Decentralizing IoT networks through blockchain

By Ben Dickson for Techcrunch

magine a washer that autonomously contacts suppliers and places orders when it’s low on detergent, performs self-service and maintenance, downloads new washing programs from outside sources, schedules its cycles to take advantage of electricity prices and negotiates with peer devices to optimize its environment; a connected car, smart enough to find and choose the best deal for parts and services; a manufacturing plant where the machinery knows when to order repairs for some of its parts without the need of human intervention.

All these scenarios — and many more — will be realized thanks to the Internet of Things (IoT). Already, many of the industries that historically didn’t fit well with computers have been transformed by the billions of IoT devices connected to the internet; other industries will follow suit as billions more enter the fray.

The possibilities are virtually countless, especially when the power of IoT is combined with that of other technologies, such as machine learning. But some major hurdles will surface as billions of smart devices will want to interact among themselves and with their owners. While these challenges cannot be met with the current models that are supporting IoT communications, tech firms and researchers are hoping to deal with them through blockchain, the technology that constitutes the backbone of the famous bitcoin.

The problem with the centralized model

Current IoT ecosystems rely on centralized, brokered communication models, otherwise known as the server/client paradigm. All devices are identified, authenticated and connected through cloud servers that sport huge processing and storage capacities. Connection between devices will have to exclusively go through the internet, even if they happen to be a few feet apart.

While this model has connected generic computing devices for decades, and will continue to support small-scale IoT networks as we see them today, it will not be able to respond to the growing needs of the huge IoT ecosystems of tomorrow.

Existing IoT solutions are expensive because of the high infrastructure and maintenance cost associated with centralized clouds, large server farms and networking equipment. The sheer amount of communications that will have to be handled when IoT devices grow to the tens of billions will increase those costs substantially.

Even if the unprecedented economical and engineering challenges are overcome, cloud servers will remain a bottleneck and point of failure that can disrupt the entire network. This is especially important as more critical tasks such as human health and life will become dependent on IoT.

Moreover, the diversity of ownership between devices and their supporting cloud infrastructure makes machine-to-machine (M2M) communications difficult. There’s no single platform that connects all devices and no guarantee that cloud services offered by different manufacturers are interoperable and compatible.

Decentralizing IoT networks

A decentralized approach to IoT networking would solve many of the questions above. Adopting a standardized peer-to-peer communication model to process the hundreds of billions of transactions between devices will significantly reduce the costs associated with installing and maintaining large centralized data centers and will distribute computation and storage needs across the billions of devices that form IoT networks. This will prevent failure in any single node in a network from bringing the entire network to a halting collapse.

However, establishing peer-to-peer communications will present its own set of challenges, chief among them the issue of security. And as we all know, IoT security is much more than just about protecting sensitive data. The proposed solution will have to maintain privacy and security in huge IoT networks and offer some form of validation and consensus for transactions to prevent spoofing and theft.

The blockchain approach

Blockchain offers an elegant solution to the peer-to-peer communication platform problem. It is a technology that allows the creation of a distributed digital ledger of transactions that is shared among the nodes of a network instead of being stored on a central server. Participants are registered with blockchains to be able to record transactions. The technology uses cryptography to authenticate and identify participating nodes and allow them to securely add transactions to the ledger. Transactions are verified and confirmed by other nodes participating in the network, thus eliminating the need for a central authority.

The ledger is tamper-proof and cannot be manipulated by malicious actors because it doesn’t exist in any single location, and man-in-the-middle attacks cannot be staged because there is no single thread of communication that can be intercepted. Blockchain makes trustless, peer-to-peer messaging possible and has already proven its worth in the world of financial services through cryptocurrencies such as Bitcoin, providing guaranteed peer-to-peer payment services without the need for third-party brokers.

Tech firms are now mulling over porting the usability of blockchain to the realm of IoT.

The application of blockchain to IoT isn’t without flaws and shortcomings.

The concept can directly be ported to IoT networks to deal with the issue of scale, allowing billions of devices to share the same network without the need for additional resources. Blockchain also addresses the issue of conflict of authority between different vendors by providing a standard in which everyone has equal stakes and benefits.

This helps unlock M2M communications that were practically impossible under previous models, and allows for the realization of totally new use cases.

Concrete uses of blockchain in IoT

The IoT and blockchain combination is already gaining momentum, and is being endorsed by both startups and tech giants. IBM and Samsung introduced their proof-of-concept system, ADEPT, which uses blockchain to support next-generation IoT ecosystems that will generate hundreds of billions of transactions per day.

In one of the first papers to describe the use of blockchain in IoT, IBM’s Paul Brody describes how new devices can be initially registered in a universal blockchain when assembled by the manufacturer, and later transferred to regional blockchains after being sold to dealers or customers, where they can autonomously interact with other devices that share the blockchain.

The combination of IoT and blockchain is also creating the possibility of a circular economy and liquefying the capacity of assets, where resources can be shared and reused instead of purchased once and disposed after use. An IoT hackathon hosted by blockchain platform leader Ethereum put the concept of blockchain-powered IoT to test, in which some interesting ideas were presented, including in the domain of energy sharing and electricity and gas billing.

Filament is another startup that is investing in IoT and blockchain with a focus on industrial applications such as agriculture, manufacturing and oil and gas. Filament uses wireless sensors, called Taps, to create low-power autonomous mesh networks for data collection and asset monitoring, without requiring a cloud or central network authority. The firm uses blockchain technology to identify and authenticate devices and also to charge for network and data services through bitcoin.

Chain of Things is a consortium that is exploring the role of blockchain in dealing with scale and security issues in IoT. In a recent hackathon held in London, the group demonstrated the use of blockchain and IoT in a case study involving a solar energy stack designed to provide reliable and verifiable renewable data, speeding up incentive settlements and reducing opportunities for fraud. The system facilitates the process in which a solar panel connects to a data logger, tracks the amount of solar energy produced, securely delivers that data to a node and records it on a distributed ledger that is synced across a broader global network of nodes.

Caveats and challenges

The application of blockchain to IoT isn’t without flaws and shortcomings, and there are a few hurdles that need to be overcome. For one thing, there’s dispute among bitcoin developers over the architecture of the underlying blockchain technology, which has its roots in problems stemming from the growth of the network and the rise in the number of transactions. Some of these issues will inevitably apply to the extension of blockchain to IoT. These challenges have been acknowledged by tech firms, and several solutions, including side-chains, tree-chains and mini-blockchains, are being tested to fix the problem.

Processing power and energy consumption is also a point of concern. Encryption and verification of blockchain transactions are computationally intensive operations and require considerable horsepower to carry out, which is lacking in many IoT devices. The same goes for storage, as ledgers start to grow in size and need to be redundantly stored in network nodes.

And, as Machina Research analyst Jeremy Green explains, autonomous IoT networks powered by blockchain will pose challenges to the business models that manufacturers are seeking, which includes long-term subscription relationships with continuing revenue streams, and a big shift in business and economic models will be required.

It’s still too early to say whether blockchain will be the definite answer to the problems of the fast-evolving IoT industry. It’s not yet perfect; nonetheless, it’s a very promising combination for the future of IoT, where decentralized, autonomous networks will have a decisive role.

First appeared at Techcrunch.com