Imagine travelling on an underground train in a large metropolitan city, and suddenly the power cuts off. The passengers are now stuck in a tunnel, a considerable distance away from the closest station, waiting for someone to come and retrieve them. To make matters worse, there is no wireless reception, meaning there is no way to contact anyone on the outside.

As cities grow and urban populations rise, networking infrastructure needs to adapt. Currently there are segmented solutions involving a wide and distinct set of last-mile technologies e.g. cellular 4G stations, WiMAX, Wi-Fi, among others, with the backbone being physical fibre-optic or ethernet hardware. Depending on which service provider a user has, access to vital services on a given device may be very limited at certain locations, either due to lack of available signal, lack of specific infrastructure from that service provider, or other various reasons. This will likely get worse with time.

Enter UrbanNet, a hybrid mesh IoT solution for large and growing urban areas, capable of:
a) monitoring radio traffic in areas such as underground stations, public parks, or large buildings, and
b) providing basic wireless connectivity to localized areas during emergencies, ensuring that citizens have necessary access to digital services.

The key components of UrbanNet are as follows: Each node will consist of a modular design, with the initial node comprising of a user-facing ESP32 Wi-Fi chip + a network-facing LPWAN chip. Under normal operation, the ESP32 chip will be responsible for monitoring 2.4GHz and 5GHz network traffic, such as pings from Wi-Fi clients like smartphones or laptops, via an attached antenna assembly to cover that spectrum. This monitoring will be silent i.e. it will not require any public interaction, and it will not log any personal data such as MAC addresses. The LPWAN chip will be responsible for communicating the monitored statistics over the mesh network. This data can then be used towards urban planning activities, as it gives precise localised traffic numbers for the spaces covered by each node.

Under the second mode, known as special operation, the nodes in that area will switch to an independent source of power. This mode activates under emergency scenarios such as area-wide power loss. The independent source of power is a Lithium Polymer power pack that was previously kept charged via the local power grid or via a solar cell (if possible). The nodes in the mesh are now running as Access Points capable of sending user data at a baud rate of around a few hundred kB/s, enough to run a low bandwidth connection out of the affected zone. This is enough for people to send out location information or any other vital data outside of the zone, or to stay up to date with vital services while inside the zone.

The dual nature of the nodes is the incentive for cities to adopt them. Network traffic monitoring and logging is already carried out by service providers, but this concept goes beyond that and acts as an independent unaffiliated network. The users of the network can still connect to it in special operation mode regardless of their prior wireless service provider, and the nodes themselves are separated from wireless infrastructure physically and logically.

The UrbanNet concept has some novel advantages:
a) the data from the nodes can be analysed and 24/7 heat maps generated. These maps can be used to show any latent trends in urban spaces as time passes.
b) the data from the nodes will be public information. This means that city authorities and local businesses can use the data to help in planning out infrastructure for future projects. Since the data is not any more complex than pings over time periods, there is no risk associated with not keeping it public information.
c) the nodes themselves are modular. The default configuration is an ESP32 and a mesh-networking chip e.g. LoRaWAN or NB-IoT, but the nodes can be modified to add 4G/5G capability, or add another mesh-networking means if it is more suitable. The power supply can also be modified from a solar-charged power pack to one charged over POE instead.

The manufacturing and sourcing of the nodes is relatively straightforward, which will help greatly in their adoption. The baseboard PCB can be manufactured in bulk, and all the networking components (ESP32, LPWAN, antennae etc) are off-the-shelf. The battery packs are off-the-shelf as well. The nodes can be assembled in an off-the-shelf electronics case and fastened to walls. Since the nodes are lightweight and there is barely any cabling required, the installation requirements and costs are at a minimum.

Cities need to adapt to their growing populations. Using analytics gathered from these inexpensive IoT devices, cities around the world can predict where there is impending wireless congestion to digital services and act before it happens. And in the case of sudden loss, the nodes themselves can act before citizens lose access to those services.


The inspiration for this idea came to me when I was travelling underground. Oftentimes, there would be a loss of signal depending on where I was in the subway network. Seldom, it would also be because of the amount of people in the area. I was thinking of how worse this could get as cities grow if adequate or correct planning to counteract this does not take place. Having a network of devices, not just on floor-level, but also spaced out so that they can keep you connected through a subway tunnel was one of the initial solutions that I was thinking of. Keeping components as little as required, preferring off-the-shelf hardware, and having the mesh network not depend on external power or data sources is what I feel is the smart way of going about building this idea. I feel this is a unique, inexpensive and novel solution that will help future urban planning greatly. This is why I believe that my applying to the IoT Innovation Challenge is worth it.


Voting is closed!