Research Summary: SoK: Applying Blockchain Technology in Industrial Internet of Things

@Larry_Bates Thanks for your comment.

In this paper, the author’s position on standardization is based on the notion that failure to conform to open standards will eventually result in incompatibility. However, one of the arguments is that on a fundamental level, standards will reduce overhead, mitigate risks, and reduce the complexity of networks. Especially within the industrial ecosystem, where there is an urgent need for solutions that “support diverse use cases” to enhance interoperability, fuel competitive advantage, and optimize performance, “open standards” could promote transparency, trustworthiness, and better security. Due to the interdependent nature of industrial organizations, which entail “multiple stakeholders”, public and private enterprises may be inclined to adopt generic standards rather than deploy case-specific solutions to overcome barriers to varied trade and partnership opportunities [1]. Moreover, the International Organization for Standardization emphasizes that “the era of proprietary solutions is over and replaced by the new era of shared economy.” Bearing in mind the shift toward Industry 4.0, standardization may “facilitate automation” essentially by simplifying industrial architectures, improving auditability, and easing the integration of “communication technologies.” That being said, though many efforts for standardization are underway, these developments are still in a nascent stage and it seems fair to assume that many of the ideas surrounding the IIoT-blockchain layer are indeed based on theoretical inferences at this point in time.

3 Likes

Hi @rlj , thank you for sharing this interesting and useful summary. Recently, the development of TinyML has been in place gradually, and it sounds like a key development field. I’d love to know how do you think about how this trend will affect the integration of IIoT and blockchain. Thanks!

Applying a blockchain database service layer shows potential for improving scalability and reducing complexity for conventional Industrial Internet of Things systems. As an abstraction of the underlying layers, the blockchain database layer can leverage cloud-based platforms, enabling the deployment of essential blockchain services without modifying existing infrastructures. While blockchain databases such as BigchainDB, ChainfyDB, Cassandra, and Modex claim to improve performance, it seems as though the appropriateness of the blockchain database layer depends on the parameters of specific applications and the scheme employed. In the case study you referenced, BlockchainDB allows for data sharing and replication to be defined by the application, whereas direct blockchain-IIoT integration requires all nodes to distribute and record all transactions. With that in mind, scalability may be improved by allowing for more efficient data management. On the other hand, by implementing blockchain-as-a-service, smart contracts can further automation in industrial environments. Overall, I would say that the overarching draw of adding a blockchain database service layer is the ease of use, fewer complexities, less investment, and user-defined blockchain features. Regardless of better scalability, incorporating the database service layer still raises the concern of standardization and independent technological developments, which ultimately could be harmful to the compatibility of the industrial landscape. Therefore, it might be more of an immediate fix for blockchain-IIoT integration rather than a long-term solution. I am interested to hear your thoughts on this. Thanks for the great question @Tony_Siu

2 Likes

I would also like to highlight the coffee shop gurus sayings to “never store data on the blockchain”. According to an article from forum.cardano.org, it clearly differentiates between a database and a blockchain. Essentially, the concept of a blockchain can be surmised as a kind of database but it was never designed to be a data storage in its inception. It may be better to think of blockchain not only as a simple decentralized database but to better utilize and develop blockchain according to its original concepts.

Regarding the issue with IoT data standardization along with “independent technology developments”, I do not think this to be an issue at all. With the subject of standardization, I would reckon that many conventional feature engineering techniques are already being deployed in commercial data science. There is even an entire profession dedicated to it called the “data engineer”. Unsupervised data preprocessing algorithms for generating N-dimensional numerical feature embeddings are widely used in sentiment analysis or in the underwriting for large insurance companies. I personally know and met researchers of research projects underway to further develop data oracles as individual preprocessing nodes called “knowledge graphs” that are currently being researched to automate the dispersed abstraction layer of IoT incumbent data! So I do not think a database requiring a single source of truth to be a short term fix nor do I think “independent technology developments” be any possible obstacle with the advent of newer and newer techniques and technologies.

Here’s a paper on unsupervised feature engineering techniques for automated sensor anomaly detection:

H. Y. Teh, K. I. -K. Wang and A. W. Kempa-Liehr, “Expect the Unexpected: Unsupervised Feature Selection for Automated Sensor Anomaly Detection,” in IEEE Sensors Journal, vol. 21, no. 16, pp. 18033-18046, 15 Aug.15, 2021, doi: 10.1109/JSEN.2021.3084970.

6 Likes

I just read an interesting World Economic Forum post that pushes back on the folk-hysteria surrounding blockchain energy consumption. Entitled “Why the debate about crypto’s energy consumption is flawed,” the post notes that crypto “consumes less energy globally than tumble driers or domestic refrigeration” yet provides enormous societal benefit that is rarely factored into discussions of the topic.

The post goes on to catalog a number of energy-consuming activities that most people would consider vital to modern society, ranging from aviation transport to air conditioning and domestic refrigeration. Bitcoin (which is the barometer used in this post) uses a tiny fraction energy by comparison to any of “socially essential” use cases.

In return, “crypto uses energy to provide an alternative, borderless and decentralized store of value,” thus providing “economic freedom to people in developing countries.” By WEF’s reckoning, the energy used by crypto is a fantastic bargain. And the post goes on to note that the still-young crypto industry is also making progress toward greater sustainability.

2 Likes

@rlj

You should know that there is no single response to this topic because it is dependent on the unique requirements of each industrial innovation. However, in order to be effective, IIoT platforms must be able to integrate with a wide range of architectures. Industrial control systems, enterprise resource planning systems, and other operational systems may be included. Furthermore, IIoT platforms must be capable of collecting and analyzing data from a range of sources, such as sensors, devices, and machines. They must also be capable of providing real-time view into industrial operations and enabling effective real-time control. Finally, in order to interoperate with a wide range of devices, IIoT platforms must offer a wide range of communication protocols and systems.

I hope this meets your expectations

1 Like