Deep learning automatically identifies features which are more important for classification. It is an enhanced form of machine learning and generally gives an output faster than machine learning algorithms when a sufficiently large amount of data is considered. An industrial system comprises inputs, processes and outputs. Inputs generally consist of raw materials, land, power, transport, cost of infrastructure and other such things considered to be investments. The process is the transformation that the raw materials undergo to be labelled as the finished product. The outputs are the finished product and the profit the producer receives from the finished product. In the current scenario, the basics are the same but the classification is changed slightly. The current scenario also includes the dynamic relation between the demand and supply, along with various factors such as review, alliances between industries and other such dynamic parameters which change randomly.
Implementing deep learning in industrial systems requires an understand of the dynamics of current industrial behaviour, and designing an automation system with respect to the same knowledge. Deep learning networks can identify the dynamics of industry if large enough data sets are provided. This means all the factories will be completely automated, and unlike the current scenario companies will suffer minimum losses on risk factor. These networks might also be able to identify potential mutually beneficial relationships amongst companies which in turn will lead to economic gain for both parties. Deep learning networks will also be able to perform risk analysis, thereby increasing the security of the companies. Another factor which leads to enhanced security and privacy is that there will be no information leakage because no human will be aware as to why a certain decision has been made and when it will change.
This special issue mainly focuses on deep learning models for security and privacy in industrial systems, addressing both original algorithmic development and new applications. We are soliciting original contributions from leading researchers and practitioners in academia as well as industry, which address a wide range of theoretical and application issues in this domain.
Topics of interest
Topics relevant to this special issue include but are not limited to:
Threats on security and privacy during transmission of data
Automation mechanism implementing deep learning
Security mechanisms in industrial systems
Sustainable computing design and analysis for IoT environments
Nature-inspired smart efficient energy hybrid systems for IoT networks
Energy-efficient networking for smart environments
Data security and privacy for Cyber physical systems
Energy efficiency in the Internet of Things
Business models and processes for the Internet of Things
Device-to-Device (D2D) communications in industrial systems
Learning for anomaly and intrusion detection
Autonomous vehicle, and wider transport system, security and privacy
Learning for critical infrastructure security
Authentication and access control for Cyber physical systems
Embedded systems security and privacy
Security and privacy in wireless sensor networks
Security and privacy in industrial control systems
Smart grid security
The relation of security and safety for smart industrial systems
Secure communications in sensor networks and the IoT
Trustworthiness management models for sensor networks and the IoT
Cyber security of industrial cyber-physical systems
Papers must be tailored to the emerging fields of deep learning models for security and privacy in industrial systems, through deployments models, challenges and novel solutions. The editors maintain the right to reject papers they deem to be out of scope of this special issue. Only originally, unpublished contributions and invited articles will be considered for the issue. The papers should be formatted according to the journal guidelines.