ICACT20210239 Question.1
Questioner: rashid_zaidi@yahoo.com 2021-02-09 오후 1:54:21 |
ICACT20210239 Answer.1
Answer by Auhor yfhou@okayama-u.ac.jp 2021-02-09 오후 1:54:21
Chrome  Click!! |
| Can you please summarize that how your models help achieving the goals of smart hospital/smart house? |
Smart hospitals and smart houses are just examples, and the technology used in 5g systems requires higher capacity. However, the available wireless resources are limited, so it is necessary to use free resources as efficiently as possible. This model is a way to achieve that. |
ICACT20210239 Question.2
Questioner: ra-maneekut@kddi-research.jp 2021-02-10 오전 8:47:46 |
ICACT20210239 Answer.2
Answer by Auhor yfhou@okayama-u.ac.jp 2021-02-10 오전 8:47:46
Chrome Click!! |
| Do the moving people/object (while transmitting or receiving data) or other electromagnetic noise source (ex. TV, radio, microwave oven) affect to the fitting model and fitting characteristic ? |
Thanks a lot for your question. Yes, the moving people/object or electromagnetic noise source will largely impact on the accuracy of fitting model espencially for video service. The reason is that audio and video applications usually have some fixed transmission patterns which is benefitful for model fitting when the interference sources are few. |
ICACT20210239 Question.3
Questioner: ra-maneekut@kddi-research.jp 2021-02-10 오후 11:29:04 |
ICACT20210239 Answer.3
Answer by Auhor yfhou@okayama-u.ac.jp 2021-02-10 오후 11:29:04
Chrome Click!! |
| Is this prediction accuracy adequate for finding spectrum hole in real-time scenario ?
What are the next steps to improve the accuracy of this model ? |
Thanks a lot for your question. Yes, the method can be used for a real-time scenario. But the prediction accuracy is perhaps not good because many traffic patterns will be included in the busy and idle duration data for a real-time case. Data categorization just gives a simple way to improve the accuracy of prediction for partial time-series data. Maybe we can use other efficient ML method such as LSTM(Long short-term memory) method to further improve the accuracy of partial data. |
