Service coverage is a fundamental limitation for CSPs. If people cannot get good service coverage then it becomes harder for CSPs to sell their products. Service coverage cannot be expanded/improved without careful planning due to cost and logistic constraints.
Currently service coverage planning is centralised and driven by various internal and external factors like regulation and prediction of consumer behaviour.
This results in overprovisioning in dense areas (e.g. city centres) and under-provisioning in rural areas. The impact of this can be seen in rural areas where significant initial investments are required for site prep and connectivity.
Then events like the COVID-19 pandemic can mess up the plans (e.g. due to work from home mobile data traffic decreased by 5% [1]) where overprovisioned sites are significantly under-utilized.
Enter the World Mobile Token
What if there was a way of providing access devices to consumers and businesses to deploy and maintain, who could then be incentivised to provide local coverage. This could be as simple as deploying a set of antennas and a box that provides connectivity to the internet.
If the platform can be compressed to a box similar to the box that most CSPs send out when you buy broadband/TV from them and you just setup an antenna and connect the box to it. Then you not only provide connectivity to your local area but also earn compensation.
One issue with doing this is that CSPs cannot handle the massive distributed transactional load it would generate. Imagine the volume of transactions as people use your node and move away. Your node then has to send data to the CSP to validate transactions and CSP has to pay you compensation. At the end of the day the people using the node should pay the people hosting the node. Why do we need CSP to play middle-person (along with a bank perhaps)? Then there is a question of ID management, privacy and securing the network.
We do know of one technology that does provide distributed, trust-less transaction support (distributed ledger technology most commonly implemented using blockchain). World Mobile Token combines distributed ledger technology with telco functions to facilitate demand-based coverage growth. Keeping the environment in mind and its focus on rural connectivity, especially in developing countries, the solution is based on low-power architecture powered by solar and battery.
The World Mobile Token (WMT) is the ‘coin of the realm’ for the network and will be used to buy and consume services on the network. So when a user does a balance top-up they will be buying WMT which they will be spending based on usage. This is promised to be a seamless and transparent operation. Part of that spend will go towards compensating node operators whether in local currency or WMT (more on this later). The supply of WMT is finite (2 billion) with a fraction of that being circulated in the beginning.
Architecture
The WMT network consists of three types of nodes:
- Earth Nodes
- Air Nodes
- Aether Nodes
Earth Nodes
These nodes contain the core logic of the WMT system. The major modules can be seen in Figure 1.
Distributed Identity – this module provides a decentralised authentication layer.
Distributed Ledger – this module contains the blockchain implementation to support the user account balances, transactions, rewards and secure storage of user data. To ensure privacy the ledger encrypts and splits information into public and private parts. The transactions are public but the details of the transaction are held privately. The good thing about the implementation is that unlike the proof of work based consensus algorithm used in Bitcoin it uses proof of stake [2] which requires you to stake some money (skin in the game!) to be trusted rather than wasting energy solving meaningless maths puzzles. At the time of writing you need to stake 100,000 WMT for an Earth Node.
Telco Functions – this module supports the utility of WMT – the telco network functions. These include: routing of data, signalling, service management (create/modify/cease), service assurance, self-healing.
Internode API – the most important component as it enables communication between all nodes and sub-systems. It can be thought of as a universal bus that touches all types of components, nodes and external systems. This is also responsible for Authentication, Accounting and Authorization functions and integration of data with the blockchain for transaction processing, smart contracts and enforcing rules.
Earth Nodes can be used for: BSS type services (i.e. buying telco services, using credit to use services etc.) or Telco type services (i.e. making a call etc.). The two main type of telco services Earth Nodes offer are: voice/sms service and Internet access. Some of the more sophisticated add-ons like content delivery are expected to be added on.
Earth Node operators are compensated using WMT for blockchain functions from what I can make out in the whitepaper. This is similar to other utility-backed coins. But I am not sure, for example, whether for Telco services Earth Nodes will be paid in local currency or WMT?
Air Nodes
Air Nodes form the access layer for the network. As the name suggests these are the Radio Access nodes. These nodes come in different hardware configurations based on capacity and location requirements.
Figure 2 shows the critical role that the Air Nodes play. These are the units that retail and business users can deploy and run to provide connectivity. The operators of these units get compensated for their role. In a normal mobile network provider all this infrastructure would be created and maintained by the provider or delegated to another organisation.
One interesting point is that Air Node operators are paid compensation in the local currency of the area of operation. This ensures all the downstream operational payments are seamless (e.g. for electricity, maintenance, connectivity).
Aether Nodes
This is where new-school telco meets the real world to provide access to the Internet and enable connectivity (e.g. voice, SMS) with users on other provider’s network. There needs to be at least one of these in a country before service can be provided there. The operator(s) of the Aether node need to have the necessary licences from the regulator in that country.
Aether Node operation requires staking 1,000,000 WMT and compensation for running it is also paid in the local currency of the operating region.
Network Architecture
Given the ‘demand’ driven nature of the operations we expect lots of Air Nodes deployed around Earth nodes linking with one or more Aether nodes. Air Nodes can also mesh with each other to provide remote connectivity (e.g. between Town C and a village near it in Figure 3).
We see two types of paths in Figure 3 – the pink one shows a call from between two subscribers of the WMT Network. The green path shows a call between a WMT Network subscriber and a subscriber on some other network.
Figure 3 shows the network architecture as I understood from the whitepaper [3]. In terms of routing the nearest Earth node is preferred to maintain quality of service. Otherwise you could make a cartel of Earth nodes that send traffic to each other (to earn compensation) even if it leads to a reduction in the overall customer experience. For example in Figure 3 the Earth Node (EN) owner in Town A can form a cartel with the EN owner in Town B so they send traffic via each other no matter if it is heading to Town C or outside the WMT Network. This will overload the link between Town A and Town B ENs and reduce the quality of connectivity.
WMT network also publishes node quality metrics that allows different selection mechanisms for route calculation and promotes transparency. For example we always want to have a diverse path through the network as long as service quality is not impacted. Why? Because here traffic processing means money means incentives. Also distributing traffic evenly means links don’t get congested.
Conclusions
This is a very interesting project which should grab the interest of operators in developing regions (e.g. MTN, Airtel, Vodafone). The Aether Node operation requires regulatory approvals which are not easy to get.
But there are many questions that I have not been able to get an answer to. Maybe I missed something in the whitepaper [3].
Q1: What happens with the spectrum for the Air Nodes? Do they operate in unlicensed spectrum (e.g. WiFi)?
Q2: How do we incentivise connectivity between the nodes because that is one of the most difficult problems when it comes to rural connectivity?
For example looking at Figure 4 (based on Figure 3) – if we assume between Village B and Town B there exists another Village A – here were are able to deploy an Air Node (for the village) using some sort of radio mesh technology. But what about Village C which has no such transition point? In WMT Network how will we incentivise people to create longer distance connectivity using mechanisms such as microwave or fibre? One option is to use community-led fibre projects (by Village C in this case) that then recoup their money via the Air Node operation. But this kind of longer distance connectivity projects can be very expensive. Therefore it will be interesting to see how they solve this issue.
Q3: Using WMT coin instead of local currency can lead to complexities and over time with Central Bank Digital Currency (CBDC) it will be easier to use that instead of WMT coin as CBDC will be easily convertible to the local currency. So how will this system evolve/work over time?
Disclaimer: this post represents my personal opinion and I am in no way associated with World Mobile Token or promoting them through this post.
Fish Eye View 