Capacitive Sensor Packet Weighting for Faster Low-Noise Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Projected capacitive sensors face challenges in distinguishing weak signals from noise, especially when noise frequencies overlap with signal frequencies, and require quick scan times, making it difficult to achieve accurate capacitance measurements.
Innovation Solution
The method involves generating data packets with varying gain, where samples at the beginning and end of each packet contribute less to the total result, using weighting techniques such as Gaussian, Hamming, or Blackman window curves, to separate noise and signal frequencies through mathematical post-processing or variable gain amplification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If quick scan is performed to maintain good responsiveness, then response speed is improved, but noise interference increases making it difficult to distinguish signal from noise
Solution Approach 1:
The patent applies periodic action by using a window function that is periodically applied to weight samples within each data packet. The window function creates a periodic weighting pattern where samples at the beginning and end of packets (which contain more noise) are weighted less, while central samples are weighted more, thereby improving signal-to-noise ratio while maintaining quick scan capability
Solution Approach 2:
The patent implements local quality by applying different weights to different samples within the same data packet. Specifically, samples at the beginning and end of packets are given lower weights while central samples receive higher weights. This local differentiation allows the system to suppress noise in specific regions (packet boundaries) while preserving signal quality in other regions (packet centers)
2Measurement precision
If noise filtering is enhanced to improve measurement accuracy, then measurement precision is improved, but scan time increases reducing responsiveness
Solution Approach 1:
The patent applies preliminary action by pre-weighting the samples with a window function before integration. This preliminary weighting of samples with appropriate weights (higher for central samples, lower for boundary samples) is performed as part of the normal scanning process, so noise filtering is achieved without requiring additional post-processing time or extended scan duration
3Device complexity
If uniform gain is applied to all samples, then processing simplicity is maintained, but noise at packet boundaries contributes equally reducing overall signal quality
Solution Approach 1:
The patent implements local quality by replacing uniform gain with a window function that provides location-dependent weighting. The window function w[n] assigns different weights based on the position n within the packet, giving lower weights to boundary samples and higher weights to central samples. This local differentiation improves output signal quality by reducing noise contribution from packet boundaries while adding minimal processing complexity through the use of standard window functions
Data Source
AI summary
A method or sensor arrangement for providing capacitive sensor detection with at least one capacitive sensor comprises a transmitting electrode and a receiving electrode. A stimulus at the transmitting electrode is generated and a signal is received from the receiving electrode and data packets are generated, each packet comprising a plurality of samples. The plurality of samples are weighted by providing less gain at a beginning and end of each packet with respect to a center of each packet; and the weighted samples are integrated to generate an output signal for each packet.