noise power estimation unit (107) estimates the noise power included in a received signal in the frequency domain. A channel characteristics calculation unit (105) estimates the channel characteristics in accordance with a reference signal included in the received signal. A tap coefficient calculation unit (108) calculates the tap coefficients for a filtering unit (109) in accordance with noise power included in the received signal and with an autocorrelation value for the channel characteristics. The filtering unit (109) calculates an estimated value for the channel characteristics via filtering processing on the channel characteristics in accordance with the tap coefficients.