Demystifying Mesh Networks
To understand what RightMesh is, let’s first understand what a mesh network is.
We’re grateful for all the interest we have been receiving from the community on our Telegram channel, and to be honest, it’s been fun to answer questions about our project over the past several months! There have been some excellent questions asked, and we’re very excited to move forward and start testing our platform on the ground later this year.
One of the questions we get a lot is how are we different from other competitive projects in the mesh networking space, which is a fair question. It’s important to note that ‘mesh networking’ is quite an abstract term to define this large space, and there can be different forms of mesh networking implementations. To answer this question, we believe it is important to have a basic understanding of what a mesh network really is.
This post intends to explain some of the basic concepts of mesh networking (especially for the non-technical reader) and answer some specific questions about what makes RightMesh different from other mesh networking projects.
What exactly is a mesh network?
A mesh network is a type of network in which the infrastructure is carried by nodes that connect with each other directly and dynamically, and cooperate with one another to efficiently route data packets.
Didn’t quite get that? Well, let’s first understand how traditional internet networks operate today. Let’s assume you want to send an email, and for that you need to have a device — be it a PC, laptop or mobile device — that connects to a network. For the PC or laptop, this usually means connecting to a Wi-Fi access point, or plugging an Ethernet cable into your device. And for the mobile device, this would mean connecting to a local cellular operator or selecting a Wi-Fi network.
From there, your devices are then connected to a central hub or node — which could be a Wi-Fi access point, a network switch, or a cellular base station, and these nodes are connected to other central hubs up in the hierarchy until the email finally reaches the destination, which typically is an email server located in the cloud.
While the advantage of this network lies in its simplicity of adding new devices to the network, this approach can be rather inefficient for a number of cases. Assume that a professor is emailing a document to her classroom of 100 students. That document needs to go up through a long path of multiple network hubs until it reaches the email server, and then is pulled down again separately by each student, for a total of 100 downloads of the same file. And in this scenario, every student requires a dedicated connection to the Wi-Fi access point, and without that they cannot get the document.
Mesh networks, on the other hand, involve network ‘nodes’ that can dynamically connect to another nodes, and this removes the dependency that one node has on any central hub or provider.
Because each node has the ability to dynamically connect to other nodes, the links between the nodes of a mesh network are typically wireless in nature. The wireless technology used between nodes in a mesh network could be homogeneous (i.e. all nodes using Wi-Fi to connect with neighboring nodes), or heterogeneous (e.g. nodes using either Wi-Fi or Bluetooth or both to connect with neighboring nodes)
Are all mesh networks the same?
Absolutely not! There are different forms of mesh networks to satisfy different needs, objectives and purposes. Generally speaking, mesh networks could be placed somewhere on two scales: fixed vs. mobile mesh networks, and closed vs. open mesh networks.
Fixed vs. Mobile Mesh Networks
A fixed mesh network is one where the nodes are stationary in position and do not move. The nodes in these networks are more commonly seen as ‘mesh routers’ that operate in Wi-Fi networks (e.g. the Cisco Meraki routers for commercial deployments or Eero routers for residential deployments), fixed point-to-point antennas (e.g. the supernodes used for the NYCMesh and guifi.net mesh networks, or the wireless transmitters to be used in Althea mesh networks), or fixed multi-radio mesh routers (e.g. Ammbr).
A mobile mesh network on the other hand is one where the infrastructure nodes are not stationary and are constantly in motion. Examples of mobile mesh networks include goTenna and Sonnet Labs.
RightMesh falls under this category as well; however, the primary difference between RightMesh and other mobile mesh networking projects is that most projects usually employ separate wireless devices or antennas to act as the network nodes in a mesh network, and users can then connect their smartphones to those devices.
In RightMesh networks, on the other hand, the smartphones themselves act as the network nodes in a mesh network.
It is also important to note that the performance of the mesh network is dependent on the type of mesh network (fixed vs. mobile), the capability of the nodes in the network, and the wireless technology used to connect nodes.
Fixed mesh networks that rely on consistent hardware for nodes and use a common wireless link will have a better quality of service than mobile mesh networks with heterogeneous nodes. However, fixed mesh networks come with the cost of deploying and maintaining an infrastructure whereas mobile mesh networks can be spontaneously created anywhere at a very low cost.
Centralized vs. Decentralized Mesh Networks
A centralized (or closed) mesh network is one that operates in a controlled environment, and is usually setup by a single organization for a specific purpose. Because it is in a closed environment, and there needs to be a certain quality of service, these networks are usually implemented as fixed networks with homogeneous wireless links. Examples of these types of networks are Wi-Fi mesh routers that connect different floors of large hotels or office buildings, or residential Wi-Fi mesh routers that are spread across a home to ensure coverage across all areas.
Decentralized (or open) mesh networks on the other hand are crowdsourced mesh networks in which the network is run by a large group of people instead of a single organization. The work needed to run the network is spread across various users to achieve a cumulative result for everyone in the network. These networks are community driven and can serve as a powerful force to empower people — especially marginalized communities which have difficulties accessing traditional networks because of cost or infrastructure barriers.
RightMesh is a decentralized mobile mesh networking platform because this technology best fits our mission which is deeply rooted in connecting the underserved.
Why does RightMesh use smartphones as network nodes instead of an external device (like goTenna or Ammbr)?
One of our initial missions at RightMesh is to connect the underserved areas of the world through crowdsourced mesh networks, and a key element for mesh networks to thrive is having a density of nodes that can cooperate together. And so, to be successful in our mission, it is extremely important that RightMesh attain a density of nodes while having low barriers to entry for new users.
We believe that having an external device to be used as a network node imposes a significant barrier of entry, especially when the cost of the device is to be borne by the user. Our goal is to have any smartphone participate as a network node just by installing an app — as simple as that.
With that said, it is much more challenging to build a mesh networking stack for a smartphone than using an external device — especially when the native operating system (Android) is not engineered for mesh networking at its core. In addition, introducing a variety of different devices may impact the quality of service.
In spite of the challenges, we believe that reducing barriers to allow more participation is a more important goal to achieve early on than providing a high quality network for only a few people.
Longer term, as we scale in emerging markets and move towards our long term vision of creating mobile mesh networks that can be used across the world and across multiple devices, we hope to backward integrate with chip and handset manufacturers to improve the quality of mesh networking nodes.
How does data get transmitted over RightMesh networks?
There are two forms of networks possible with RightMesh that determine how data is transmitted over the mesh network: mobile adhoc networks and delay tolerant networks.
Mobile adhoc networks (MANETs) are those configurations in which a number of nodes are connected together, and information is passed from source to the destination over a number of hops. These networks form autonomously and are self-forming, self-healing and self-regulated.
When RightMesh nodes come into close proximity with each other they can form MANETs dynamically without any internet connection, and users can interact with each other through local services and apps built on top of the RightMesh platform. A use case for such a scenario could be university students interacting with each together using a messaging and sharing app that runs on top of a RightMesh MANET formed using the smartphones of students. This type of MANET is local, completely offline, and can only be used at the university campus.
It is also possible to have nodes in a MANET which act as ‘internet sharing nodes’ providing internet bandwidth to other nodes in the network. A user can ‘buy’ data from the internet sharing node and subsequently send or receive information from the internet that will be transferred to it through various mesh nodes. The internet sharing node would be the gateway between the local mesh network and the internet. This becomes a mobile mesh network where data is not only transferred between neighboring nodes, but also to and from the internet through the internet sharing nodes.
Delay tolerant networks (DTNs) are another form of mesh networking where nodes in a network may not have continuous connectivity, and therefore will not have an instantaneous end-to-end path. But they can pass on information to another nodes when they are in range.
This employs more of a ‘store and forward’ approach where data is incrementally moved and stored throughout the network in hopes it will eventually reach its destination. Applications that need to distribute data or content to recipients are better suited to use DTNs.
It’s also possible to have a blend of both networks. For example, disparate local MANETs could be linked together through superpeers to form regional mesh networks. However, in the absence of any internet sharing nodes, information could be passed from one MANET to another using DTN and through nodes that move from one MANET to another.
(Note: Phase 1 of the RightMesh platform will support MANETs and internet sharing nodes, and DTNs will be supported in the next phase.)
Do we really need mesh networks? Won’t traditional networks connect the entire world eventually?
We believe this is wishful thinking.
It is true the communication networks that have emerged and evolved over the last several decades have greatly improved life in all sectors, bringing voice, data, and the internet to billions of people all over the world. And this has resulted in new and improved ways for people to interact with each other.
This has also led to a handful of massive corporations accumulating large amounts of profit from these networks over these decades and accordingly, are only interested in actions that will lead to even more profits.
The reality is that there are vast areas and communities around the world which are not even close to receiving the benefits that connectivity the internet provides because there is no incentive for these corporations to serve these areas that equate to low profitability and high operational costs.
This is where mesh networks play an important role. Since they are crowdsourced and are run by the users themselves, networks can be more accessible and affordable for everyone and can potentially help narrow the digital divide that is so prevalent across the world.
Why is blockchain and the RMESH token important for RightMesh? Does it even need it?
Technically, a mesh network does not require a blockchain. And centralized mesh networks do not have any need for the blockchain. Decentralized mesh networks; however, pose a different need because of their inherently crowdsourced nature, and the need to be run by a group of anonymous people who provide work for a community good.
The core of RightMesh is indeed mobile mesh networking; however, it is the blockchain that allows the network to be truly open, decentralized and community managed.
Our vision is to create a truly decentralized mobile data network where participants have more freedom of choice to decide how they would like to be connected, for what service, and at a rate that is affordable to them. In this regard, the blockchain does several important things for RightMesh:
- It facilitates trust between various users in the network and allows the network to work as a true peer-to-peer sharing economy (unlike Airbnb or Uber which rely on central organizations).
- It allows for economic incentives for users in the network to provide ‘connectivity’ between devices in a frictionless manner in the form of digital tokens.
- It establishes a free market where resources can be bought and sold by a large group of anonymous users, and prices can be set by users themselves based on forces of demand and supply.
Mesh networking may be the core, but the blockchain and token provide the fuel to power the network!
Networks Can Be Meshy!
Mesh networks are vastly different from each other, and the answers to these questions determine what differentiates them:
- What are the hardware devices that act as nodes in the mesh network?
- Are the nodes fixed or mobile? If mobile, are the networks that are formed MANETs or DTNs?
- What is the wireless link used between nodes?
- What is the purpose of the mesh network? What problem does it intend to solve?
- Is the mesh network controlled by a single entity or is it crowdsourced and community managed?
For more information on how the RightMesh platform and protocol work, please have a look at our recently updated RightMesh Technical Whitepaper!
Thanks to Dr. Jason Ernst and Dr. Lucien Loiseau for reviewing and providing their valuable feedback on this post.
