Social Sensing using Ethereum Blockchain
Introduction
It was just the last decade that gave us 3G mobile internet and the iPhone, allowing us to access and share digital content with our friends and family using our mobile phones. Sites like Facebook, LinkedIn, Instagram introduced us to social networking to keep in touch with our friends, colleagues, and people in different circles.
Turns out, our activity on the social networking sites is a treasure trove for data analytics. After acquiring the data, we can analyze, report and act on the newly learnt information.
In fact, the data being collected from our web activity has been doubling every 40 months and these days big data storage can exceed Petabytes or Exabytes. Data analytics programs help us fight spam, recommend us which movies we might like and tackle more serious issues like predicting forest fires and financial fraud. But all of that analysis and reporting happens in centralized databases and programs.
Another approach is in-situ data processing. Instead of collecting data, and processing in centralized databases, we send computation to sources that create data. That brings us to social sensing.
Sensing Network
So what is social sensing? Social sensing is a relatively new concept which was coined just about the same time Ethereum appeared. The modern smartphones are not just communication tools — they are also computational tools, capable of working with mobile networks to global cellular networks. For example, smartphones now measure noise that closely matches professional acoustic equipment. They also can pin-point GPS coordinates accurately, and have powerful multi-core processors capable of computing complex calculations.
A smartphone can be used as a wireless control device that can unify monitoring and control over other kinds of electronic devices making them mobile sensors carried by humans rather than placed at static locations.
So, instead of collecting data for processing later, we could perform necessary computations in-situ and just forward the results securely. The term social sensing is used to describe the individual-level sensor data analysis where people and sensors in different places can easily share sensor data and cooperate with other sensors.
Ethereum Blockchain
Now why stop at analyzing the data? What if we could act and enforce rules automatically from what the data tells us? That’s precisely what smart contracts help achieve.
Ethereum can be viewed in two ways:
- A blockchain, which is a type of distributed ledger or decentralized database that keeps records of digital transactions
- A platform called Ethereum Virtual Machine (EVM) with its own instruction set which is used to write and execute programs called smart contracts.
Complex logic can be expressed in smart contracts just as you would in a high-level programming language like Java, C# or JavaScript. Smart contracts are executed on every Ethereum node simultaneously and the results are stored within the blockchain.
So, if we could implement rules to be enforced in a smart contract, a sensing network can not only analyze and report on telemetry data but it can also act on the results automatically and enforce any necessary rules!
Decibel.LIVE
Decibel.LIVE is a noise pollution monitoring startup that leverages smart contracts to react to the sensory information in a predictable manner. We are building a social sensing platform on the blockchain — two seemingly unrelated fields of computing.
We use smartphones and microcontroller based IoT devices deployed around the noise sites to capture noise events from multiple sources. The sensors form a local, private and secure mesh network which forwards the results such as average loudness per minute in dB(A) units.
Some challenges in a social sensing network and how we are addressing them:
- Security and Privacy — our most important goal is to perform privacy-preserving sensor data collection and analysis. To this end, we use P2P communication protocols to ensure end-to-end encryption, beyond hardware-level encryption. Noise events are collected and stored on IPFS by pseudonymous IDs.
- Energy Efficiency — microcontrollers that use minimal power and software running on local storage resilient to communication failures.
- Veracity — Smartphone apps and noise instruments have built-in calibration tools to ensure accuracy of readings. Devices participating in the mesh network are intelligent enough to compute attenuation rates to pin point sources of noise.
We are currently building an MVP and refining the implementation based on the results and feedback from testing.
In the upcoming articles we’ll explain our hardware and software products, including PCB designs in detail. We’ll show how they help citizens and local agencies curb noise pollution and help create smart cities.
For more information, visit us at https://www.decibel.live and chat with us on Slack https://decibellive.slack.com