Functional Requirements

Several organizations help businesses in using IoT for solving all industry-specific and long-standing issues. IoT solutions are developed by them that helps to collect data, connect things, along with deriving all insights with scalable as well as open solutions to increase revenue, improve productivity, and reduce costs. IoT helps to improve the digitalization of the society along with the economy with the connection of objects, all people with one another through communication and connected medium (Helmi et al. 2021). A client approaches IoT Innovative for designing an IoT system for a cashier-less store within the UK. This paper will provide the functional requirements, challenges associated with such a system, and overall security as well as physical architecture of this IoT system for the client.

This cashier-less process has evolved over the last decade, along with there are several variations upon the theme. There is the use of portable scanner models, and the scanner devices within the stores are connected with the account of the shopper. Such devices could be used by the customers for scanning all products as they include them in the cart along with paying for those products while leaving. Self-checkout has been the staple of the stores for some time now. The customers stand in a line, scan as well as bag their products, and make payment at the counter (Demir et al. 2019). Merchant apps could be used by the shoppers on their mobile devices for scanning all products and paying using the credit cards or their devices at the dedicated aisle around the exit. Functional requirements of the IoT infrastructure remain the same regardless of the technologies used by the users for every function. There are several functional requirements of this IoT infrastructure, such as:

• Diverse Connectivity: Connectivity’s most familiar form for IoT and the internet is Ethernet. Additionally to Ethernet, all IoT devices could connect with the use of a broad variety of all other technologies. An objective of the connectivity is that such IoT infrastructure supports several connection modes, both wireless and wired, that includes RFID, Z-Wave, ZigBee, LTE, EDGE, NFC, and GPRS (Pathak 2020).
• Leverage applications: The applications of IoT infrastructure are rapidly emerging for all businesses within virtually all industries, along with home users. Such applications provide several capabilities of automation for making this IoT solution much valuable. Such middleware and software applications help all businesses to reduce costs, improve all regulatory compliances, along with increasing efficiency. For achieving such goals, this IoT infrastructure needs to be compatible with all applications specific to those industries.
• Manage many devices: Information is collected by IoT sensors about the conditions in the vicinity, like moisture level and temperature. Specific tasks performed by IoT actuators, like turning things off or on, along with recording crucial information about the triggers as well as subsequent reactions (Denuwara, Maijala and Hakovirta 2021). This IoT infrastructure’s functional requirement is that this possesses the ability in managing a range of different devices.
• Generate a huge volume of data: The devices perform the tasks along with a report on all tasks performed by them. With the connection to the IoT infrastructure as well as to one another, these devices would transmit detailed information about all actions. Hence, there would not be any need for any human intervention within the process. All devices would send data for analysis along with storage in real-time (Lee and Lee 2018). Hence, this IoT infrastructure would be able in supporting storing a huge volume of data.
• Powerful Analytics: Huge volume of data possess the potential in providing unprecedented insights into consumer preferences as well as behaviours. Unlocking such insights needs tools of powerful analytics. The major functionality of this IoT infrastructure is capable of incorporating along with offering compatibility with the analytics solutions, which would translate a huge volume of data into actionable as well as useful insights.

IoT is connecting many more devices daily, and it would change the overall way daily tasks are being carried out by people. However, with all such benefits, there are several risks of using IoT devices. There are several privacy issues of IoT infrastructure that should be addressed properly.

• Huge amount of data: A huge volume of data which IoT devices could generate is quite staggering. There exist numerous data points daily generated from the households, and such creates several more entry points for the hackers along with leaves sensitive data vulnerable (Alhassoun, Uddin and Venkatasubramanian 2020).
• Eavesdropping: The hackers or manufacturers could use the connected devices for virtually invading the home of any individual. It is accomplished by intercepting a few unencrypted data from the smart meter devices for determining the activities.
• Unwanted public profile: Collected data is used by the organizations that the users willingly offer for making employment decisions. The data gathered from all IoT devices come from sensors that include thermometers, accelerometers, and microphones. Such granularity allows in the creation of additional information through the inferences of machine learning along with other techniques of analysis that could generate results which wouldn’t be possible with all coarser data (Altuk and Kablan 2020). The case must be given to the purposes where data would be used when this is collected from the individuals having no choice.
• Consumer confidence: All the issues could put a dent in the desire of the consumer for purchasing all connected products that could prevent IoT infrastructure to fulfil all true potentials.

For all reasons to make IoT infrastructures feasible for broader adoption, this is crucial in making this low cost along with increasing the overall supported devices. There are many security issues of IoT infrastructure that should be addressed before achieving the objective of broader IoT adoption.

• Authentication: It is a huge issue due to the overall quantity of devices. Authenticating all devices isn’t an easy task to do. Due to all features of quick computation along with energy efficiency, depending upon private key cryptographic primitive, a security mechanism could be used (Abbu, Fleischmann and Gopalakrishna 2021).
• Identification: Identification is crucial if this is original or any malicious code. There is need of few references to be available. Device identification could be performed by considering their hardware or physical addresses by deploying through IPv6, as all identifications are crucial for proceeding towards data management and authentication.
• Data management: Identifying numerous devices along with their addressing could be considered a huge issue within IoT. Managing such devices along with their addressing would be tough for IPv6 also (Gauri et al. 2021). There exist several methods which could be used to identify all objects within IoT. A few of these are the identification of bar codes. There is also the use of object identification based on vision, such as NFC and RFID technologies, to scan purposes.
• Heterogeneity: One of the biggest issues is this device heterogeneity issue. The issues should be properly tackled for making IoT more reliable as well as secure. All objects should be differently tackled, that makes this tough for applying for a single solution. This would be quite a difficult task for securing all kinds of devices from all kinds of attacks. All devices’ needs should be fulfilled for keeping them functioning. Device heterogeneity could affect several other aspects like difficulty in identification, privacy, and integration. For removing the heterogeneity issue, there could be the use of IDRA architecture that is designed for integrating all devices.

            The growth of IoT is rapidly picking up the pace. Most of such connectivity would be concentrated within the cities, where smart devices along with sensors would be scattered over stores. Such increased connectivity would transform the process of cashier-less stores for the better in several ways which are not imaginable now (Raj, Raman and Subramanian 2022). However, few associated along with unique challenges with this IoT infrastructure should be addressed properly.

• Interference: Almost every IoT device depend upon wireless technologies for connectivity as well as communication. Radio frequency (RF) is used by all current wireless technologies as a transport medium. RF’s issue is that this is intended for the communications of far-field; however, this IoT ecosystem would be a span of a few feet only. It would cause huge issues of interference when several devices are connected along with to begin filling up all frequency spaces (Sukhani, Qomariyah and Purwita 2022). There is a need for new solutions which would allow IoT networks of short-range for coexisting at some feet of one another without interfering with one another or causing collisions.
• Device management: This IoT ecosystem’s management becomes a nightmare when there is an increase in the overall number of devices. It is both an issue of security administration as well as functionality. Considering this security side issue, controlling updates along with credentials for a large number of devices become problematic. This is easy in managing security for a smaller number of connected devices. However, in the case of a huge number of devices, managing updates along with maintaining passwords for all individual devices could become quite a difficult task. It could be solved with the use of a centralized system for security administration. This system has an access portal which helps the users in viewing all devices within this IoT network, keeping track of along with allowing all updates for all individual devices and using password manager systems for controlling credentials for every device. A centralized system is also needed for the issues of functionality (Shankar et al. 2021). Such a system could provide the hub for every device in being recognized along with their functionalities controlled as well as linked to one another.
• Energy consumption: Consumption of energy is another huge issue with this IoT infrastructure. There would be more wireless transmissions if more devices were connected to this IoT network, that would be quite energy-consuming. There exist several technologies using techniques of low energy transmission; however, they aren’t complete along with, are not secure and quite slow. There is a need for new technologies which would be quite low on the consumption of energy. It would provide also acceptable data rates as well as be deployable within different devices. IoT devices could be deployed within unattended environments as well as would depend upon the energy of the battery to work, where these would need to continue in working for much longer periods without any hitches.
• Network traffic: With numerous devices sharing this same IoT network, improving efficiency and controlling traffic would require smart management of transmission of data along with the allocation of bandwidth between device groups and devices. It could be quite vital within smart cities, where activity and congestion within different areas would vary with time. The smart systems using machine learning could help learn the activity patterns along with allocating resources intelligently at several different times depending on this network’s changing needs (Shah 2018). Having this system could help in reducing costs of equipment and make the most effective use of the infrastructure of the network without uselessly scaling up.

Technical Security and Privacy Requirements

The reactive system’ principles define state-of-the-art technology within this IoT infrastructure. As IoT device is sensing along with actuating the physical systems, a few of them are crucial architectures. Communication is done for IoT devices over diverse networks, along with a few devices changing location. IoT is the broad spread strategy, along with it is needed everywhere for fully-fledged deployment (Abbu, Fleischmann and Gopalakrishna 2021). Hence, such technology needs an effective and widely accepted medium for all operations. An effective medium to deploy this IoT infrastructure is wireless medium. RFID is the technology that could be used for transmitting data using radio frequencies. The process would be performed with the use of RFID tags which needs to be deployed at all required spots. Such tags make communication to all RFID readers. RFID technology would offer great potency to this IoT environment. RFID tags’ miniature size would help its deployment within any area (Caruso 2020). RFID’s benchmark innovation is the protocols of anti-collision, which are serving as the boon for avoiding any errors. RFID would help to simplify several activities for this IoT infrastructure.

IPv6 is the communication protocol that could be used for this IoT infrastructure. It encapsulates all IPv6 long headers within small packets of IEEE 802.15.4 efficiently. It supports several low bandwidths, length address, power consumption, mobility, different topologies, scalable networks and unreliability. It could be connected to other IP networks directly without any intermediate entity like proxies or translation gateways. 6LoWPAN communication is based upon IPv6 that allows low-power closed wireless networks in interfacing with the internet, the global network, along with implementing more advanced services of intelligence (Pallitto 2018). This control system of this IoT infrastructure consolidates all information for recognizing all contexts of the real world, giving proper optimized feedback to this environment with the use of actuators. It allows the nodes in directly connecting with the internet with the use of open standards. With several wireless networks with low power designed for tackling certain issues, it is crucial that this system has the defined area that it addresses.

Crucial to secure IoT infrastructure is to understand this architecture along with its components. It should be ensured that appropriate measures are taken for bolstering this IoT network. There are several crucial components included in this IoT architecture.

• Actuators and sensors: The actuators are the devices acting as robotic controls. Actuators have been used by several attacks of IoT, like printers, as the launching point into the network of the business. This is quite imperative in ensuring that access of the device is controlled through settable passwords, monitor along with containing all executable codes run by such device, and encrypting all data kept locally (Yildiz 2019).
• Actuator and sensor network: It is local networking connecting all devices to analytics’ first layer. This is analogous to the LAN within the use cases. The network to be used is wireless. The traffic across the network, along with its network components like access points, should be protected. Close attention should be provided to the network access control. It needs emphasis upon account management. It should be ensured that there exist only some individuals having the ability in configuring all network components (Dominguez 2019). The traffic needs to be encrypted for avoiding all attacks.
• Bus: Most initiatives don’t consolidate several traffic streams in the IoT bus. It should be expected in evolving as the complexities to manage along with scale multiple initiatives of IoT is realized.
• Analytics platform: Data of IoT is totally useless unless this is analyzed as well as acted upon. Several initiatives of IoT consist of one analytics platform. Such an analytics platform is based on cloud computing. For securing the platform, multifactor authentication should be established for access, all anomalous behaviour should be monitored, and encryption should be enabled where possible (Von Briel 2018).
• Analytics network: Analytics would be run on-premises. However, for making this IoT infrastructure, the platform should be connected along with security measures should be implemented (Srivastava, Kishore and Dhingra 2021). Considering how the multi-cloud security could have the same distinct issues, it might need special focus upon IoT applications.

Figure: IoT System Architecture (Source: Author)

End-to-End (E2E) solution will be used in this IoT architecture for addressing the privacy of the users. All security requirements are balanced by the privacy architecture with the privacy requirements. PETs are integrated by such design following an easy configuration manner into this IoT architecture. This architecture protects the overall privacy as close to that source. Such an approach prevents violation of privacy within consecutive domains along with might reduce the signalling overhead (Mason and WRLC 2020). In this device domain, all data could be dropped as well as aggregated. Aggregation involves the fusion of data of several different sensors along with time aggregates.

Unique and Associated Challenges

Dropping involves the removal of all unnecessary metadata. Such operations could be performed by this IoT gateway. In this connection domain, all data sources could be anonymized. In this cloud domain, all stored data could be used for analyzing new data’s influence. As authentication could clash with the privacy requirements, authenticity could be included through the IdP. Authorization could be decoupled from this IdP through pseudonyms that are issued by Trusted Third Party (Barile, Polese and Sarno 2018). Every PET is controlled through the policies. The privacy boundaries along with default settings would be specified by the service providers. The desired protection levels of privacy would be selected by the customers in such boundaries within this device domain.

Conclusion

            IoT Innovative has provided the design of IoT infrastructure that would provide for cashier-less stores within the UK. The system would help the customers in picking up their shopping along with walking out of that store where payments, as well as shelf stock, are handled automatically through this aid of this IoT system. This IoT system will provide reasonable assurances that privacy, as well as security considerations, are inherent within this design. This paper provides the functional requirements, privacy and security requirements, all associated challenges, proposed and state-of-the-art technologies, and overall privacy as well as security architecture of this IoT infrastructure. 

References

Abbu, H., Fleischmann, D. and Gopalakrishna, P., 2021, June. The Case of Digital Transformation in Grocery Business: A Conceptual Model of Digital Grocery Ecosystem. In 2021 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC) (pp. 1-9). IEEE.

Abbu, H.R., Fleischmann, D. and Gopalakrishna, P., 2021, March. The Digital Transformation of the Grocery Business-Driven by Consumers, Powered by Technology, and Accelerated by the COVID-19 Pandemic. In World Conference on Information Systems and Technologies (pp. 329-339). Springer, Cham.

Alhassoun, N., Uddin, M.Y.S. and Venkatasubramanian, N., 2020, September. Multi-Network Provisioning for Perpetual Operations in IoT-Enabled Smart Spaces. In 2020 IEEE International Conference on Smart Computing (SMARTCOMP) (pp. 89-97). IEEE.

Altuk, V.E. and Kablan, A., 2020. A critical approach to accounting in the digital era: assessment of the impacts of industry 4.0 on financials. In Digital Business Strategies in Blockchain Ecosystems (pp. 439-456). Springer, Cham.

Barile, S., Polese, F. and Sarno, D., 2018. Grocery retailing in the I4. 0 era. Symphonya, (2), pp.38-51.

Caruso, C.E., 2020. How AI can revolutionize modern industries: an analysis of the strategic and ethical implications in business and healthcare.

Demir, K.A., Turan, B., Onel, T., Ekin, T. and Demir, S., 2019. Ambient intelligence in business environments and internet of things transformation guidelines. In Guide to Ambient Intelligence in the IoT Environment (pp. 39-67). Springer, Cham.

Denuwara, N., Maijala, J. and Hakovirta, M., 2021. The impact of unmanned stores’ business models on sustainability. SN Business & Economics, 1(10), pp.1-27.

Dominguez, C.R., 2019. Collaborative Object Identification in Multi-Modal IoT Systems (Doctoral dissertation, Carnegie Mellon University).

Gauri, D.K., Jindal, R.P., Ratchford, B., Fox, E., Bhatnagar, A., Pandey, A., Navallo, J.R., Fogarty, J., Carr, S. and Howerton, E., 2021. Evolution of retail formats: Past, present, and future. Journal of Retailing, 97(1), pp.42-61.

Helmi, R.A.A., Hishamuddin, M.N.B., Jamal, A., Johar, MGM and Sim, LF, 2021, April. Quantum Application: A Smart Cashier-less Shopping Application with Biometric Payment Authentication. In 2021 IEEE 11th IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE) (pp. 157-162). IEEE.

Lee, S.H. and Lee, D.W., 2018. Changes in distribution industry using information and communication technology. Journal Of Advanced Information Technology And Convergence, 8(1), pp.23-30.

Mason, R.B. and WRLC, C., 2020. Project 2019/28: Status and Challenges of Automation in the South African Retail Sector, Including Automated, Self-Service Checkouts.

Pallitto, R.M., 2018. Irresistible bargains: Navigating the surveillance society. First Monday.

Pathak, B.K., 2020. INTERNET OF THINGS ENABLED ELECTRONIC MARKETS: TRANSPARENT, COLLABORATIVE, AND AUTOMATED. Issues in Information Systems, 21(4).

Raj, P., Raman, A.C. and Subramanian, H., 2022. Cognitive Internet of Things: Enabling Technologies, Platforms, and Use Cases. CRC Press.

Shah, S., 2018. Product marketing in the era of Internet of Things (Doctoral dissertation, Massachusetts Institute of Technology).

Shankar, V., Kalyanam, K., Setia, P., Golmohammadi, A., Tirunillai, S., Douglass, T., Hennessey, J., Bull, J.S. and Waddoups, R., 2021. How technology is changing retail. Journal of Retailing, 97(1), pp.13-27.

Srivastava, V., Kishore, S. and Dhingra, D., 2021. Technology and the Future of Customer Experience. In Crafting Customer Experience Strategy. Emerald Publishing Limited.

Sukhani, K., Qomariyah, N.N. and Purwita, A.A., 2022, February. CleverCart: An Eco-Friendly Shopping Self-Checkout System with IoT sensor. In IOP Conference Series: Earth and Environmental Science (Vol. 998, No. 1, p. 012041). IOP Publishing.

Von Briel, F., 2018. The future of omnichannel retail: A four-stage Delphi study. Technological Forecasting and Social Change, 132, pp.217-229.

Yildiz, S., 2019. Artificial intelligence based e-commerce types and consumer behaviours (Master’s thesis, Sosyal Bilimler Enstitüsü).