ICACT20210200 Slide.18
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
This slide show Performance evaluation of Average estimation error of different estimators.
|
|
ICACT20210200 Slide.17
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
This slide shows Training result of Network situation estimator.
|
|
ICACT20210200 Slide.16
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
This slide shows Fitting accuracy of each regression order.
|
|
ICACT20210200 Slide.15
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
|
ICACT20210200 Slide.14
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
|
ICACT20210200 Slide.13
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
Finally, I conclude this presentation.
This presentation proposed a method to estimate the retransmission rate of the background traffic for various scan rates through an empirical analysis based on a computer simulations considering two different Wi-SUN sensor networks.
We confirmed that the proposed method can estimate the retransmission rate of the target network with the average error lower than 5.8% regardless of the situation of the target network and the scan rate.
|
|
ICACT20210200 Slide.12
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
This slide shows the average estimation error of the retransmission
rate of background traffic for each scan rate alpha and beta when the estimators are trained for both Cases 1 and 2.
Scan rate alpha is the scan rate when observed scan response delay
Scan rate beta is estimation target of retransmission rate of background traffic.
This result shows that the average estimation error is lower than 5.8% regardless of the situation of the target network and the scan rate.
|
|
ICACT20210200 Slide.11
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
This slide shows training result of Retransmission rate estimator.
According these training results of estimators, the shapes of curves are different between two network situations.
So, Network situation estimator will helpful for improving the estimation error by selecting the network situation of the target network.
|
|
ICACT20210200 Slide.10
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
This slide shows training result of COR estimator.
|
|
ICACT20210200 Slide.09
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
COR estimator and Retransmission rate estimator estimate their estimation target (which is COR or retransmission rate) of the background traffic.
These values are estimated from the mean of scan response delay using a regression function.
This function is corresponding to the estimated situation of target network and the scan rate which is used for observing the scan response delay.
This figure shows a estimation example of COR of background traffic for scan rate alpha and network situation Case 2.
|
|
ICACT20210200 Slide.08
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
This slide shows training result of Network situation estimator.
These figures shows the regression function for mean of scan response delay and its variance.
These functions are modeled using the least squares method with regression order 3.
According to these results, the shapes of curves are different between two network situations.
So, Network situation estimator can find the index which gives the minimum distance between observed variance of scan response delay and Estimated one.
|
|
ICACT20210200 Slide.07
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
Network situation estimator estimates the situation of the target network by finding an index which gives the minimum distance between observed variance of scan response delay and Estimated one.
This figure shows a estimation example for scan rate alpha when observed scan response delay.
|
|
ICACT20210200 Slide.06
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
The proposed method consists of three kinds of estimators.
Network situation estimator aims to estimate the network situation of the target network.
COR (Channel Occupation Rate) estimator aims to estimate the COR in the background traffic.
Retransmission rate estimator aims to estimate the retransmission rate of the background traffic for various scan rates at the estimated COR.
|
|
ICACT20210200 Slide.05
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
This slide shows Network Situations of Simulations.
Following situations of the target network are assumed.
Case 1 is assumed with different transmission intervals of the background traffic.
Case 2 is assumed With different number of terminals which transmit the background traffic.
|
|
ICACT20210200 Slide.04
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
The scanner needs to determine the proper scan rate from the current scan rate and obtained scan response delay.
In order to realize this, the scanner needs to estimate the retransmission rate of the background traffic at a scan rate different from the current scan rate.
Therefore, we study a method to estimate the retransmission rate of the background traffic for various scan rates through followings.
First, we conduct a computer simulations considering a Wi-SUN sensor network with two different network situations.
Second, we derive the relationship (which is regression function) between the scan response delay and the retransmission rate of background traffic based on the simulation results.
|
|
ICACT20210200 Slide.03
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
According to reference [5], Scan success rate easily decreases in Wi-SUN which has a small network capacity.
The reason is that the scan traffic collides with the background traffic and reaches the limit of the number of retransmission in the target network.
In this situation, controlling the scan rate will improve the degradation of the retransmission rate.
However, the degradation of the retransmission rate and the scan processing time are in trade-off relationship.
In order to treat this trade-off, the scanner need to grasp the retransmission rate of background traffic due to the scan performing.
|
|
ICACT20210200 Slide.02
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
Network scan is useful to find vulnerable IoT devices for cyber-attacks.
However, when the network scan is performed toward wireless networks with a small network capacity like LPWA (for example, Wi-SUN, LoRa), these are easily congested.
In order to finish the network scan without disturbing the background traffic quickly, the network scanner needs to keep the retransmission rate of the background traffic due to the scan traffic at an allowable level.
|
|
ICACT20210200 Slide.01
[Big slide for presentation]
[YouTube] |
Chrome Text-to-Speach Click!! |
 |
I'm Kenta Suzuki from ATR, Japan.
Today, I introduce about "A Study on Estimation of Retransmission Rate of Background Traffic for Various Scan Rates with Scan Response Delay."
|
![[Big slide for presentation]](/upload/2021/0200/Presentation/15/スライド15.jpg){kind=link}
![[Big slide for presentation]](/upload/2021/0200/Presentation/14/スライド14.jpg){kind=link}