TY - GEN
T1 - BSLR
T2 - 2024 IEEE International Symposium on Antennas and Propagation and INC/USNCURSI Radio Science Meeting, AP-S/INC-USNC-URSI 2024
AU - Hameed, Hira
AU - Ishabakaki, Prisila Alex
AU - Farooq, Muhamamd
AU - Fatima, Aisha
AU - Arshad, Kamran
AU - Assaleh, Khaled
AU - Imran, Muhammad Ali
AU - Abbasi, Qammer H.
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Communication is a key means of exchanging information, but many people are unable to engage in verbal communication. These individuals often use non-verbal modes, like sign language. Existing British Sign Language (BSL) systems, primarily camera-based, face challenges such as poor illumination and privacy concerns. This paper introduces the use of Wi-Fi signals to detect BSL, with data represented as Channel State Information (CSI). The CSI is denoised and 15 features are extracted. Six classes Drink, Eat, Help, Stop, Walk, and Empty, are identified, corresponding to situations with the subject seated. Machine Learning (ML) models like Neural Network Pattern Recognition, Ensemble (Bagged Trees), KNN (Weighted KNN), and Naïve Bayes (Kernel Naïve Bayes) process this information. These models effectively classify the gestures, with Neural Network Pattern Recognition achieving a classification accuracy of 100%.
AB - Communication is a key means of exchanging information, but many people are unable to engage in verbal communication. These individuals often use non-verbal modes, like sign language. Existing British Sign Language (BSL) systems, primarily camera-based, face challenges such as poor illumination and privacy concerns. This paper introduces the use of Wi-Fi signals to detect BSL, with data represented as Channel State Information (CSI). The CSI is denoised and 15 features are extracted. Six classes Drink, Eat, Help, Stop, Walk, and Empty, are identified, corresponding to situations with the subject seated. Machine Learning (ML) models like Neural Network Pattern Recognition, Ensemble (Bagged Trees), KNN (Weighted KNN), and Naïve Bayes (Kernel Naïve Bayes) process this information. These models effectively classify the gestures, with Neural Network Pattern Recognition achieving a classification accuracy of 100%.
KW - British Sign Language
KW - Machine Learning
KW - RF sensing
KW - Wi-Fi Signal
UR - https://www.scopus.com/pages/publications/85207051047
U2 - 10.1109/AP-S/INC-USNC-URSI52054.2024.10686089
DO - 10.1109/AP-S/INC-USNC-URSI52054.2024.10686089
M3 - Conference contribution
AN - SCOPUS:85207051047
T3 - IEEE Antennas and Propagation Society, AP-S International Symposium (Digest)
SP - 345
EP - 346
BT - 2024 IEEE International Symposium on Antennas and Propagation and INC/USNCURSI Radio Science Meeting, AP-S/INC-USNC-URSI 2024 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 14 July 2024 through 19 July 2024
ER -