Z-Mesh
ICN based IoT Network protocol
Scalable
Interoperability is a prerequisite for scalability. Z-Mesh allows heterogeneous devices to integrate, communicate and share information with each other.
Next gen IoT
Scholars agree the IP protocol is bad fit for IoT devices. Z-Mesh, being a Information Centric Networking architecture, is a better fit for IoT because of the named data approach.
Actuation
The value of an IoT solution comes from act of reacting on the sensor-data. Z-Mesh supports direct two-way communication, even for sleepy devices.
Runs on anything
Being physical layer independent means that Z-Mesh runs on anything from constrained wireless devices to mains-powered devices. See Metcalfe's law below.
Multi-vendor IoT
Many IoT solutions are vertical and causes vendor lock-in. Separating the physical layer from the application layer allows for easy replacing of components.
Ultra low power
Battery-driven Z-Mesh devices wake up, broadcast data and go back to sleep. The Network takes care of the caching and routing.
Privacy-by-Design
Cloud optional; Data stays local/on-prem, encrypted when stored and when in transit.
Cost savings
Integrate once; Add a new sensor or app and benefit from the Network Effect. Scaling is cheap.
Vendor independence
Z-Mesh is an open and royalty-free IoT-protocol with Built-in Device- and data-flow management.
# The problem Z-Mesh is solving
TLDR: Interoperability, Scalability & Vendor-neutrality
Vertical sensor-to-cloud IoT solutions are able to get you started with collecting and visualizing data quickly, but this is where the problems start. IoT solutions creates value when you actually do something with the sensor-data and as vertical IoT solutions are incompatible, you end up with expensive custom integration projects. In fact, Metcalfe's law (see below) states that the value of a network is only released when all the IoT devices can communicate directly, this is what Z-Mesh has been designed for.
The IoT landscape is dominated by fragmented, proprietary, supplier-specific ecosystems. While effective within a particular ecosystem, such an approach limits the ability to scale and integrate, constraining the impact of IoT deployments and driving up costs.
-- McKinsey 2021 report (opens new window)
# How to achieve interoperability?
Metcalfe’s law (opens new window) when applied to IoT, says, the value of an IoT network (solution) is proportional to the square of the number of addressable pieces of Content Names. That is: If all Content Names can be retrieved by any Consumer, you have maximum value.
Unique Content Names (addressing) allows Content Producers and Consumers to be properly routed and connected within the network, enabling communication and data exchange between them. Without unique Content Names (addressing), the network would not be able to effectively route and deliver Content, limiting the overall connectivity and value of the network.
As the network grows, the number of connected devices and systems increases exponentially. Robust addressing schemes (Content Naming) are necessary to accommodate this growth and ensure that the network can scale effectively, maintaining the network effect as the number of connected entities expands.
Unique Content Names (addressing) enables interoperability between different devices and systems within the network. This interoperability is crucial for realizing the full potential of the network effect, as it allows diverse entities to seamlessly communicate and collaborate.
Unique Content Names (addressing) facilitates the management and control of the network, allowing administrators to monitor, configure, and troubleshoot individual devices or systems. Effective network management is essential for maintaining the stability and reliability of the network, which is a key factor in achieving the network effect.
# Privacy-by-Design
In Host based communication models, like LoRaWAN and IP, the sensor-data is tunneled in an encrypted channel to it's destination. If the sensor-data is to be forwarded to multiple destinations, or just cached for lator retrieval, then the sensor-data resides on the forwarding device unencrypted. Obviously this is a huge security risk. Digital twins are forwarding/caching devices and sensor data resides unencrypted inside, ready for intruders to take a copy.
With Z-Mesh, being an ICN architecture, the sensor-data itself is encrypted and has a name attached to it. This allows any device to ask for sensor-data with that name or to forward or cache the sensor-data. Only devices with the encryption key can decrypt and use the data. With Z-Mesh, data is always transmitted and stored in encrypted form, this provides the data-producer full control over who can use the data.
# Why not IP?
The Internet protocol is successful because it is royalty-free networking layer standard that allows direct communication using the same naming-scheme like sensor.your-domain.org. However, it is a bad fit for IoT as it does not support offline devices, lacks support for mobile data-producers, is too complex for constrained devices and is a host-centric networking architecture in which data exchanged in channels.
Z-Mesh is an Information-Centric Networking (ICN) architecture where devices communicate via name-based data. The object of encryption is the data, not the channel. This makes Z-Mesh able to support one-to-many event-driven communication, offline devices, mobile data-producers and constrained devices.
# Metcalfe's law
The value of a telecommunications network is proportional to the square of the number of connected compatible communicating devices
-- Bob Metcalfe (opens new window)
Two telephones can make only one connection, five can make 10 connections, and twelve can make 66 connections:
This is why the IP protocol has been so successful. Z-Mesh is following in it's footsteps.