UWB Clock Drift Estimation Using Trellis Path Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ultra Wide Band (UWB) impulse radio systems face challenges in accurately estimating clock drift due to quartz crystal inaccuracies, which can lead to synchronization loss, especially in low-cost, low-power devices intended for ad hoc sensor networks, as existing methods like Early-Late Gate Synchronizer circuits require high sampling frequencies, resulting in high power consumption and limited accuracy.
Innovation Solution
A trellis-based method for estimating clock drift that samples UWB pulse train signals, calculates trellis information, and processes it to determine the path metric, allowing for accurate drift estimation with reduced complexity and memory usage, suitable for both coherent and non-coherent receiver architectures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Early-Late Gate Synchronizer circuits are used for clock drift estimation, then synchronization accuracy is improved, but power consumption increases due to high sampling frequency requirements
Solution Approach 1:
The patent changes the fundamental parameter of how drift estimation is performed by using a trellis-based maximum likelihood approach that processes sampled signal values through dynamic programming, replacing the high-frequency sampling requirement with a computationally intensive but low-power signal processing algorithm that achieves the same synchronization accuracy
Solution Approach 2:
The patent substitutes the hardware-based Early-Late Gate Synchronizer circuit with a software-based trellis processing algorithm, replacing the mechanical/electrical synchronization mechanism with an information-processing approach that achieves synchronization through computational analysis of sampled signal characteristics
2Measurement precision
If high sampling frequency is used in ELGS circuits, then drift estimation accuracy is improved, but device complexity and memory requirements increase
Solution Approach 1:
The patent applies partial action by sampling the UWB signal at standard lower frequencies and then using trellis processing to extract drift information from a limited number of sampled points, achieving accurate drift estimation without requiring excessive sampling frequency or processing all possible signal parameters
Solution Approach 2:
The patent segments the drift estimation problem into discrete states represented by the trellis diagram, where different paths through the trellis represent different drift scenarios, allowing the system to process and evaluate multiple drift hypotheses through dynamic programming without requiring high-frequency sampling
3Reliability
If ELGS circuits are used for synchronization, then synchronization capability is improved, but cost increases due to high performance requirements
Solution Approach 1:
The patent employs cheap standard crystals for clock generation in both transmitter and receiver, accepting that these low-cost components will have drift variations, and then uses software-based trellis processing to compensate for the drift, replacing expensive high-precision hardware with affordable components plus computational correction
Data Source
AI summary
A method is for estimating drift between a first clock used in a digital transmission processing of a first Ultra Wide Band (UWB) pulse train signal and a second clock used in a digital reception processing of a second UWB pulse train signal resulting from a transmission of the first UWB pulse train signal. The method may include sampling the second UWB pulse train signal, and calculating trellis information representative of a trellis having reference paths respectively associated to different reference values of the drift and including sample transitions of a sampled third signal from the sampled second UWB pulse train signal. The method may further include processing the sampled third signal along the trellis for obtaining a path metric for each processed reference path, and selecting the processed reference path having a greatest path metric, the drift being the reference value associated to the selected processed reference path.