Modem Chip Interference Whitening Using Reliable PDSCH Samples
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing interference whitening techniques in NR communication systems face challenges with low whitening performance due to limited samples in DMRS-based filters and high power consumption and unreliability in PDSCH-based filters, necessitating the development of a high-accuracy whitening filter using a large number of reliable samples.
Innovation Solution
A modem chip selects reliable resource elements (REs) from the PDSCH to generate a whitening filter based on channel estimates, employing a processor to generate noise plus interference vectors and filters, and determine sample REs for improved interference whitening.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If whitening filters are generated based on DMRS, then the process is simple and fast, but the whitening performance is low due to a small number of samples
Solution Approach 1:
The patent segments the PDSCH REs into two groups: those used for generating whitening filters (first REs) and those used for data transmission (second REs). This segmentation allows the system to utilize a subset of REs for filter generation, achieving sufficient whitening performance without requiring all PDSCH REs, thus resolving the contradiction between sample quantity and performance.
Solution Approach 2:
The patent extracts and utilizes only the necessary portion of PDSCH REs (first REs) for generating whitening filters, while leaving the remaining REs (second REs) for data transmission. This extraction approach obtains sufficient samples for accurate whitening filter generation without consuming all available REs, thereby improving whitening performance while maintaining data transmission efficiency.
2Measurement precision
If whitening filters are generated based on PDSCH, then the number of samples is large, but the reliability of samples is low and power consumption is significant
Solution Approach 1:
The patent segments PDSCH REs into first REs for filter generation and second REs for data transmission. By processing only the first REs for whitening filter generation, the system reduces computational load and power consumption while still obtaining sufficient samples for accurate whitening, thus resolving the contradiction between performance and energy usage.
Solution Approach 2:
The patent applies partial action by using only a portion of the available PDSCH REs (first REs) for generating whitening filters, rather than processing all REs. This partial processing approach provides sufficient samples for accurate whitening while significantly reducing the computational resources and power consumption required compared to processing all PDSCH REs.
3Measurement precision
If all PDSCH REs are used for generating whitening filters, then the number of samples is maximized, but the data transmission capability is reduced
Solution Approach 1:
The patent segments PDSCH REs into two distinct sets: first REs dedicated to whitening filter generation and second REs dedicated to data transmission. This segmentation enables the system to maximize whitening performance by using sufficient samples from the first REs while preserving data transmission capability through the second REs, effectively resolving the contradiction between these two functions.
Solution Approach 2:
The patent extracts a subset of PDSCH REs (first REs) for whitening filter generation, leaving the remaining REs (second REs) available for data transmission. This extraction strategy ensures that sufficient samples are obtained for accurate whitening while maintaining the data transmission capability through the unused second REs, thus resolving the contradiction between whitening performance and data transmission productivity.
Data Source
AI summary
A modem chip includes a transceiver that receives receive a reference signal that is allocated a plurality of first resource elements (REs) and a data signal that is allocated a plurality of second REs, and a processor that generates generate a first noise plus interference vector, based on the reference signal and a channel estimate matrix, generates a first filter, based on the first noise plus interference vector, the first filter being configured to filter interference noise in the data signal, and generates a plurality of estimate values for data by performing interference filtering on the data signal, based on the first filter. The processor further determines at least one sample RE among the second REs, based on sample selection information, and, based on the at least one sample estimate value corresponding to the at least one sample RE, generates a second filter.