Distance Estimation Using Multi-Scale Transform for Burst Noise Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional distance estimation methods in wireless communications struggle with accuracy due to environmental noise, particularly burst noise, which can lead to significant errors in determining the distance between nodes, as they fail to accurately identify and remove noise components, often affecting the validity of signal strength measurements.
Innovation Solution
A method that employs multi-scale transform to project received signal strengths onto multiple scales, allowing for the accurate identification and removal of burst noise components, and utilizes a multi-scale empirical mapping relationship to estimate real-time distances, incorporating both timely and historical signal strength data for precise distance determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a filter is used to remove burst noise before distance estimation, then the distance estimation process can proceed, but it is easy to remove valid signals and/or impossible to completely remove burst noise, resulting in offset in the distance estimation result
Solution Approach 1:
The patent segments the received signal strength sequence into multiple scales using multi-scale transform (e.g., wavelet transform). This decomposition separates the signal into different frequency components and time scales, allowing burst noise to be identified and removed from specific scales while preserving valid signals in other scales. The segmentation enables selective noise removal without affecting the entire signal uniformly.
Solution Approach 2:
The patent extracts burst noise components from the received signal strength sequence by analyzing multi-scale transform results. By identifying abnormal values in specific scales and time positions, the method extracts and removes only the burst noise portions while retaining valid signal components. This extraction approach prevents the loss of valid signals that would occur with traditional filtering.
2Ease of manufacture
If conventional filtering is applied to remove burst noise, then the processing is simple, but the position of burst noise cannot be accurately determined, leading to incomplete noise removal and offset in results
Solution Approach 1:
The patent transforms the one-dimensional signal strength sequence into a multi-dimensional representation through multi-scale transform. This adds time-scale and frequency dimensions, enabling the identification of burst noise not only by amplitude but also by its temporal and spectral characteristics. The additional dimensions provide more information for accurate noise position determination while maintaining computational feasibility.
3Ease of operation
If burst noise is not taken into account in distance estimation, then the estimation process is straightforward, but when burst noise occurs, the distance estimation result becomes inaccurate
Solution Approach 1:
The patent performs preliminary action by conducting multi-scale transform and burst noise removal before the distance estimation process. This preprocessing step prepares the signal by eliminating burst noise components in advance, ensuring that the subsequent distance estimation uses clean data. The preliminary noise removal prevents accuracy degradation while keeping the main estimation process simple.
Solution Approach 2:
The patent introduces multi-scale transform as an intermediary process between signal reception and distance estimation. This intermediary step transforms the raw signal into a multi-scale representation where burst noise becomes distinguishable from valid signals. The transform acts as a mediator that facilitates noise identification and removal without complicating the final distance calculation.
Data Source
AI summary
Disclosed is a method of estimating the distance between two nodes. The method includes obtaining an original real-time strength sequence, its elements being strengths of signals between the two nodes at respective time slots of a first time period; conducting multi-scale transform with respect to the original real-time strength sequence so as to acquire plural real-time strength component sequences which are results of the original real-time strength sequence projected onto plural scales; removing at least one burst noise element in each real-time strength component sequence so as to get plural updated real-time strength component sequences; estimating, based on the plural updated real-time strength component sequences and a multi-scale empirical mapping relationship, an empirical real-time distance, the multi-scale empirical mapping relationship being used for expressing correspondence relations between strength components and distance components; and determining, based on at least the empirical real-time distance, a real-time distance between the two nodes.