SINR Prediction Using Time-Frequency Region Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radio communication systems face challenges in predicting channel quality due to time selective and frequency selective fading, which limits real-time channel quality assessment and impacts reliability and efficiency, as the Reference Signal (RS) is not intensively present in both domains.
Innovation Solution
A method and device that set time and frequency domain thresholds to divide the time-frequency space into regions, calculate differences between the channel to be estimated and a referenced RS, and determine the SINR based on corresponding SINR determination manners for each region, ensuring accurate prediction by compensating for transmit power spectral density differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Reference Signal (RS) is transmitted intensively in both time domain and frequency domain, then real-time channel quality measurement on all spectrums is improved, but radio communication resources are consumed excessively
Solution Approach 1:
The patent divides the time-frequency space into multiple regions based on time domain thresholds and frequency domain thresholds. Each region is associated with a specific SINR determination manner, allowing selective measurement and prediction strategies for different segments of the channel spectrum, thus reducing overall resource consumption while maintaining measurement accuracy where needed.
Solution Approach 2:
The patent performs preliminary classification of the channel to be estimated into different time-frequency regions before determining SINR. By pre-defining regions and their corresponding determination manners, the system avoids exhaustive measurements across the entire spectrum, thereby conserving radio communication resources while ensuring accurate channel quality assessment.
2Quantity of substance
If RS is transmitted sparsely to save resources, then radio communication resources are conserved, but real-time channel quality prediction accuracy deteriorates
Solution Approach 1:
The patent introduces time-frequency region classification as an intermediary mechanism between sparse RS measurements and channel quality prediction. By categorizing channels into different regions with distinct determination manners, the system can accurately predict channel quality even with limited RS transmissions, as each region has a tailored prediction strategy that compensates for measurement sparsity.
Solution Approach 2:
The patent changes the prediction approach based on the time-frequency region parameters. Different determination manners are applied to different regions, allowing the system to adapt prediction accuracy to the specific characteristics of each region. This parameter-based adaptation ensures high prediction accuracy with sparse measurements by focusing computational resources on critical regions.
3Device complexity
If traditional filtering methods are used for channel quality prediction, then implementation is simple, but prediction accuracy under time-selective and frequency-selective fading deteriorates
Solution Approach 1:
The patent segments the channel prediction problem into multiple time-frequency regions, each handled by a specific determination manner. This segmentation allows the system to address time-selective and frequency-selective fading in a structured way, improving prediction accuracy without requiring overly complex unified filtering methods across the entire spectrum.
Solution Approach 2:
The patent implements dynamic prediction strategies by associating different determination manners with different time-frequency regions. The prediction approach adapts to the specific characteristics of each region, allowing the system to respond dynamically to time-selective and frequency-selective fading conditions rather than using a static filtering method for all channels.
Data Source
Figure 1~2
Figure 3~4
AI summary
The disclosure provides a method for predicting Signal to Interference plus Noise Ratio (SINR) and a device for implementing the method. The method includes: setting a time domain threshold and/or a frequency domain threshold, dividing a time-frequency distance space into different regions according to the set time domain threshold and/or frequency domain threshold, and setting an SINR determination manner for each time-frequency region; calculating a time domain difference and/or a frequency domain difference between a channel to be estimated and a Reference Signal (RS), determining a time-frequency region to which the time domain difference and/or the frequency domain difference belong(s), and determining the SINR of the channel to be estimated according to the SINR determination manner corresponding to the time-frequency region to which the time domain difference and/or the frequency domain difference belong(s). In the disclosure, the relevant fading characteristics of an RS in time domain and frequency domain are fully considered, the SINR of the channel to be estimated is predicted combining the SINR feature of the RS that was transmitted previously by the system, and the prediction result is more accurate.