Getty Images
Using blockchain to improve security for IoT devices
Network and device security and data integration make protecting IoT deployments difficult. Organizations can achieve better security with random neural networks and blockchain.
Large-scale IoT deployments are currently within our grasp technologically in the form of smart cities.
Smart cities have promised to make our lives more efficient. With connected devices and information controlled by an intelligent system, cities will never have traffic jams again. Connected technology can also make cities energy efficient with different sensors identifying waste. Despite the tech industry talking about it for a decade and a half, it has not come to fruition.
Most deployments focus solely on the distribution of hardware without considering the needs of the hardware. Because of this, deployments have failed or been long forgotten.
In this article, we will discuss a sample implementation of the infrastructures and the issues of large-scale connectivity and device security. One type of decentralized ledger, blockchain, can help solve the security problems IoT creates.
What are the challenges?
Three key problems plague IoT deployments.
The first is securing the network. In some IoT deployments, 5G networks provide the connectivity. These large networks present a very tempting target to hackers because of the volume of devices used on them. In addition, potentially sensitive data flow through the network which hackers can also target.
The second problem is data integration. How does data from different sources interact? Some of this data could be personal user data, such as a social insurance number; commercial data owned by an organization; individual usage habits while in a store; or public data, such as the current state of traffic lights.
All these pieces of data must interact on some level; however, this poses challenges. These include privacy, data ownership and data security. In addition, data ownership comes into question. For example, at a traffic intersection information about the cars there will be captured, but who owns that data? Is it the owners of the respective cars or the government that set up the sensors to collect the data?
The third and final problem is insecure IoT devices. This issue is important because a compromised device could also be used to attack other devices in the network. Organizations must monitor and regulate interactions between devices.
How to protect IoT deployments
One security measure to address IoT security integrates random neural networks with decentralized ledger technology to provide users with anonymous authentication and auditing capabilities.
First, a random neural network encrypts and decrypts data between two parties in a highly secure way. The neural network solves the problem of the network being hijacked. Even if a hacker were to hijack or eavesdrop on a network, the data will be useless because of the neural network encryption.
Second, blockchain fills in a large portion of the security requirement. It can be used as an auditing system to handle authentication, access control, block query, de-identification, encryption and key management. Blockchain can also prevent distributed denial-of-service attacks against IoT devices because it offers a distributed voting system for managing power and bandwidth requirements.
IoT devices are constrained when it comes to power and computational resources. Ideally, blockchain deployments should be integrated with the cloud to achieve optimal results. It also should address the lack of trust between IoT devices, remove single points of failure, and allow for the prediction of interzone movement via the use of IoT crypto tokens and Markov prediction models. The result of this implementation is anonymous authentication and auditing capabilities. Blockchain does most of the heavy lifting from an infrastructure perspective; however, that is by design.
As with any implementation, there are drawbacks. The main drawback of using blockchain for a major portion of this implementation is scalability in terms of computational and space requirements. When implementing this solution, consider the way blockchain serves the IoT devices.
Conclusion
In this article, we went over some implementation problems facing large-scale IoT deployments. These challenges included network security, data security and privacy. In a sample security implementation, we explored the combination of random neural networks with blockchain technology to produce a strong large-scale IoT infrastructure. The most important thing to remember is that no matter how promising a piece of technology is, it needs its supporting infrastructure to function optimally.
About the author
Neil Okikiolu is a computer scientist, roboticist and the founder of Simius, a consumer-focused IoT cybersecurity company.
