The Fog computing is the extension of cloud computing which helps in managing calculation and storage services with end devices and data centres. “The processing of the information takes place in local smart devices in the fog computing” (Yi, 2012). Intelligent transport system is used for managing real-time traffic over the network. The designing and implementation of fog computing is equipped with many challenges. The issues and problems are raised in the area of model analysis, in developing protocol design, allocation of workload balance, consumption and controlling of energy, controlling optimization and communication, testing of prototype, integration of the system, implementation of application, and others.
Purpose and Justification
“The fog devices are less trustworthy as they are deployed at the edge of the network” (Khalid, 2016). The privacy and security issues are associated with the implementation of fog computing. Fog Computing is the technology which is based on decentralized computing infrastructure for the distribution of computing resources and services at the edge of the network. The data is transferred from the data source to the cloud. The fog computing works on improving the efficiency and reducing amount of data transportation to cloud for analysing, processing, and storing. In this paper, we are going to focus security and privacy issues which are faced by the fog computing.
From the research it has been analysed that it is the new paradigm which provides the virtualized phase for the allocation of resources to manage storage, computation, and services at the customer end. “The communication between decentralized and heterogeneous devices is classified as fog computing” (Wen, 2014). The edge location is the advantage for managing central focus in fog computing.
Conceptual or theoretical Framework
Fog nodes are the smart devices which provides standard in virtualization. Fog computing is categorised as Software as a service (SaaS), Platform as a Service (PaaS), and infrastructure as a Service (IaaS). “In the deployment model fog can be categorised as public fog, private fog, hybrid cloud, and community fog” (Lu, 2017). The overview of the fog computing is generalized in the diagram below
Research and system development methods:
- Reliance, Verification, and Authentication:
In fog computing, unusual parties served as suppliers which result into diversification in deployment choices such as wireless carrier or internet service suppliers put the fog in the hand of infrastructure, suppliers of cloud services which plays an important role in the extension of cloud services at the edge of the network, and end customers can change local private cloud to fog by making use of auxiliary resources. “The existence of fake fog node raises the concern to security and privacy of customer data” (Allerin, 2016). It is hard to predict the fake nodes in fog computing. Another issue is to detect complex trust situation call. “Authentication and verification are the primary security concern of fog computing to provide trustworthy services massive strength of end customers” (Borkar, 2014).
Wireless network security is the major concern area for fog computing to manage remote framework of devices. Jamming and sniffing are the major attacks associated with the fog computing. The common problem is to detach network traffic management from general traffic management. The access of cloud server from end fog nodes should be properly managed. The software defined network helps in execution and management of resources associated with fog computing. Cloud watch, traffic isolation and prioritization, arrangement of access controller on SDN, updating of fog router, and others are some of the security issues which are associated with SDN.
- Protected and safe storage of data and its computation:
It is hard to analyse the correctness of the data which is outsourced over the fog node. Unapproved parties can exchange data in the scenario of fog network. To overcome this problems, auditable data storage services is used for setting data over the cloud or fog computing. Homo-morphic encryption technique is used for the verification of correctness of data to permit secure transmission of data among untrusted parties. Random mask method is used by the third party for authentication and verification of data.
Data Collection or system design methods:
Questionnaires, interviews, direct observation, and reporting are some of the methods which are used for conducting research on security issues related with fog computing.
- Confidentiality and privacy of data:
The confidentiality of data and instance should be deployed in relation with location, instance, zone, data, and others which are involved in fog computing. Privacy preserving methodologies have been used in the online social network and wireless network for maintaining privacy of the data. “Sensitive data is delivered to end devices or other contraption devices by sensor” (Li, 2015). Homomorphic encryption technique is used for allocating data to the end fog devices.
Access control is the major concern to ensure security and confidentiality of the data to the end user. “Standard access control is used for outsourcing the data on the cloud” (Kaur, 2014). Symmetric key management is the versatile method used to transmit data over the cloud. Attribute based encryption technology is a fine grained access control which helps secure transmission of data from data source to end devices. Policy based resource access control is the method used in fog computing to support interoperability between fog resources.
Intrusion detection method is used for reducing attacks on the virtual machines which are placed in the fog network. Flooding strike and port checking are the methods used for measuring smart grid system. “Sniffing activities can be identified in fog networking by using the technology of intrusion detection system” (Bharti, 2012).
Security, scalability, open, autonomy, programmability, Reliability, availability, serviceability, agility, and hierarchy are the ten pillars of the fog computing. Verifiable computing helps in the transmission of data over the untrusted parties which take participation in the communication.
Analysis of Data:
Fog system uses the technology of scrutinizing of log files and customer information for recognizing attacks and sniffing activities which can takes place over the fog network. Differential privacy is the mechanism used for non-disclosure of information. Privacy preserving algorithms are used to prevent prohibition of resources at the end devices. Fog devices uses usage pattern for utilizing the services of the fog. Location privacy prevents the disclosure of location of the fog client.
The fog computing is defined as the extension of cloud computing which supports networking of resources at the edge. The figure below is the proposed model for overcoming security issues in Fog computing:
It is hard to analyse the correctness of the data which is outsourced over the fog node. The encoding of the fragile data shall be done before the outsourcing of data from customers to fog nodes.
Fog Computing is the technology which is based on decentralized computing infrastructure for the distribution of computing resources and services at the edge of the network. Verifiable computing helps in the transmission of data over the untrusted parties which take participation in the communication. The existence of fake fog node raises the concern to security and privacy of customer data. Authentication and verification are the primary security concern of fog computing to provide trustworthy services to massive strength of end customers.
Yi, S. (2012). Security and privacy issues of fog computing: A survey(1st ed.). Retrieved from https://www.cs.wm.edu/~zhengrui/papers/wasa15-fog.pdf
Khalid, A. (2016). Privacy and security problems in fog computing (1st ed.). Retrieved from https://www.caeaccess.org/research/volume4/number6/fakeeh-2016-cae-652088.pdf
Lu, R. (2017). Special issues security and privacy challenges in Emerging fog technology (1st ed.). Retrieved from https://www.mdpi.com/journal/sensors/special_issues/SPCEFC
Wen, S. (2014). The fog computing Paradigm: Scenarios and security issues (1st ed.). Retrieved from https://pdfs.semanticscholar.org/4289/40fa3f81f8d415c26661de797a77d8af4d43.pdf
Allerin, J. (2016). What is fog computing? (1st ed.). Retrieved from https://www.allerin.com/blog/what-is-fog-computing
Borkar, D. (2014). Fog computing: A new concept to minimize the attacks and to provide security in cloud computing environment (1st ed.). Retrieved from https://esatjournals.net/ijret/2014v03/i09/IJRET20140309018.pdf
Bharti, M. (2012). Securing user data on cloud using fog computing and decoy technique (1st ed.). Retrieved from https://www.ijarcsms.com/docs/paper/volume2/issue10/V2I10-0022.pdf
Li, C. (2015). A survey of fog computing: Concepts, applications ,and issues (1st ed.). Retrieved from https://www.cs.wm.edu/~liqun/paper/mobidata15-fog.pdf
Kaur, M. (2014). Fog computing proving data security: A Review(1st ed.). Retrieved from https://www.ijarcsse.com/docs/papers/Volume_4/6_June2014/V4I6-0126.pdf.