FIT5003 Software Security 3 (S2 2024)R

Java Python FIT5003 Software Security Assignment 3 (S2 2024)

Total Marks 100

Please Check Moodle for the Due Date

1 Overview

The learning objective of this assignment is for you to perform. penetration testing, thread modelling, and write ethics for hacking.  The lab setup employed in Lab10 (Penetration Testing) can be utilized for this assignment.

2 Submission

You need to submit a report (one single PDF file) to describe what you have done and what you have observed with screen shots whenever necessary.  Please follow the template of report wherever provided. Typeset your report into .pdf format (make sure it can be opened with Adobe Reader) and name it as the format: [Your Name]-[Student ID]-FIT5003-Assignment.pdf.  Please do not submit any extra files, all screenshots or code (if applicable) should be embedded in the report.

Late submission penalty: Late submissions incur a 5-point deduction per day.  For example, if you submit 2 days and  1 hour late, that incurs  15-point deduction.   Submissions more than 7 days late will receive a zero mark. If you require extension or special consideration, refer to special consideration form. Kindly note that no member of the teaching team is authorized to grant extensions or special considerations. Therefore, refrain from seeking assistance on this matter from any teaching team member. Please adhere to the guidelines provided in the link mentioned.

Zero tolerance on plagiarism: If you are found cheating, penalties will be applied,i.e., a zero grade for the unit.  University polices can be found at https://www.monash.edu/students/academic/ policies/academic-integrity

For each question (Q1, Q2, and Section 5), a Generative AI statement must be included to indicate the extent of its use.  This statement must specify whether Generative AI tools were employed in the response. If no AI tools were used, the statement should explicitly state that.  Ensure a total of three statements, one for each designated section.

3 Penetration Testing [50 Marks]

The learning objective of this part is to learn the process of conducting a standard penetration test and sub- sequently compose a formal report detailing the identified vulnerabilities. The examination will be executed on virtual machines deliberately designed to be vulnerable, publicly accessible for educational purposes. You may leverage walkthroughs created by other testers as reference material; however, direct replication of text or screenshots from these walkthroughs is strictly prohibited.  While utilizing a walkthrough for guid- ance is permitted, the report should be an original composition.  External resources, beyond the provided walkthrough, can be consulted and referenced appropriately. It is important to note that the penetration test report will be checked for plagiarism through Turnitin.

Download one of the below Virtual Machines (VMs) and perform. penetration test on it. The goal of the test is to make an attempt to compromise the VM.

HACKINOS: 1 (https://www.vulnhub.com/entry/hackinos-1,295/)

CENGBOX: 1 (https://www.vulnhub.com/entry/cengbox-1,475/)

BASIC PENTESTING: 1 (https://www.vulnhub.com/entry/basic-pentesting-1,216/)

DEATHNOTE: 1 (https://www.vulnhub.com/entry/deathnote-1,739/)

Q1 (50 marks): Identify at-least 3 vulnerabilities in the selected Virtual Machine and write a report. The report should be in the following format:

Executive Summary (Max 300 words) - (10 Marks)

{Briefly explain the penetration testing results, e.g.  was the goal achieved?  if yes, how?  you can also provide high-level recommendations here. }

Vulnerability List (Max 200 Words) - (4 Marks)

{Create  a table with columns:   Vulnerability Name,  Severity  and  Page No.} (Utilize CVSS3.0 calculator for calculating the severity of the issue)

Details of Vulnerabilities

Chosen   three   vulnerabilities    should   be   written    in   the    following   format   -    (36   Marks)

{Severity} (e.g. High)

{Vulnerability Name e.g. SQL Injection}

Vulnerability

{Describe the vulnerability, exploit it and write step by step guide on how to re-produce the exploitation with screenshots} (Max 400 Words)

References

{add references here, for further reading, e.g. Heap Overflow}

Risk

{Explain risk here} (Max 200 Workds)

Recommendation

{Make theoratical recommendations here} (Max 200 Words)

4 Threat Modelling [30 Marks]

A pharmaceutical company has developed a system to diagnose an illness using a wearable device and machine learning (ML) models. Diagnosis tests are performed by clinicians using a mobile application and the patients are asked to do certain activities while wearing the devices.  The motions captured from the wearable devices are sent to a mobile app via Bluetooth and then sent to a cloud API for processing over internet.

The cloud API collect the data and process them using ML models. The result reports processed by ML models are saved in a database in the cloud. The clinicians can pull the reports from the cloud API and view them using the same mobile app.

Q2 (30 Marks): To complete thread modelling of above scenario, perform. the following:

•  Draw a DFD (it can be second or a third level DFD) for the above system and identify the trust boundaries. (10 Marks)

•  Identify at-least 3 threats, including an Information Disclosure threat, and suggest mitigation strategies for it. (Max 500 Words) (12 Marks)

• Add the mitigation strategy to the DFD. (8 Marks)

5 Ethics in Hacking [10 Marks]

Developing an Ethical Hacking Policy is essential. Your task is to communicate guidelines to ethical hackers in your company (fictitious) regarding appropriate hacking conduct, prohibited activities, and behaviors classified as unethical.  List a minimum of five policy directives.  Kindly ensure your response falls within the 150 to 500 word limit.

6 Report Completion and Quality of Presentation [10 Marks]

The remaining 10 marks are allocated to the quality and clarity of the report         

内容概要:本文介绍了基于贝叶斯优化的CNN-LSTM混合神经网络在时间序列预测中的应用,并提供了完整的Matlab代码实现。该模型结合了卷积神经网络(CNN)在特征提取方面的优势与长短期记忆网络(LSTM)在处理时序依赖问题上的强大能力,形成一种高效的混合预测架构。通过贝叶斯优化算法自动调参,提升了模型的预测精度与泛化能力,适用于风电、光伏、负荷、交通流等多种复杂非线性系统的预测任务。文中还展示了模型训练流程、参数优化机制及实际预测效果分析,突出其在科研与工程应用中的实用性。; 适合人群:具备一定机器学习基基于贝叶斯优化CNN-LSTM混合神经网络预测(Matlab代码实现)础和Matlab编程经验的高校研究生、科研人员及从事预测建模的工程技术人员,尤其适合关注深度学习与智能优化算法结合应用的研究者。; 使用场景及目标:①解决各类时间序列预测问题,如能源出力预测、电力负荷预测、环境数据预测等;②学习如何将CNN-LSTM模型与贝叶斯优化相结合,提升模型性能;③掌握Matlab环境下深度学习模型搭建与超参数自动优化的技术路线。; 阅读建议:建议读者结合提供的Matlab代码进行实践操作,重点关注贝叶斯优化模块与混合神经网络结构的设计逻辑,通过调整数据集和参数加深对模型工作机制的理解,同时可将其框架迁移至其他预测场景中验证效果。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值