# 碩論 (2021/5/26) 第17次
> [碩論 (2021/1/6) 第1次](https://hackmd.io/lQUIFZCdSHOLwzvwI9WsOw?both)
> [碩論 (2021/1/20) 第2次](https://hackmd.io/IbJDXXacSaetaHRpinuqpA?both)
> [碩論 (2021/1/27) 第3次](https://hackmd.io/mzur4HIZRF66sjkMear1rQ)
> [碩論 (2021/2/3) 第4次](https://hackmd.io/HBvT3a8tS9CxdUcCcQVkqg)
> [碩論 (2021/2/24) 第5次](https://hackmd.io/MvviEMO_TyG1Xqd5sQTSPw)
> [碩論 (2021/3/3) 第6次](https://hackmd.io/X4KKgcnnQMieS-dvl1lYyA)
> [碩論 (2021/3/17) 第7次](https://hackmd.io/r2EZTug-Q3OHScgs0eMi5w)
> [碩論 (2021/3/25) 第8次](https://hackmd.io/uxTGxw7LQTGkd2CU1gUIAQ)
> [碩論 (2021/3/31) 第9次](https://hackmd.io/1fXLcGDBSC2txhP51G7nAQ)
> [碩論 (2021/4/7) 第10次](https://hackmd.io/ipFw-4TxRd2xTbNQ6-DfMA)
> [碩論 (2021/4/14) 第11次](https://hackmd.io/KA_dhLtxT1C3dSmKeXhKhg)
> [碩論 (2021/4/21) 第12次](https://hackmd.io/B7WYI8G1TnaPpYYKmuVWzw)
> [碩論 (2021/4/28) 第13次](https://hackmd.io/tPyYVp9MQ32MOA7secTYiw)
> [碩論 (2021/5/5) 第14次](https://hackmd.io/QqtAESfUTpCyMKwFa6GwHg)
> [碩論 (2021/5/12) 第15次](https://hackmd.io/-5rUjPcoQJeirb-chas7MQ)
> [碩論 (2021/5/19) 第16次](https://hackmd.io/f3l9hmcLTqGItRMu9QGAPw)
> [碩論 (2021/5/26) 第17次](https://hackmd.io/9OAqe43tQOur8J8-deBMZw)
> [碩論 (2021/6/2) 第18次](https://hackmd.io/xDj3opLhSg2P5K6MpYJU6g)

###### tags: `碩論地獄`

:::info
[TOC]
:::

## Yen aka 心好累，好想畢業

### 雲環境中基於虛擬機自我檢查偵測惡意行為之研究
> 英文題目:Using Virtual Machine Introspection to Detect Malicious Behavior for Cloud Environment

1. (Yin)
2. (古)
3. (Yen)
4. (Cen)
5. (槍熊)
6. (Riz)
7. (晁誌)
8. (祐綸)
9. (韋德)
10. (晨偉)
11. (宗琪)
12. (議員)
13. (日日)

* time
* Intro - Introduction :
* Motivation & Goals :
* Related Work :
* System :
* Experiment :
* Conclusions :
* Total :

* 老師
---

## Gu aka 原諒你幾回，再讓你幾回

### 車載網路於異構邊緣雲中基於身分隱私之服務下載機制
> 英文題目：A Service Download Mechanism Based on Identity Privacy for VANET in Heterogeneous Edge-Cloud Environment

1. (Yin)
2. (古)
3. (Yen)
4. (Cen)
5. (槍熊)
6. (Riz)
7. (晁誌)
8. (祐綸)
9. (韋德)
10. (晨偉)
11. (宗琪)
12. (議員)
13. (日日)

* time
* Intro - Introduction :
* Motivation & Goals :
* Related Work :
* System :
* Experiment :
* Conclusions :
* Total :

* 老師

---

## Yin aka 戒酒了拉
### 基於深度學習之惡意舒服加密封包偵測
> 英文題目：A Deep Learning Approach for Malicious and Encrypted Packet Detection

1. (Yin)
2. (古)
3. (Yen)
4. (Cen)
5. (槍熊)
6. (Riz)
7. (晁誌)
8. (祐綸)
9. (韋德)
10. (晨偉)
11. (宗琪)
12. (議員)
13. (日日)

* time
* Intro - Introduction :
* Motivation & Goals :
* Related Work :
* System :
* Experiment :
* Conclusions :
* Total :

* 老師

---

## Cen aka yoyo球
### 5G網路中針對分散式舒服切片移動攻擊之防禦
> 英文題目: TODO

1. (Yin)
2. (古)
3. (Yen)
4. (Cen)
5. (槍熊)
6. (Riz)
7. (晁誌)
8. (祐綸)
9. (韋德)
10. (晨偉)
11. (宗琪)
12. (議員)
13. (日日)

* time
* Intro - Introduction :
* Motivation & Goals :
* Related Work :
* System :
* Experiment :
* Conclusions :
* Total :

* 老師
---

## Gun Bear （槍熊）

### 基於SDN網路之Slow HTTP服務阻斷攻擊偵測
> 英文題目: Slow HTTP DoS Attack Detection in SDN Network

1. (Yin)
2. (古)
3. (Yen)
4. (Cen)
5. (槍熊)
6. (Riz)
7. (晁誌)
8. (祐綸)
9. (韋德)
10. (晨偉)
11. (宗琪)
12. (議員)
13. (日日)

* time
* Intro - Introduction :
* Motivation & Goals :
* Related Work :
* System :
* Experiment :
* Conclusions :
* Total :

* 老師

---

## Riz
### In-Network traffic classification using programmable data plane

1. (Yin) Why do you want to split the layer to different switch? What's the advantage if you put your model in one switch?
2. (古) If the incoming packet is ipv6, how will the system handle
3. (Yen) How many types can BNN classify at most?
4. (Cen) if original ipv4 header has data, where do you store the bnn data?
5. (槍熊)
6. (Riz)
7. (晁誌) If the packets are unknown, how can you label it and retrain the model again
8. (祐綸) how do they define first layer sitch or other layer switch? and how do know first layer sitch or other layer switch? the distance from source to destination or another? In addition, different source and destination may have different path.
9. (韋德) How will you evaluate your system's overhead?
10. (晨偉) What if there p4 switch is not enough for hidden layer?
11. (宗琪) What does the output of each switch mean when you split hidden layers into many switches? Does the only last switch classify a packet? Does a packet have to run through every switches until the last switch?
12. (議員)
13. (日日)

* time
* Intro - Introduction : 6m5s
* Motivation & Goals : 1m57s
* Related Work : 4m16s
* System : 11m54s
* Experiment :
* Conclusions :
* Total : 24m14s

* 老師
* Co-advisor Lee
---

## Template

1. (Yin)
2. (古)
3. (Yen)
4. (Cen)
5. (槍熊)
6. (Riz)
7. (晁誌)
8. (祐綸)
9. (韋德)
10. (晨偉)
11. (宗琪)
12. (議員)
13. (日日)

* time
* Intro - Introduction :
* Motivation & Goals :
* Related Work :
* System :
* Experiment :
* Conclusions :
* Total :

* 老師

## template

* time
* Intro - Introduction :
* Background :
* Related Work :
* Motivation & Goals & comparison table:
* Workflow - System :
* Experiment Plan :
* Conclusions :
* Total :