As an MSc research student, you are expected to understand how other areas of computer science relate to your chosen specialism. You must have a grasp of significant current open problems in other specialisms, as well as an understanding of how techniques and methods from other research areas can be applied within your specialism.
In this assignment you must provide an overview of two MSc Research specialism lectures (AI / Networking / Cyber Security / Software Engineering / Data Science) which are NOT the specialism you discussed in Assignment 1.
For each of these two areas, you must discuss the research question presented within the relevant specialist lecture for this specialism, propose some research approaches to investigate this question and identify further work you might undertake which builds on this. You must also explore your personal strengths in this area.
Part 1: Robotics and AI
Robotics is an important interdisciplinary branch of the advanced computer science that helps in reducing the efforts of humans by doing their tasks automatically. AI is termed as the intelligence of computer science that is incorporated in robotics systems to improve the accuracy of work. However, management of human robots interaction (HRI) is essential to continue the superior functionality of robots by eliminating ethical issues associated with robots in safety performances. Social robot (SR) is designed technically with the help of AI to interact with humans. SR is very effective in eliminating or handling social hazards without direct involvement of humans. The presence of social responsibilities such as gesture movements, head orientation, and gazed directions are essential to produce SR and maintain its social effectiveness (Ghazali et al. 2018). Details of credibility of SR in safety performances in support of other works and approaches that can fit with research will be provided here.
- Is the robot acting like a functional social being would?
Combination of AI with robotics is an important invention of computer science that can be utilised for different tasks. Robots use AI algorithms using which they can easily learn and adapt to the surrounding situation. This learning of SR is essential as using them different safety performance can easily be done with a little involvement of humans. SR is accurately designed for interaction, maintaining collaboration and working with humans in interpersonal and different socio-affective levels (Piçarra & Giger, 2018). Interest in arranging this specific question is that it has strength in discussing the role of AI in handling commands of SR and maintaining functionality for different societal safety performances. SR provides benefits in doing different social activities as well as guiding children during their learning. Effective HRI and design of SR provides a strong influence in continuation of varieties safety functionalities within a society (Holthaus et al. 2019). SR works in the form of collaboration with humans and assists in an autonomous process in different important tasks.
The question here is maintained based on the credibility issues of SR during services to any safety performance especially due to failure of proper HRI algorithms. SR is not designed as replication or an alternative to human efforts as they have huge issues with functionalities sometimes (Horstmann & Krämer, 2019). Research is going to determine domain-specific behaviour strategies from SR through which accuracy in functioning during safety performances can be possible. According to Banks (2018), proper management of coding in AI and its implementation in robotics plays a positive role in healthcare, the hospitality sector and even in performance improvement in social works. Therefore, a proper AI algorithm and its use in SR manufacturing can be answered with the support of this question.
The issue of AI algorithms during SR manufacturing creates problems in its use in different safety performances.
SR is an autonomous machine that has enough influence in maintaining social works including safety performances using AI. Identification of different diverse attitudes of people is essential for enhancing the persuasiveness of any SR and its functionalities (Ghazali et al. 2018). SR are highly embodied agents for maintaining autonomous performances by maintaining a proper HRI. In contrast, Peter & Kühne (2018) have said that the communication medium needs to be maintained properly between SR and humans whether that communication is verbal or even non-verbal. Therefore, it can be assured that communication is important between SR and user to improve functionality of SR in various essential safety performances. The presence of different advanced automatic sensors and the advancement of AI features have supported the work of robots and humans together safely in different fields such as in healthcare or even in military service (Bankins & Formosa, 2020).
Figure 1: User adaptiveness model
(Source: Martins et al. 2019)
Digital training to SR is important as using that digital knowledge SR can work for safety performances after using an adaptive strategy mentioned in the above image. AI is useful to modify the coding and algorithms using which advanced performances from an SR can be achievable. AI and its use in robotics are successful due to the presence of a neutralisation strategy in advanced technologies as AI can easily perform based on its knowledge and other value systems (Froese & Taguchi, 2019). The presence of AI in advanced SR helps in the reduction of reaction time clarifies a complex task easily and can function 24*7 times without any serious interruption. In contrast, Wang & Siau (2019) have opined that advancement in robotics, different advanced machine learning, and AI can influence jobs of humans at a high level that is a serious issue for using robots. Therefore, it can be considered with the fact that SR is effective for performing tasks however; it has a negative influence on job management.
The literature that is selected here is good for determining the influence of AI in SR however; they are lacking information on the actual usage strategy of AI in robotics. Further, possible issues with SR and accurate mitigation strategies are still not clearly mentioned in the literature. An essential open problem is lack of details regarding accurate functionality of SR using AI causes issue gaining knowledge. Another open problem is absence of issue mitigation strategy of SR for maintaining HRI in safety performances.
As per the concepts mentioned in the research paper, it can be determined that the importance of HRI and the use of SR in safety management are explained here. Primary research has been conducted based on this particular research paper through which responses of respondents have been taken for determination of safety hazards associated with SR. Total thirty participants have been chosen based on experience of issues as respondents for the conduction of data collection (Holthaus et al. 2019). The major aspect for conduction of this specified research is to establish a prior link between credibility of SR in maintaining a large range of safety performance to eliminate social hazards. SR is determined as a good companion for elderly adults that offers effective utilities along with companionship (Breazeal et al. 2019). Two types of variables are granted here such as dependent and independent variables for analysing data accurately. Independent variable indicates two types of behaviour of SR such as per the various social norms (AN) and violation of different norms (VN). An experimental area is selected for SR functionalities as mentioned in this specific paper mentioned in the image below.
Figure 2: Selected area
(Source: Holthaus et al. 2019)
The advantage that is the main outcome of choosing this approach is gathering accurate and real-time authentic data that has helped to determine issues with SR. An appropriate initial connection of different social credibility with SR functionalities is maintained, which is another advantage for using such an approach. Violation issue of SR is easily identified using the responses of respondents. Further, identification of social hazards using this approach has added advantage here. Disadvantage includes lack of availability of in-depth knowledge due to the absence of secondary sources. Further, mitigation strategy of issues using such an approach is not clear here which is a critical disadvantage. The importance of alerts regarding safety during the use of SR is not accurate using such an approach that is a serious disadvantage. Another important disadvantage is the absence of statistical significance in analysis of such primary data accurately that vets barriers in this research.
An alternative approach that can be effective is the use of secondary data along with the use of primary data can be beneficial to maintain supremacy in result. Presence of a secondary source will be effective to compare the observed results from primary analysis to secure authenticity. Secondary research will be less expensive and data will be easily available that will save time along with maintaining authenticity.
The outcome from the question can be highly effective in analysis of exact use of AI for managing functionalities in SR. Knowledge about AI and SR can be helpful to generalise the strategy of functionality improvement to make SR eventful in safety performances. The advancement of AI has changed the behaviour of SR as human types that have physical presence along with presence of psychological functionalities (Giger et al. 2019). Questions that are generated will be helpful to answer algorithms of Ai that can be used for improving SR performance such as human and HRI. Identification of ethical problems and their solutions can easily be available from that question.
Ethical problems and their strategy of identification for SR use have been mentioned which is good enough to strengthen knowledge on AI and SR. This strength will be fit for applying knowledge in the arrangement of research in future especially on the specific topic.
Bankins, S., & Formosa, P. (2020) ‘When AI meets PC: Exploring the implications of workplace social robots and a human-robot psychological contract’. European Journal of Work and Organizational Psychology. 29 (2) pp.215-229. https://www.researchgate.net/profile/Sarah-Bankins/publication/333398306_When_AI_meets_PC_exploring_the_implications_of_workplace_social_robots_and_a_human-robot_psychological_contract/links/5d009e2e299bf13a384ea23a/When-AI-meets-PC-exploring-the-implications-of-workplace-social-robots-and-a-human-robot-psychological-contract.pdf
Banks, J. (2018) ‘The human touch: practical and ethical implications of putting AI and robotics to work for patients’. IEEE pulse. 9 (3) pp.15-18. https://ieeexplore.ieee.org/abstract/document/8358087/
Breazeal, C. L., Ostrowski, A. K., Singh, N., & Park, H. W. (2019) ‘Designing social robots for older adults’. Natl. Acad. Eng. Bridge. 49 pp.22-31. https://robots.media.mit.edu/wp-content/uploads/sites/7/2021/04/Breazeal_et_al_2019_Designing-social-robots-for-adults.pdf
Froese, T., & Taguchi, S. (2019) ‘The problem of meaning in AI and robotics: still with us after all these years’. Philosophies. 4 (2) pp.14. https://www.mdpi.com/2409-9287/4/2/14/pdf
Ghazali, A. S., Ham, J., Barakova, E., & Markopoulos, P. (2018) ‘The influence of social cues in persuasive social robots on psychological reactance and compliance’. Computers in Human Behavior. 87 pp.58-65. https://www.researchgate.net/profile/Emilia-Barakova/publication/325348471_The_influence_of_social_cues_in_persuasive_social_robots_on_psychological_reactance_and_compliance/links/5e87294f299bf1307975173d/The-influence-of-social-cues-in-persuasive-social-robots-on-psychological-reactance-and-compliance.pdf
Giger, J. C., Piçarra, N., Alves‐Oliveira, P., Oliveira, R., & Arriaga, P. (2019) ‘Humanization of robots: Is it really such a good idea?’. Human Behavior and Emerging Technologies. 1 (2) pp.111-123. https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/hbe2.147
Holthaus, P., Menon, C., & Amirabdollahian, F. (2019, November) ‘How a robot’s social credibility affects safety performance’. In International Conference on Social Robotics. (pp. 740-749). Springer, Cham. https://uhra.herts.ac.uk/bitstream/handle/2299/22060/safetycredibility.pdf?sequence=1&isAllowed=y
Horstmann, A. C., & Krämer, N. C. (2019) ‘Great expectations? Relation of previous experiences with social robots in real life or in the media and expectancies based on qualitative and quantitative assessment’. Frontiers in psychology. 10 pp.939. https://www.frontiersin.org/articles/10.3389/fpsyg.2019.00939/full
Martins, G. S., Santos, L., & Dias, J. (2019) ‘User-adaptive interaction in social robots: A survey focusing on non-physical interaction’. International Journal of Social Robotics. 11 (1) pp.185-205. https://ap.isr.uc.pt/wp-content/uploads/mrl_data/archive/112.pdf
Peter, J., & Kühne, R. (2018) ‘The new frontier in communication research: Why we should study social robots’. Media and Communication. 6 (3) pp.73-76. https://www.cogitatiopress.com/mediaandcommunication/article/viewFile/1596/1596
Piçarra, N., & Giger, J. C. (2018) ‘Predicting intention to work with social robots at anticipation stage: Assessing the role of behavioral desire and anticipated emotions’. Computers in Human Behavior. 86 pp.129-146. https://www.sciencedirect.com/science/article/pii/S0747563218301900
Wang, W., & Siau, K. (2019) ‘Artificial intelligence, machine learning, automation, robotics, future of work and future of humanity: A review and research agenda’. Journal of Database Management (JDM). 30 (1) pp.61-79. https://www.researchgate.net/profile/Keng-Siau-2/publication/333423274_Artificial_Intelligence_Machine_Learning_Automation_Robotics_Future_of_Work_and_Future_of_Humanity_A_Review_and_Research_Agenda/links/5cf48f4b92851c4dd0240f42/Artificial-Intelligence-Machine-Learning-Automation-Robotics-Future-of-Work-and-Future-of-Humanity-A-Review-and-Research-Agenda.pdf
Cyber security (CS) is a code-based technology generated on the advancement of computer science using which safety features can be installed for computer systems and networks. A probe request (PR) is a control frame that is sent to any Wi-Fi network for understanding the network information. These two are critical challenges in terms of maintaining modern-day network security as using this two; any attacker can steal confidential data about networks. WLAN has a unique MAC address that is not secure in maintaining the safety of any wireless network. Attackers use this gap to incorporate codes to steal data effectively and hamper systems or networks. Attackers use advanced techniques to remove the security fences in wireless networking. DoS is the favourable technique of attackers to replicate access point (AP) in wireless network (Srinivas et al. 2019). This study will discuss the issue of MAC spoofing by providing an open question along with an analysis of existing works on this security issue.
- What are the intelligent as well as realistic methods for recognition of rough probe request frames in WLAN?
PR is a technical threat that uses different data rates and uses 802.11 capabilities to bypass AP in any wireless connection. This threat is remote based and is not understandable without proper security analysis of systems and networks. According to Benson et al. (2019), the CS issue indicates strongly that industries are not immune to serious cyber-attacks that can spoil their reputation. CS can be different types such as the security of an application, attack of malware, DoS attack, SQL injection and so many other advanced techniques of computer science. This specific research question on CS and PR will enhance the knowledge about PR attack and possible types of solution that causes interest generation for this question. MAC layer that is present in Wi-Fi is based on the 802.11 protocol system and works in terms of requests from users and responses from systems (Ratnayake et al. 2011). The security policy of 802.11 system is mentioned in the below image that indicates different agreements for maintaining safety.
Figure 3: Security system of 802.11
(Source: Ratnayake et al. 2011)
Therefore, this question has a positive influence in answering actual strategy of the 802.11 wireless system in maintaining security of Wi-Fi and MAC addresses.
PR is based on PRFA types of strategy using which attackers target flaws in connectivity or MAC designing and access information of any specific wireless network. PR sends a comparatively high amount of requests to AP and forces AP to reply to that request (Rizzi et al. 2020). The presence of malware in systems and the use of pirated software enhance the chance of CS in networks. The use of STA and ANN are two important facts using which attacks can be determined in Wi-Fi.
Drawbacks in maintaining security in MAC addresses of Wi-Fi can cause cyber-attacks and other PR attacks that destabilise system performance.
WLAN, beacons, and different online response messages are comparatively less protected due to flaws security. Security in cloud storage facilities is important for which Wi-Fi security is essential as cloud details can be easily accessed using wireless technologies. Attacks of PR and CS risks are two major parts that target those wireless systems to stay connected with the network for gathering information. PR is a type of security breach that forces an AP to respond against any PR (Hong et al. 2018). The issue with randomised classification of MAC address causes a serious issue of security that initiates PR and CS risk as mentioned in image 4.
Figure 4: Classification of MAC address
(Source: Hong et al. 2018)
In contrast, Ohki & Otsuka (2017) have suggested that since PR captures MAC addresses using which it can easily detect location of a wireless network and receptive point for safe access. Therefore, it is important from the above discussion that a PR safety arrangement is necessary without which safety can be never maintained due to MAC address issues. According to Li et al. (2020), any non-encrypted PR can easily be captured using a scanning device that is termed as Wi-Fi sniffers (Image: 5).
Figure 5: Wi-Fi PR using 802.11
(Source: Li et al. 2020)
An issue of cyber-attack is a technique using which attackers can intrude into a network to destabilise performance. CS is a modern computer science tactic that safeguards hardware, software, and other connectivity-related infrastructure (Srinivas et al. 2019). CS is important to make information and devices private from other users and even from any attacker. In contrast, Sun et al. (2018) have suggested that CS is based on the installation of a tool using which unauthorised intrusion is done by attackers. Therefore, it is true that CS is looming threats that can influences a networking system and destroy security to spoil the reputation of any networking security system. Categories of CS in systems are mentioned in the below image that can help to identify actual issues of CS and PR.
Figure 6: CS categories
(Source: Sun et al. 2018)
Drawbacks of this used literature indicate minimal information on PR attack strategy and CS mitigation process. The first open problem is the lack of support for information related to PR and CS. Another open problem is lack of a solution strategy to mitigate the remote attack of PR and risk of CS.
4. Research approach
Strategy of sample frame analysis that is primary data is the selected strategy as mentioned in this selected provided paper for determining facts of PR and CS. Three types of frames are taken here such as management, effective control and data using which analysis has been done (Ratnayake et al. 2011). Variables are selected considering the WIDS for analysis of large data for determination of the influence of PR and CS in Wi-Fi. The respective analysis is based on a selection of 1000 samples of present frames with two names. The first name is based on the user frame that is named as 1 (genuine) and another is for the attacker namely 0 (non-genuine). Sequence numbering of users is used here for synchronising software using STA. The test is based on a request sent to a designated AP using an SSID that is normal in PR attacks (Wang et al. 2018). MATLAB technique is utilised here as an analysis framework for those 1000 frames using which authentic outcomes are achieved in this research. The utilisation of 802.11 system and WLAN technologies for testing the results are mentioned in this given paper. Synchronisation of software has helped to detect issues of PR and CS in the testing systems.
The advantage of determination of MAC spoof using data analysis has helped in maintaining accuracy. Advantage of using different variables with potential to determine the possibility of STA spoofing is a positive outcome (Ratnayake et al. 2011). Result authenticity is an advantage that has supported the success of selecting this primary approach. Availability of precession in the framing technique and effectiveness has strongly eliminated doubts in respective research outcomes. Utilisation of technology such as MATLAB for analysis and accurate determination of MAC spoofing in WLAN has added advantage for selecting this approach. However, the main disadvantage is an absence of comparison with secondary sources. Improper frame arrangement for issue detection is another disadvantage associated with this approach.
Collection of secondary data in association with such primary frame analysis can be an alternative to add flexibility in PR issue determination. Such a vast database can be able to determine the actual influence of PR and CR and their mitigation strategy. This strategy will be accurate, as authentic sources will be utilised only to detect PR systems. Improvement of knowledge using secondary sources along with primary ones will add an extra advantage in getting genuine outcomes.
The concept of PR and CS is essential using which management of future security in Wi-Fi and MAC addresses can be possible. Questions that contain terms such as PR and WLAN can be answered effectively using which gathering of knowledge about functionalities of WLAN and MAC can be identified. PR messages have advantages using which geographical features of respective users can be identified (Redondi & Cesana, 2018). Analysis of this paper supports gathering knowledge on WLAN functionality and possible security gaps in MAC through which attackers can spoof into a network. Detailed processes of PR attacks and CS elimination can be gathered after the solution of the research question.
Strengths include the availability of frame strategy using which effective PR issues can be separated. This strength will be fit for arranging research in future, as it will help to analyse data accurately. Further, the concept of PR and CS in WLAN can provide strength to implement concepts in a research analysis in future.
6. Reference list
Benson, V., McAlaney, J., & Frumkin, L. A. (2019) ‘Emerging threats for the human element and countermeasures in current cyber security landscape’. In Cyber Law, Privacy, and Security: Concepts, Methodologies, Tools, and Applications. (pp. 1264-1269). IGI Global. https://repository.uwl.ac.uk/id/eprint/4448/1/Human%20element%20Editorial%20Chapter_v3.pdf
Hong, H., De Silva, G. D., & Chan, M. C. (2018) ‘Crowdprobe: Non-invasive crowd monitoring with wi-fi probe’. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies. 2 (3) pp.1-23. https://www.researchgate.net/profile/Hande-Hong/publication/327757136_CrowdProbe_Non-invasive_Crowd_Monitoring_with_Wi-Fi_Probe/links/5c516e4c458515a4c749bfeb/CrowdProbe-Non-invasive-Crowd-Monitoring-with-Wi-Fi-Probe.pdf
Li, Y., Barthelemy, J., Sun, S., Perez, P., & Moran, B. (2020) ‘A Case Study of WiFi Sniffing Performance Evaluation’. IEEE Access. 8 pp.129224-129235. https://ieeexplore.ieee.org/iel7/6287639/8948470/09138409.pdf
Ohki, T., & Otsuka, A. (2017) ‘A Study on Autoencoder-based Reconstruction Method for Wi-Fi Location Data with Erasures’. In Proceedings of the 2017 on Multimedia Privacy and Security. (pp. 13-18). https://www.researchgate.net/profile/Akira-Otsuka/publication/320743700_A_Study_on_Autoencoder-based_Reconstruction_Method_for_Wi-Fi_Location_Data_with_Erasures/links/5a0083fa0f7e9b62a1500119/A-Study-on-Autoencoder-based-Reconstruction-Method-for-Wi-Fi-Location-Data-with-Erasures.pdf
Ratnayake, D. N., Kazemian, H. B., Yusuf, S. A., & Abdullah, A. B. (2011) ‘An intelligent approach to detect probe request attacks in IEEE 802.11 networks’. In Engineering Applications of Neural Networks. (pp. 372-381). Springer, Berlin, Heidelberg. https://link.springer.com/content/pdf/10.1007/978-3-642-23957-1_42.pdf
Redondi, A. E., & Cesana, M. (2018) ‘Building up knowledge through passive WiFi probes’. Computer Communications. 117 pp.1-12. https://core.ac.uk/download/pdf/154335220.pdf
Rizzi, A., Granato, G., & Baiocchi, A. (2020) ‘Frame-by-frame Wi-Fi attack detection algorithm with scalable and modular machine-learning design’. Applied Soft Computing. 91 pp.106188. https://www.sciencedirect.com/science/article/pii/S1568494620301289
Srinivas, J., Das, A. K., & Kumar, N. (2019) ‘Government regulations in cyber security: Framework, standards and recommendations’. Future Generation Computer Systems. 92 pp.178-188. https://www.researchgate.net/profile/Srinivas-Jangirala/publication/328183318_Government_regulations_in_cyber_security_Framework_standards_and_recommendations/links/5c1d53d892851c22a33d339e/Government-regulations-in-cyber-security-Framework-standards-and-recommendations.pdf
Sun, N., Zhang, J., Rimba, P., Gao, S., Zhang, L. Y., & Xiang, Y. (2018) ‘Data-driven cybersecurity incident prediction: A survey’. IEEE communications surveys & tutorials. 21 (2) pp.1744-1772. https://ieeexplore.ieee.org/abstract/document/8567980/
Wang, W., Chen, J., Hong, T., & Zhu, N. (2018) ‘Occupancy prediction through Markov based feedback recurrent neural network (M-FRNN) algorithm with WiFi probe technology’. Building and Environment. 138 pp.160-170. https://www.sciencedirect.com/science/article/am/pii/S0360132318302464