SNR Estimation Using Absolute Value Sum Approximations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing signal-to-noise ratio (SNR) estimation methods in wireless receivers face challenges in accurately calculating SNR due to the inability to directly cancel noise power from total received signal power, leading to computational complexity and bias issues, especially in low SNR regimes.
Innovation Solution
The method involves determining absolute value sum approximations of receive power and noise power, followed by bias compensation to estimate SNR in the decibel domain, allowing direct combination of SNR from multiple antennas without converting back to the linear domain, thereby reducing computational complexity and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If absolute value sum approximation is used to reduce hardware complexity, then device complexity is reduced, but measurement precision deteriorates because noise power cannot be directly canceled
Solution Approach 1:
The patent introduces an intermediary variable (noise power estimate) that is separately calculated and then used to compensate the total power measurement. This mediator enables the cancellation of noise power effect without requiring complex hardware operations, thus maintaining low device complexity while improving measurement precision through the intermediate compensation step.
Solution Approach 2:
The patent changes the parameter representation from direct power values to logarithmic domain (dB) values. By working in the logarithmic domain, the system can perform SNR estimation using simple addition and subtraction operations on the approximated values, thereby maintaining measurement precision while keeping hardware complexity low.
2Measurement precision
If direct subtraction of noise power from total received signal power is attempted, then measurement precision improves, but device complexity increases due to computational requirements
Solution Approach 1:
The patent uses cheap approximation methods (absolute value sum) instead of expensive precise calculations. The approximated values are 'disposable' in the sense that they are computed once and then used in simple algebraic manipulations to achieve the final SNR estimate, avoiding the need for complex real-time computations while maintaining acceptable precision.
Solution Approach 2:
The patent transforms the computational problem from the linear domain to the logarithmic domain, where complex division and subtraction operations become simpler addition and subtraction operations. This parameter change reduces computational complexity while preserving the essential measurement precision through the mathematical relationship between the domains.
3Measurement precision
If SNR is estimated in linear domain and then converted to decibel domain, then measurement precision is maintained, but device complexity increases due to additional conversion steps
Solution Approach 1:
The patent inverts the conventional approach by working directly in the decibel domain from the beginning, rather than computing in the linear domain and then converting. This inversion eliminates the need for logarithmic conversion steps while maintaining measurement precision, as the absolute value sum approximation and noise compensation can be directly performed in the logarithmic domain using simplified algebraic operations.
Data Source
AI summary
A method and apparatus are provided. The method includes receiving a signal sequence and a noise sequence, determining a first absolute value sum approximation of a receive power of the signal sequence, determining a second absolute value sum approximation of a noise power of the noise sequence, and determining a signal-to-noise ratio (SNR) based on the first absolute value sum approximation and the second absolute value sum approximation.