Blog Post

Mesh networks could bridge the gap between internet capacity and demand

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Despite the billions of dollars spent every year on internet infrastructure, the market and our mobile lifestyle are increasing the gap between capacity and demand. Consumers often face slow data connections or high prices for broadband internet access…when it is available. This problem may only increase as manufacturers of mobile phones, tablets, TVs, cars and kitchen appliances continue to hop on the internet connectivity bandwagon.

According to IDC’s research, in the next three years at least five billion new mobile devices will try to connect to the internet and to one another — and that’s not even counting the new “things” of the internet of things.

As Google and Facebook look to drones, balloons and satellites to deliver the internet to the world, it’s worth remembering that one of the most powerful pieces in the quest for global internet access isn’t up in the sky. It’s down in our pockets.

The humble smartphone’s ease of use, connectivity and widespread adoption puts it in a good position to bring reliable internet access to everyone. Considering that most smartphones already have the components to act as routers, one can imagine a new generation of networks, where each device becomes an active part of the network.

Mesh networks, as they’re called, are built out of a patchwork of wireless-enabled devices — smartphones, tablets, computers, even wearable devices that connect with one another directly in a peer-to-peer mode. These networks configure the data-sharing capabilities within those devices, so that the devices can share and receive not just data, but the connection itself, becoming pieces of the network’s piping. A group of devices linked along a mesh network becomes an incredibly powerful intranet. And if one of the network devices connects to the internet, they can share that access across all the other devices as well. It’s like crowdsourcing the network.

Mesh networks aren’t hard to create. In Greece, about 1,000 people linked their rooftop Wi-Fi antennae to create a completely off-the-grid internet — over a connection that delivers at an astounding 150 MBps — complete with blogs, a classified site akin to Craigslist and streaming movies. Computer owners in rural Spain, fed up with being ignored by major ISPs, took matters into their own hands to create Guifi, a mesh network boasting 21,000 members.

Since mesh networks are easy to create, they’re prime candidates for bringing connections to the world’s most internet-poor regions. Take the case of Daniel Hastings, an American teacher in Somaliland. One of Africa’s most desolate regions, Somaliland had no fiber optic cable as of the end of 2013, and yet Hastings was able to deliver quality internet connectivity to his school via a mesh network he essentially built on his own.

There’s another critical advantage: mesh networks are decentralized. Since every device in the network is a node, the removal of any one device won’t break the system. Mesh networks remain reliable even in a natural disaster. For example, a mesh network was critical in reintroducing network connectivity in Red Hook, Brooklyn after Hurricane Sandy.

Decentralization also gives mesh networks a particular advantage in areas where government is the critical barrier to internet access. Repressive regimes manipulate the web via ISPs. They command the pipes to control access and information (undoubtedly part of the reason North Korea only has one ISP). Mesh networks sidestep that problem by creating a system that has a limitless number of nodes and no central management. It’s possible to take control of cable-delivered broadband, or even to jam signals from drones or balloons. Overtaking a sprawling network of millions of mobile devices is a far harder task.

The idea gained enormous credence in 2011, when the Egyptian government shut down the country’s internet access during the Arab Spring. In response, hacktivists banded together to form the Open Mesh Project, a global organization dedicated to building mesh networks that governments couldn’t destroy.

But I’m not just talking about the developing world, which is expected to have more mobile internet users than the developed one this year. I’m talking about building networks out of our smartphones everywhere. In 2014, creating a mesh network is easier than ever. All it takes is software.

This software is already here. You may get it from an app developer, an operator or the manufacturer of your device. As this happens, operators can offload huge amounts of data traffic through the least expensive and most efficient off-ramps allowing them to satisfy the demand while realizing the ROI on their heavy investments. Device manufacturers and app developers can solve network discontinuity issues. End-users get better, faster and more reliable internet access. Everyone benefits from more reliable and faster connectivity. It is a win-win-win.

Micha Benoliel is the CEO and co-founder of Open Garden, the developer of mesh networking apps that connect over five million people worldwide. Open Garden also powers the mesh-network technology in One Laptop Per Child’s XO Tablet.

13 Responses to “Mesh networks could bridge the gap between internet capacity and demand”

  1. Michael Teal

    “Mesh networks aren’t hard to create. In Greece, about 1,000 people linked their rooftop Wi-Fi antennae to create a completely off-the-grid internet — over a connection that delivers at an astounding 150 MBps — complete with blogs, a classified site akin to Craigslist and streaming movies.”

    A completely off-the-grid internet is called an Intranet or an Extranet just so you know.

    • With 5 million users, Open Garden is the largest mesh networking technology made available to the public today.

      What exists in Greece even if it implies a specific hardware shows the potential for such networks, thank you for sharing.

  2. Michael Teal

    I don’t want to be against a wireless mesh network but after I thought about it for a while I think it would only become useful as a back up network to a hard wired network. Because of the scarce availability of wireless spectrum bandwidth it will eventually become congested when everyone is streaming Netflix all at once to their 4K or 8K TV. You can always lay more fiber optic cable to create more capacity but you can’t create more wireless spectrum to add capacity in a wireless mesh network. Also just imagine how many wireless APs from everyones cell phones that would create. It would create a disaster when hundreds of cellphones are simultaneously occupying and competing for the same airwaves. The 2.4 GHz WiFi spectrum is already congested almost everywhere especially when channels 1, 6, and 11 are the only non-overlapping channels in that band. If you have too many hotspots competing for the same channel everyone suffers from that.

  3. Michael Teal

    How is this proposal of using cell phones as a mesh network any different then using WiFi direct? WiFi direct can already transfer large files between laptops, cell phones, and tablets. And we also have the ability to use tethering over Bluetooth, USB, and WiFi.

    This proposal just sound like all it would become is an Intranet of cell phones similar to how WiFi direct works. How would the Intranet connect to the Internet?

    • Wifi direct is just a wireless interface like Bluetooth, Bluetooth LE or LTE. Open Garden build an overlay network that uses the availability of any of these interface to route traffic. The magic comes from the seamless discovery and pairing which normally works against the formation of dynamic mobile mesh networks.

      • Michael Teal

        would’t every cell phone in this network need to be in a close proximity to at least one networked phone to make this work? If so using longer range spectrum such as 2.4 GHz WiFi would be better then using a shorter range Bluetooth spectrum. Not to mention the data transfer rate of speed is better over WiFi.

  4. Hora66

    Well, maybe I’m not the right kind of visionary… But the only issue such a mesh could address is last mile (in certain cases). There always has to be some uplink to the Internet and you will not save a single Mbps here unless you install HUGE cache somewhere in the mesh. Also, I’d be really upset if my phone (with limited mothly data) accidentally became the uplink to the Intenet for all those fellows around…

  5. Madlyb

    Yeah…I already don’t trust my ISPs and only use Open WiFi as a last resort and then only for non-secure activities Why would I even consider letting my packets bounce through any number of random strangers’ systems.

    I agree that Mesh provides ways of connecting people that are more fault tolerant and robust, but until the state of networking security is much higher, there is no way I am going to use one.

    • @Madlyb I understand your concern on this. Accessing Internet through someone else device is probably more secure than connecting a public open Wifi you don’t know. You can take care of your own security with a VPN or by using https.

  6. Mesh networking is an insecure solution to a very specialized problem, not a general-purpose method of extending Internet service to end users. While it may seem to make sense to install a mesh in a home of office, the price of LAN bandwidth is effectively nothing whether it’s provided by Ethernet or Wi-Fi. The demand for actual Internet connectivity – through Internet Exchanges or priviate peering points – is growing, but much less fast than the capacity of the existing infrastructure. ISP bandwidth to the end user grows by 100% every three years, and the bandwidth demands of the IoT are very modest.

    People who have actually created mesh networking technologies – like me, for instance – are much less enthusiastic about their possibilities than non-technical people are. The store-and-forward delay kills you.

  7. I’m confused and maybe somebody can help me understand what’s being proposed. In my mind, the “mesh” would only create a private network, then through Internet sharing, provide wider access. I can’t imagine sharing my cellular data, even if allowed, would be a very useable speed beyond like 2 other users. I also don’t see how this helps with the supply/demand problem. You’re going to have just as many people accessing the Internet, right? The mobile mesh also has the battery life problem of its members needing to be “always on”.

    • OpenGarden

      @jeffdalydose the system as a whole can optimize your battery life as different radios have different power consumption. Let say you are low on battery but someone next to you is full on battery, instead of using your expensive 4G radio to check your emails, you can be using your BT radio and connect to the Internet through the peer next to you that is full on battery.
      BT consumes 50Mw when a 3G link would consume approx. 1000 Mw.
      We always need more capacity so the mesh is caching content that can enable you to access it faster without the need to connect back to a server on the Internet. In addition, it can solves the spectrum crunch by virtualizing the available spectrum: when you are on the mesh you can access to the capacity of your carrier but also to the capacity of other carriers simultaneously. That way it creates more available capacity according to a space and time.