Hey guys, wanted to put this together to give you a little more insight about what my team and I have been working on down in Medellin for the past year.

 
 
 

The basic idea is that we are building large, community-wide Wi-Fi networks that serve user in their homes (like a fixed-line service) as well as when they are out and about (like a mobile operator).

We are specially targeting informal settlements because the crazy density of these communities allows us to spread the cost of infrastructure out in a way that an economically viable and super scalable business can be created.

 

Below are some picture of our work with a bit more of a deep dive into the technical stuff below that. Happy to answer any question and/or really nerd out on network engineering with you guys!

 
 
 

— Technological innovation —

The red_ team believes that if technical risks can be overcome, a highly profitable, highly scalable service delivery model can be designed to serve users in dense urban environments, particularly those in informal settlements. The innovation is comprised of three principal aspects:

  1. An innovative user experience

  2. A cutting-edge network architecture

  3. A novel business model

What follows is an account of the innovative elements of each of these three aspects with regard to technology.

(1) Innovative user experience

red_ believes that client-side authentication and billing software can be created that enables users to securely connect to the red_ Wi-Fi network and to let them pay in real time for data. The authentication will rely on WPA2 - Enterprise and RADIUS protocols to carry out its function. This way, user devices will use a client app on their phone to automatically and securely connect to the red_ network. This authentication software will eliminate the need for captive portals, access management, WPA passkeys, and unsecured Wi-Fi hotspots. Connections will remain private and encrypted end-to-end.

While similar philosophies have been applied to network authentication [see https://www.eduroam.org/] they have not been coupled with real-time billing. However, the red_ team believes that it is possible to design a system to do this by creating software that interfaces with the network controller and mediates access to a client app on each user’s device, charging each user for Wi-Fi access on a per-packet basis.

(2) Cutting-edge network architecture

red_ 's thesis around creating last-mile networks is that by leveraging emerging wireless technology and software-defined networking, both the capital cost and the operational complexity of deploying infrastructure in dense urban environments can be drastically reduced. 

In order to test this thesis, the principal technologies red_ will utilize are:

  • Software-defined networking (SDN)

  • IEEE 802.11ax (Wi-Fi 6)

  • 60 gigahertz (GHz), multipoint-to-multipoint (MPtMP), self-organizing mesh radios

heex mesh with caption.png

The network will be comprised of two subsections: the user subsection and the backhaul subsection.

The user subsection will be a mesh of Wi-Fi access points using the emerging IEEE 802.11ax standard. 802.11ax has many benefits in terms of data rate, latency, range, and spectrum efficiency that all come together to make it a viable method of service delivery for end users [13]. These advancements give access points both the ability to perform in dense urban environments and to make Wi-Fi meshing far more practical than it was with previous generations of Wi-Fi.

A certain number of these Wi-Fi nodes will have an additional radio that will be used to connect to the second subsection of the network: the backhaul subsection. The backhaul subsection will be comprised of mmWave radio units that operate in the unlicensed V-Band (57 GHz - 64 GHz). red_ believes nodes with wide fields of view (>300°) can provide tens of Gbps of switching capacity and dynamically adapt to the network environment (eg. load, node failure). red_’s thesis is that backhaul layer radios can be connected in a true mesh, with paths that can be routed through different radios and the ability to use multi-path routing within the mesh for load-balancing and resiliency. Finally red_ believes that a mechanism can be designed to immediately provision new nodes via the SDN, eliminating the need for radio planning or manual alignment in order to enable plug-and-play installation in under an hour. Finally, a certain number of these nodes will be connected to a fiber point of presence.

Within the network design red_ envisions, hardware will act solely as hardware switches and routing will be abstracted to a server-based control plane. red_ believes that by adopting a software-defined LAN approach from the start, infrastructure cost can be lowered, the complexity of installs and network provisioning can be made less complex, and network management can be iterated on at the pace of software development. 

This novel approach to network design—in combination with emerging technologies—is red_’s attempt at creating a solution that can economically service high-density, low-income neighborhoods with broadband access.

(3) Novel business model

The above technologies, if proven viable, will allow red_ to decouple internet service from complicated pricing plans and instead offer a “pay-as-you-go” model that has been widely proven to work in developing markets [14]. If red_ is successful, Wi-Fi will be available ubiquitously for the first time in urban, informal settlements. However, business model innovation is highly dependent on a mitigation of the technical risks described above and a demonstration that the business model is financially feasible.