Is Internet of things enough for Africa?

NAIROBI, KENYA: There are a lot of discussions on the Internet of Things (IoT) and its contribution to solving problems of our world. But are IoT technologies sufficient to address the grand challenges of Africa?

Whether we are referring to your smartphone or a sensor tracking rain fall, the IoT is a means to an end, which more often than not is about optimisation. If an athlete wants to optimise his or her running route they will need to track the distance and monitor the altitude – this can all be done with a phone or GPS watch.

If a farmer wants to optimise water and fertiliser usage, sensors can be buried in the earth to collect moisture and mineral samples.

But that is where the capabilities of sensors end, collecting the data. To then optimise the running path or the farm the data needs to be pushed to the cloud, combined with other data sources and then analysed by AI and machine learning analytics to bring the best results.

Steps of Optimisation for IoT:

The successful path to optimisation has the following steps in continuous iterations, regardless of the solution:

Monitoring:  As of today, IoT is bringing a myriad of opportunities to us such as monitoring the air quality, traffic flow speed, fire detection, power usage of a device etc. If you have a look at solutions being implemented, the majority of the projects fall into this category.

Management: In some cases, it may be necessary to manage the environment via IoT Nodes. For example, if the water level at a water reserve is lower than warning level, it may require stopping water delivery at non-critical areas.

Optimize it: After collecting information about the monitoring point, before taking management action, analysing and comparing the data to historical data or other sources helps to make more accurate decisions. For example, even if the water reserve is below average levels (but above critical level), but a heavy rainfall is expected in the next 24 hours, it alters the decision making process.

Some areas that should be considered in IoT projects:

For successful IoT project, the following areas should be considered;

Transmission of the IoT data: it is one of the potentially big costs in any long term project budget. Also, physical factors and the value of the data will impact the more effective solution. For example, if very critical data should be collected from a remote location in real-time and if there is no GSM coverage, very expensive satellite based communication should be considered. But, if the data is not critical to collect in real time, some other technologies should be considered. So, it is important to select most suitable communication technology

Physical Security: Sometimes, it will be necessary to place IoT nodes in public places, for example, air quality detectors or traffic monitoring sensors. One point to consider is the physical security of the devices. Because one of the most important cost of any IoT project are the devices and installation costs. Sometimes, it may require locating them in places which are more secure; but perhaps not the most optimum place to collect data.

Supplying power to IoT Nodes: In countries like Kenya, providing sufficient power for IoT devices needs to be carefully considered. Many projects across Africa use solar powered IoT nodes, but then one must consider if connecting data in the evening is important or not. In addition, the solar cells on the devices will need be kept dust free. This is why many scientists at IBM are exploring various nanotechnology innovations to keep sensors low-powered to run off standard batteries for years at a time.

Operation of the IoT Nodes: For most projects, the numbers of nodes are in the hundreds and sometimes thousands, making operationalization monitoring a challenge. If an IoT device is removed, failing or is tampered with, it is important to be notified as soon as possible with relevant information (i.e., Location, model, responsible team, etc.) or you will risk receiving corrupted data. Also in some networks the nodes are like a chain and if one of the links breaks, so does the entire network. Inventory management and monitoring the health of the IoT Nodes is paramount.

Confidentiality and immutability of collected IoT data: One of the areas that should be carefully evaluated is the confidential nature of the data collected from IoT Nodes. Naturally, collecting rainfall or pollution data is not confidential.

On the other hand, energy usage of a house can give excellent insight about if households are on vacation or not.  Or healthcare data, such as tracking the spread of tuberculosis.

In addition, the immutability of the data is essential. For example, certain malaria and HIV tests need to be kept at room temperature. If they stay too long at extreme temperatures it will damage the test. This is why technologies such as blockchain are being used with the IoT to assure users that the data cannot be tampered with.