LPWA Distance Estimation via Neural Network Transfer Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
LPWA networks face challenges in accurately estimating distance and position due to low time resolution of narrow band signals, especially in multi-path environments, which limits their effectiveness in geolocation applications.
Innovation Solution
A method using a deep neural network to estimate distance by measuring the transfer function of the transmission channel over a virtual frequency band, combining forward and backward transfer functions to improve time resolution, and training the network with labelled data to predict the line of sight distance between nodes in LPWA networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of stationary object
If narrow band signals are used in LPWA networks to propagate over long distances, then transmission range is improved, but time resolution and distance estimation accuracy deteriorate
Solution Approach 1:
The patent transforms the problem from the time domain to the frequency domain by measuring transfer functions across multiple frequency points. Instead of directly measuring time of arrival with narrow band signals, the system measures channel characteristics in the frequency domain and uses a neural network to map these frequency measurements to distance estimates, effectively changing the dimension of measurement from temporal to spectral.
Solution Approach 2:
The patent introduces a deep neural network as an intermediary between the raw frequency domain measurements and the final distance estimation. The neural network learns the complex relationship between multi-frequency transfer function characteristics and propagation distance, acting as a mediator that translates frequency domain observations into accurate range estimates without requiring direct time domain measurement.
2Use of energy by moving object
If narrow band signals are used in LPWA networks, then power consumption is reduced, but position estimation accuracy deteriorates
Solution Approach 1:
The patent changes the measurement parameters from time domain characteristics to frequency domain characteristics. By measuring transfer functions at multiple frequency points instead of measuring time of arrival directly, the system maintains compatibility with low-power narrow band LPWA signals while extracting richer channel information that enables accurate distance estimation through the neural network.
3Adaptability or versatility
If traditional trilateration from radio signals is used, then geolocation capability is enabled, but distance resolution deteriorates due to multi-path effects
Solution Approach 1:
The patent converts the harmful multi-path effects into beneficial information. Instead of trying to eliminate or ignore multipath components, the system measures the complete channel transfer function including all paths at multiple frequencies. The neural network learns to extract the direct path distance information from the composite channel characteristics, effectively using the multi-path structure to improve measurement robustness.
Data Source
AI summary
The present invention relates to a method for estimating a distance between nodes of an LPWA network. First and second nodes, respectively, successively transmit a narrow band signal at a plurality of carrier frequencies to each other. The second and first nodes, respectively, receive the transmitted signals and demodulate them into baseband signals. Complex values representative of a forward-backward function of the transmission channel between two nodes are obtained from the baseband demodulated signals. These complex values are then provided to a previously supervisingly trained neural network, the neural network giving an estimation of the distance separating the first node and the second node. The present invention also relates to a method for estimating the position of a node in an LPWA network using the above-noted distance estimation method.