TY - GEN
T1 - Busy/Idle Duration Prediction for Video and Audio WLAN Traffics Using Autoregressive Predictor with Data Categorization
AU - Hou, Yafei
AU - Kawasaki, Shun
AU - Denno, Satoshi
N1 - Publisher Copyright:
© 2022 Global IT Research Institute-GiRI.
PY - 2022
Y1 - 2022
N2 - Due to huge access from massive devices and peoples such as hospitals, railway stations and airports, wireless local area network (WLAN) is required to have high spectrum efficiency (SE). One of the most intensively researched techniques for wireless LAN systems is cognitive radio (CR) technique which is expected to solve such issue by modeling and predicting of channel status from the current statistics information of spectrum usage. In this paper, we investigate the prediction performance of busy/idle (B/I) duration of two major and widely used wireless services: video service; and audio service using an auto-regressive (AR) based predictor. We first investigate the modeling of their busy/idle duration and analyze their predictability based on predictability theory. Then, we categorize the durations of recent B/I statuses with their ranges to make the duration of the next status be distributed into different sets or streams with different ranges. From their predictability and prediction performance using the low-complexity AR-based predictor, we can confirm that data categorization can largely improve the prediction performance of partial time-series data.
AB - Due to huge access from massive devices and peoples such as hospitals, railway stations and airports, wireless local area network (WLAN) is required to have high spectrum efficiency (SE). One of the most intensively researched techniques for wireless LAN systems is cognitive radio (CR) technique which is expected to solve such issue by modeling and predicting of channel status from the current statistics information of spectrum usage. In this paper, we investigate the prediction performance of busy/idle (B/I) duration of two major and widely used wireless services: video service; and audio service using an auto-regressive (AR) based predictor. We first investigate the modeling of their busy/idle duration and analyze their predictability based on predictability theory. Then, we categorize the durations of recent B/I statuses with their ranges to make the duration of the next status be distributed into different sets or streams with different ranges. From their predictability and prediction performance using the low-complexity AR-based predictor, we can confirm that data categorization can largely improve the prediction performance of partial time-series data.
KW - Autoregressive predictor
KW - Channel status prediction
KW - Data categorizationn
KW - WLAN traffic
UR - http://www.scopus.com/inward/record.url?scp=85127485543&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85127485543&partnerID=8YFLogxK
U2 - 10.23919/ICACT53585.2022.9728885
DO - 10.23919/ICACT53585.2022.9728885
M3 - Conference contribution
AN - SCOPUS:85127485543
T3 - International Conference on Advanced Communication Technology, ICACT
SP - 259
EP - 264
BT - 24th International Conference on Advanced Communication Technology
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 24th International Conference on Advanced Communication Technology, ICACT 2022
Y2 - 13 February 2022 through 16 February 2022
ER -