How fragile is the internet in Africa?

In the past few months, internet and international telephone services were significantly disrupted across many African nations due to sub-sea cable damage. In any typical year, around 100 incidents are reported globally, and normally affect one cable at a time. Two separate events just weeks apart caused the loss of multiple cables on both sides of the African continent. Damage to a single cable will often go unnoticed as cable operators have agreements in place to reroute traffic. So, why did the internet fail in Africa?

In this article, I will explain what happened and explore some actions aid organisations can take to mitigate against future events.

24th February 2024:: Three cables were cut in the Red Sea (Seacom/TGN-EA, EIG, and AAE-1). The initial speculation stated that the cuts were carried out deliberately by the same rebels who are targeting shipping in the Red Sea in support of the Gaza war. This theory was soon disregarded in favour of a more likely cause which is assumed to be the anchor the crew of the MV Rubymar deployed following a missile attack. The ship eventually sank. The outage of the three cables caused major disruption to internet services across the East and Southern African region. Services started to improve in the following days as ISPs worked together to re-route traffic via alternative cable.

14th March 2024: Four cables were damaged off the west coast near to Abidjan. This time an underwater
rockslide was suspected to have caused damage. Services in West and Central
Africa were severely disrupted. Whilst mobile networks continued to operate for
local and national calls, the ability to make international calls and use data

Resilience and recovery: In a report published by the internet society in April 2024. The organisation covered the events in detail and explained that timelines for repairing cables ranging from 4 to 6 weeks. The Infrapedia map is a global map showing sea cable routes. In the screenshot below, the cable routes for West Africa are clearly shown.

Returning to the question from the top of this article “why was Africa severely impacted”, this map demonstrates two factors contributing to the high risk of losing connectivity.

  • Low number of sea cables serving African countries. Some countries more resilient than others. Liberia has just one cable, but Nigeria is in a stronger position with 7 cables.
  • Multiple cables routed through the same place will heighten the risk of major disruption as a single event can damage multiple cales at the same time.

Multiple cable damage is exactly what happened in the Red Sea and the Atlantic. West Africa was more severely affected due to the low volume of cables in the area.

What can organisations do to reduced the impact of a connectivity outage? When a major outage happens, both internet access and ability to make international calls will be lost. This also affects mobile networks as well as fibre connections. As these events are completely outside the control of aid agencies, there is not much IT teams can do other than to monitor the situation and inform staff.

There are some actions organisations can take to reduce the impact when connectivity is lost. It is possible in many nations to implement back up services which can be activated during an outage. Here are a few suggestions.

  • VSAT – This is satellite-based internet. Bandwidth will not perform to the same level as a fibre connection, but it will allow organisations to perform essential online tasks. VSAT is technology that organisations need to buy in advance and have contracts in place which enables services to be switched on at short notice.

    VSAT relies on ground stations to link satellites to the internet. If the ground station is in the same or nearby country where the internet has failed, VSAT will not work. When sourcing VSAT services, its best to select services where ground stations are in Europe and other locations with a high density of internet infrastructure.

    There are some countries where internet is strictly regulated, and ISPs are required to use ground stations in the same country. In these circumstances, VSAT will not be an appropriate solution.
  • Portable Satellite Communications – Satellite telephones can be held in reserve to allow internal calls to be made. Inmarsat BGAN and Thuraya IP are compact units (small notebook size) which can be used to access the internet. Inmarsat and Thuraya satellites connect to the internet via ground stations in the EU, USA, UEA and other global north locations. This technology has a reputation for being expensive, but in recent years, affordable packages have become available. (BGAN Link provides 30GB of data for around $1000 per month).BGAN and Thuraya IP operates at around 350Kb/Sec, which is nowhere near fibre speed but sufficient for a small team to carry out essential work.
  • StarLink – During the outages of February and March 2024, many NGO leaders were calling for StarLink to be sourced. StarLink has a high profile in the media as it has been strongly promoted by Elon Musk its CEO. This new technology delivers internet at high speed and low cost.

    At the moment StarLink can only add limited value in Africa as it can only be used in Nigeria, Kenya, Rwanda, Mozambique and Malawi. More countries are working through legal agreements with other countries, so this list will expand.

    Based on my research, there is only one StarLink ground station in Nigeria. As this was one of the countries affected by the outage, it was assumed that the StarLink services may have also been affected by the cable cut. During a recent conversation with StarLink, the sales team told me that StarLink performed as normal during the recent outages, but I have not been able to independently verify this statement.

    One of the reasons that StarLink may have performed well maybe down to the way data can be transferred between satellites via 100GB/Sec laser links. As StarLink is new to Africa, it’s feasible that performance may have been good, but how well would the StarLink network perform during a major cable cut in the future when there are significantly more users accessing StarLink?
  • ISPs: What can your internet supplier do to keep services running during a cable cut?  IT and Procurement teams should ask more technical questions about how the services are provided. Is it possible for supplier to have back up connections to keep limited services running when sea cables are cut. As an example, ISPs in Liberia used to rely on large VSAT systems to provide internet services. When Liberia was connected to the ACE cable in 2012, many ISPs retained their VSAT systems as a backstop.

    Specific question to ask ISPs;

    • How many sea cables do your supplier by internet services from? In some of the recent outages, some ISPs performed better than others as they were able to move traffic quickly down the remaining operational routes.

      Does the supplier operate an independent VSAT back up?  If so, what is the capacity and where is the ground station?
    • In the event of a major cable incident, what service levels can the supplier guarantee via its back up infrastructure?

Conclusion: Most organisations rely upon internet-based services for running daily tasks from managing HR to running procurement. When the internet goes offline, this has a significant impact on operations. As recent events have demonstrated, total loss of internet access is a possibility and more likely in places where cable density is low.Sea-Cables are vulnerable to damage from ships, trawlers, and seismic events. It is also important to understand that the risk of damage to infrastructure by bad actors is possible. In 2022, it was suspected that the Nordstream gas pipeline in the Baltic sea was sabotaged by bad actors. Some nation states have made it clear that they consider sea cables as a legitimate target during a war.

So, organisations need to take the risk of internet outages seriously, especially in locations where sea cables are limited to low numbers. This risk of outages need to be mitigated by investment in back up connections such as satellite-based solutions. VSAT is possibly the most solid option, but new LEO technologies like StarLink will have an increasing role to play in the future.

GPS Jamming & a solar lantern review

In this month’s article, I am covering two topics. First, GPS jamming which has become a significant challenge in some parts of the world to aid organisations and beyond. The impact can be significant on navigation systems and telecoms. Over the past few years, we have seen an uptick in jamming, so I want to make people aware of how this can impact aid operations and personal security.

The second topic is about personal solar lighting. I have seen many products over the years, but recently innovation from “Mpowerd” has caught my eye. Recently I conducted some tests on three light systems and was very impressed!

GPS was the first GNSS technology and was funded by the Defence Department in the USA. GPS was a game changer for navigators as not did it provide precise positioning for weapon systems, but it became popular with the marine and aviation sector. Eventually as technology improved, GNSS technology became embedded into many personal devices to enable people to find their way around cities.

It is assumed that miliary drones and missiles will use GNSS to find their way to a target. To help defend countries from incoming attack, GNSS signals are often jammed by transmitters designed to produce strong signals to disrupt the signals transmitted from GNSS satellites.

Recent media reports confirm that GNSS jamming is being used by Russia in cities near to the border with Ukraine as part of its military operations. Further north in the Baltic region, commercial airlines have reported GPS disruption which is presumed to be caused by Russia.

GNSS jamming is also taking place in the Eastern Mediterranean by Israel to disrupt incoming missiles.

Jamming may have a significant impact on commonly used technologies which rely on GNSS systems. In recent weeks, I have heard first hand reports from;

  • Lebanon –Car “sat-nav” systems and mobile phones reporting wrong positions.
  • Gaza – The Emergency Telecoms Cluster have reported significant jamming in the Gaza strip which has frustrated their ability to map services.  

GNSS jamming will have a significant impact beyond frustrating everyone’s ability to navigate around places using google maps or similar. Jamming operations may have a serious impact on safety of aid workers and their ability to run humanitarian operations.

Tracking technologies such as the Garmin products are widely used for fleet management as well as personal tracking, especially in hostile environments. GNSS jamming can reduce the accuracy of tracking technologies or may prevent them working completely.

Satellite telephones such as Inmarsat, Thuraya and BGAN also rely on GPS. Before a call can be made, the internal GPS inside the handset will calculate the user’s position. If the GPS signal is jammed, the satellite telephone may not be usable until a position can be obtained. The only exception is the Iridium range of satellite phones (including PTT). Whilst some of the advanced Iridium handsets may have GPS built in, location information is not needed to make a call. Users should be aware that pressing the red distress button on a Iridium handset may transmit an incorrect location.

In any crisis context, the aid sector may use mapping systems (GIS) to map an area with information about where organisations are operating and what services are being offered to the affected community. In a war context like Gaza or Ukraine, bombing has significantly altered many neighbourhoods and previous streets are unrecognisable. Without functioning GPS systems, it is almost impossible to create new maps.

GPS is also an accurate timing source and used for some IT systems. Whilst not a normal feature in the aid world, GPS timing is used in banking and trading systems to synchronise trading systems. Whilst the risk is low, in theory GPS jamming in the right location could disrupt ATM machines and other banking services.

If GNSS signals are being jammed, there is very little that can be done to restore accuracy of the technology. But if you are experiencing issues with technology that rely on GPS and other GNSS systems, GPSJAM GPS/GNSS Interference Map is a good resource to check if you think your navigation, tracking and telecoms tech is being affected. The data on the site is not real time and only shows average accuracy of GPS for the previous day and further back in time.

Data is sourced from the aviation industry where aircraft systems automatically report data in most parts of the world (See the FAQ section on the website for more detail). Note that information for Ukraine does not exist as there are no commercial flights over the country.

The following screenshot shows an example of jamming which is taking place in the eastern end of the Mediterranean sea.

Over the past 20 years, I have tried out wide range of solar lights. Lanterns with built in solar panels are not new and there are many companies making these products. I have tested many. Some are great, but some are poorly made from cheap materials and break easily.

In a humanitarian context, solar lanterns are very useful in places where power is not available.

NGOs running education programmes will sometimes provide lanterns to students so they can continue study at home. Some of the larger lanterns can also be used to charge up smart phones.

I have covered this topic previously and felt that there would be nothing new to report. Its nice to be proven wrong!

Mpowerd have designed a range of lanterns that can be compressed flat.  In the picture, the smaller green light (Luci outdoor 2.0) is fully compressed. The larger light (Luci Base) to the left has been inflated. This light has a larger battery and a port for charging a mobile phone.

At the back is the Luci Solar String which is suitable for lighting a larger area with its multiple LED light bulbs.

When compressed, the lanterns take up less space than competing products. I estimate that up to 6 Luci outdoor 2.0 units could fit into the same space as one of its main competitors.

Build quality is excellent and these products are likely to last much longer than some of the cheaper alternative brands I have tested in the past. Whilst the inflatable units can be inflated by mouth, I recommend using a pump to prevent the build up of condensation over time. Note: Due to the rugged design of the Luci range, the build up of condensation should not cause any damage.

Over a few evenings during the UK winter, the lights were tested in low and high-power mode. Lights were switched on at 7PM and lasted through the night in low power mode. In high mode, the lights stopped working at around 5AM. In low power mode, the lights lasted through the night.

In the picture below, the Luci Outdoor 2.0 presents (1) a harsh white light which is suitable for reading. The larger Luci Base (2) has a softer tone. The Luci String system base unit (3) provides power for the soft toned string lights (4)

For more information and pricing, please visit Mpowerd via the link below.

MPOWERD | Solar Inflatable Lanterns, String Lights and More – MPOWERD Inc.

How to connect communities

Last month, I raised my concerns about how the Humanitarian Sector are pressing ahead with digital solutions but not considering the connectivity infrastructure needed to facilitate it. In this article, I will shine a light on how to build reliable and safe connectivity services for communities.

Setting up public Wi-Fi in public is complex and good planning is essential. Over the past 10 years, the Emergency Telecoms Cluster (ETC), Nethope and other agencies have provided connectivity in refugee settings across the world. I have supported some of these projects and there have been plenty of lessons the ICT community have learnt along the way.

Here is a high-level view on how to deliver a successful service and get it right first time. 

All successful projects must start with the community and other humanitarian actors in mind. Firstly, if digital programming is to be delivered in the setting, early conversations need to be started so that you have a good knowledge about any ICT4D solutions before meeting with community members. From a system design perspective, it’s important to understand how much bandwidth per person is needed for services to operate correctly.

The next step it to meet up with key local people from the communities to understand what sort of devices people use to access the internet, the type of information and services the communities would like to access and the best locations to locate equipment. Permissions may also need to be sought from key people to install equipment.

An important part of community engagement is to explain which services will be delivered. You should also be clear about what services may not be provided. In most projects, there may not be sufficient bandwidth to support entertainment and it’s likely that streaming platforms such as YouTube may be blocked. Managing expectations correctly must be a priority as if services are over promised, communities will become frustrated. Setting expectation correctly in advance will lead to a smooth implementation. Signage in local languages will also help when services are installed.

The community assessment will give engineers an idea of how many people will access systems at the same time. Headcount information combined with bandwidth requirements for ICT4D services will help technicians to decide how much total bandwidth will be needed to serve the community. User numbers will also be used to identify the correct technologies to source. Generally commercial equipment will be needed to delivery services rather than the cheaper routers used in domestic settings.

This is the part of the project where mistakes made could lead to unintended consequences. Firstly, technicians must ensure that Wi-Fi services are equitable. In some refugee settings, there could be multiple communities, so it’s essential that each community has equal access. Organisations may limit providing services in public locations such as libraries or schools where everyone has access, but even in public areas, care must be taken so that Wi-Fi does not cause issues for other services.

In 2021, Wi-Fi was installed in Mahama Camp, Rwanda to provide connectivity to a library building. On the opposite side of the road was the entrance to the clinic. When the service went live, crowds gathered to use the Wi-Fi and obstructed access to the clinic. Fences were soon installed to keep the entrance clear for patient access. So, key lesson learnt from this experience is to be mindful that Wi-Fi will attract a gathering of people where it could cause inconvenience to other services and activities nearby.

One other important factor to consider when locating IT equipment is the security of these systems. Core technologies should be installed in secure locations with access to stable power and Wi-Fi access points placed in locations where they cannot be easily removed.

As part of the system design, stable power must be provided to run the technology. Many sites will have unstable power of in some cases no power at all. Where there is intermittent power, battery backup systems such as UPS will ensure that services will continue during power outages. Renewable energy systems can also be a great solution for securing stable power. Solar energy is very popular, but space will be needed for panels. For locations where space is limited, wind power could be an option to consider if the local climate is suitable.

Think about how the community will be able to charge their devices. In many settings this may not be a problem at all as small shops within the camps may have solar charging which people can access for a small fee. If there are no means for people to charge technology, it is easy to provide USB charging stations as part of the project.

This is the topic where I get very frustrated with some organisations. Over the years I have seen countless examples where expensive equipment is placed in crisis settings. As communities start to use the Wi-Fi, they start to depend on it for communications and education. After a period (normally a year), the service fails because there is no funding for renewing the contract with the internet provider.

In a refugee camp, where NGOs provide water, shelter, and food, they don’t take all this away with no notice. The same ethos must now be applied to all services including Wi-Fi. For me, the lack of budgeting for longer term provision is a massive failure and, in my opinion, if a project cannot be funded for the longer term, it should not be started at all.

The manufacturers of Wi-fi equipment such as Cisco are massively generous with hardware, but generally do not fund connectivity. The telecoms sector could be doing much more in this space to support such projects by providing free of discounted services. They are often massive multinational organisations with the means to afford supporting humanitarian assistance.

In a nutshell, whilst we can deliver great services at pace, more work is needed on the funding side to keep services running beyond the first year!

Bandwidth to run the Wi-Fi services is some locations can be expensive, especially if satellite technologies are used. With a high volume of users, it will be necessary to limit the number of websites the community can access. As previously mentioned, streaming services such as YouTube are likely to be blocked.

Priority must always be given to essential services which are important to members of the community the project supports. is a good example of a service that will be of great value to refugees. This website has been created in a wide range of languages to provide communities with information about claiming asylum, accessing health care and many more services. This service has been operating in Greece for many years and has been showcased as a major success.

Education Technology (EdTech) is another important service where Wi-Fi systems have been set up to allow students to access to approved education content. In projects where we support a displaced population, the continuation of education for children is vital. Across Ukraine, there are many of examples of such services where educational content approved by the Ministry of Education is delivered every day.

In the assessment phase of a new project, when engaging local communities, also engage government officials such as Health and Education as well so that services can be provided to support these important thematic areas.

Next, but possibly most importantly, Wi-Fi services must be designed and delivered with privacy and security always in mind. In many cases the communities we support will have been displaced because of armed conflict. Alongside the physical wars, there is cyber war taking place. Bad actors (some government sponsored) will attempt to hack IT systems being used to support displaced communities. The activities can range from denying access to services to more sinister attempts to get information about people in the community you are supporting.

Cyber Security and data protection must be included into all projects where services are provided to communities. The threat is very real and there is plenty of evidence supporting this.

As I write this article, war taking place in Gaza where over one million people have been displaced. Before the war, the population in Gaza had regular access to services including Internet.

Famine is building up in Gaza as access for aid agencies is limited. When a ceasefire does take hold, the Aid Sector will need to move in at pace to deliver medical supplies, food, water, and other important non-food items. Alongside these lifesaving services, the community will also need access to digital services and connectivity. As an aid community, are we ready to deliver this?

We need to talk about connectivity for digital programming!

Across the Aid Sector (UN and NGOs) there is amazing progress made in the digital programming space (also known as ICT4D and T4D). Digital applications have a positive impact on communities through various services such as mobile money and cash vouchers, information sharing, feedback services and much more. Despite all these amazing programmes, I am becoming very concerned that many organisations in this sector are just focusing on digital programming and neglecting its enabler which is power and connectivity. In this article, I am going to advance the argument about why infrastructure-based services to communities remain not just important but essential for the success of Digital Programming.

Back in 2015, I was part of the Emergency Telecommunications Cluster (ETC) working group where the ETC 2020 strategy was launched. One objective of ETC2020 was to deliver “Service to Communities”.  Until ETC2020, the organisations focus was to facilitate connectivity and telecoms for aid agencies during major disaster responses. It was recognised that to provide connectivity to affected communities were becoming increasingly important. Around the same time, some organisations including NetHope were already at work in this space and providing connectivity to communities in the refugee camps in Greece.

Using the Nethope Wi-Fi services, IRC and Mercy corps was able to provide the website which informed refugees about humanitarian services and how to navigate complex asylum application processes. Later, this approached become the Signpost project Signpost–Who We Are — Signpost.

In addition to the Wi-Fi and the hosted information, charging points were set up in camps so that refugees could charge up their devices.

Today, the range of digital services to communities continue to increase. In 2022, NGOs and UN agencies provided SIM cards and information for people escaping war in Ukraine in public settings such as railway stations.

Around the world, there are many examples where digital programming is delivering benefits to communities. Frequently, some NGOs neglect to provide the infrastructure and services which are essential to facilitate digital programming. It is often said “This is not our business” or “Somebody else will do it” or “there is 4G everywhere” Unfortunately there are too many examples where these arguments will not stand up. In places where connectivity has not been provided, digital programming has not reached its full potential impact.  

The Gaza war have caused massive infrastructure damage and displaced most of the population. Internet services have been severely impacted. Mobile telephone networks are congested to such an extent that it is difficult to make calls and messages sent by SMS have been reported to take as long as 4 hours to arrive. Before the war started, mobile networks only supported 2G services (voice and SMS only).

Before the War, the Gaza mobile networks were limited to 2G services (voice and SMS) which limited digital programming to SMS based services. There will be a huge piece of work needed to initially provide modern connectivity in public places where aid is delivered. Longer term, the mobile networks need to be repaired and upgraded to modern 4G or 5G.

I want to dispel one myth about e-sims in Gaza.  It was suggested to me that we could facilitate the mass deployment of e-sims to the community to promote access to digital services. Unfortunately, this “Global North” is ill informed and will not help due to the state of telecoms across Gaza.

Typically, in any natural disaster or “hot war” situation, infrastructure breakdown is inevitable. Whilst the aid sector looks to local business to re-establish telecoms, the reality on the ground is very mixed. The telecoms sector in the Philippines strengthened its resilience against volcanic activities, cyclones, and earthquakes. In recent years, there have been various cyclones and other events which has called on the telecoms sector and government to implement their new disaster response plans. The preparedness planning has delivered good results as services are routinely being restored quickly after cyclones and seismic events.

As digital programming becomes more ubiquitous, the aid sector must do more to facilitate connectivity. For Save the Children International, I have recently developed a “Services to Communities” (S2C) approach where our in-house local expertise is used to deliver connectivity to affected populations. Clearly individual NGOs and UN Agencies will not have the capacity to connect a complete population, but they cab provide Wi-Fi hotspots in limited areas such as IDP/Refugee camps, clinics and schools and other public places supported by aid agencies and local partners.

This year (2024), I am running a project to preposition Wi-Fi and connectivity kits in some high-risk countries across the Global South. In my designs, I have included a solar energy module to power the Wi-fi technology and provide charging points for the community. Satellite kits are on standby in 4 regional locations so they can be brought in when a crisis the local internet services are destroyed.

Nethope training in Panama

In addition to pre-positioning equipment, we also need trained people available to deploy the kits and deliver services. Since 2016, some organisations including Save the Children International, Nethope, UNICEF and the ETC have been delivering high quality training to local ICT staff in all regions. The Save the Children “Technology for Emergencies” or (T4E) is a good example of how we have localised deployment of technicians instead of sending people on long flights from more developed countries.

My call to action is for the sector to not forget the importance of delivering the infrastructure needed to provide connectivity. There is no point in focusing 100% on digital solutions if communities do not provide connectivity, power to run the technology or charging points for communities. Digital programming needs to be more open to establishing partnerships with the infrastructure teams that exist in all ICT Departments.

In the months ahead, I hope that the situation in Gaza will become calm so that we in the Aid Sector can get in and do our work. Once we do gain access to the communities, there will be huge needs and it will be vital to provide safe and secure internet access so that the communities can access various services ranging from education to mobile money.

In my next article, I will be taking a deeper dive into the best practices for delivering services to communities (S2C)

Direct to satellite to mobile telecoms

A few years ago, I wrote an article about new project from AST Space Mobile which uses satellites to provide services to standard mobile telephones. Since then, a lot of progress has been made, but whilst satellite-based connectivity for standard mobiles is not yet fully mainstream, new services are getting closer as other tech companies also develop solutions. Based on what I have seen reported over the past year, we will not have to wait much longer for these new services, but availability in the “Global South” will take longer.

In February, The GSMA (trade association for mobile networks) will be holding its annual trade show called the Mobile Word Congress (MWC). In the lead up to MWC many network operators and handset makers use this event to launch new products and services. For 2024 we are already seeing plenty of announcements from the innovators planning direct to satellite to mobile services. More announcements may be in the pipeline, but currently I am aware of four direct mobile to satellite options emerging. Below we will explore four options. Two of these options rely on standard satellite signals and require compatible mobile phones. AST Space Mobile and StarLink use standard LTE/4G technology.

AST Space Mobile was the first operators to go public with to promote direct to satellite LTE services. AST Space Mobile initially secured funding from the Vodafone group which is a multi-national company. Recently they announced further funding from AT&T and Google.

AST tested its Blue Walker-3 satellite in April 2023 with the first call being made from Texas USA to the Rakuten Group in Japan using a standard Samsung S22 Android mobile.

Already, the company has agreements and understandings in place with around 40 partners including Orange, Telefonica and MTN, although it is expected that the first commercial operations are likely to be launched in the USA. The appearance of MTN in the partnership list will be good for Africa as MTN is a major player in many countries throughout Africa. The commercial work with the AST network appears to be the most advanced at this time, but SpaceX with its StarLink network could accelerate its partnership model quickly.

As a long-established satellite telephone operator, Iridium has had a bumpy start in its direct to mobile roadmap when its partnership with the chip maker Qualcomm failed. They have launched Project Stardust which will be based on open standards. Unlike the full LTE functionality offered by AST Space Mobile, Project Stardust uses standard L Band channels and will be initially limited to SOS and SMS text services with data arriving in the future. In addition to smartphones, the service is being designed so that it can be accessed by smaller devices like smart watches. Iridium plans to start testing in 2025 and then launch a commercial service in 2026.

Whilst this innovation is planned to be a direct mobile to satellite service, it does rely on a bespoke chipset being added to each handset. Whilst the tech is “Open sourced” the service will not be LTE/4G compatible but will operate direct on the Iridium L Band. There is an advantage to this approach as it uses the standard Iridium frequency spectrum and will be easier to roll out globally as there are less regulatory challenges.

SpaceX is possibly now the largest operator of satellite globally and well known for its StarLink internet service. Whilst StarLink has focused on its low-cost internet services, in January 2024, they tested services via one of their new and recently launched LTE/4G capable satellites and early testing is showing some great results. Very soon, text services will become available and in 2025, services will be expanded to data and voice. As this service is LTE/4G technology based, services can be accessed by any standard smartphone if the SIM card is registered to a telecoms partner like T-Mobile (USA), Salt (Switzerland). To date, there does not appear to be any global south partners but given that StarLink is operating in many Global South but this could change in the months ahead.  

With the launch of the Apple iPhone 14, Apple included a basic satellite SOS function via the GlobalStar satellite network. The Apple approach is not LTE based, but still useful for people who may be in distress in remote locations. Apple is working with GlobalStar to provide internet access in the future. Whilst the SOS function is very useful, Apple technology is very expensive and the GlobalStar network is limited to the Americas, Europe Australia and parts of Asia and the Middle East. Apple satellite SOS will not work in most parts of Africa.

The answer is a definite yes! But it’s important to weigh up the various options before selecting any of the technologies. The best option to select will depend on where the technology will be used and what functionality will be required. Currently the Apple SOS feature is free of charge but likely to become a pay for service after the end of 2024. This service would be an excellent choice for people who occasionally venture into remote locations in developed countries, but the iPhone 14 or newer is needed. The Apple / GlobalStar service is the only Direct satellite service operating at this time.

Iridium will be a good option as its global and provided by a well-established global satellite network. Whilst Iridium is promoting open standards, unfortunately to access this service, phones need to be sourced that has the compatible technology built in. When available Iridium could be the best alternative to buying a standard satellite telephone. In the International Aid sector, when Iridium goes live, this option could be my first choice for security telecoms.

Finally, we are left with StarLink and AST Space Mobile which is pure direct to satellite LTE/4G services. In a nutshell, these services will be compatible with any standard GSM phone. Based on current announcements from these operators, services can only by bought through a national telecoms partner. From a “Security Telecoms” point of view, for now we must assume that telecoms operators in some countries could be forced to disable services by local governments in the same way as they do for terrestrial networks. So, for now, it’s important not to give up traditional satellite telephones such as Inmarsat, Iridium and Thuraya.

Longer term, it is likely that Direct Satellite to Mobile will get through the various regulatory challenges and could push traditional satellite operators aside. Out of the big three, Iridium is in a good place for the new services, but Inmarsat and Thuraya are unlikely to be offering LTE services from their GEOS fleet and there orbits are significantly higher.

Direct to satellite services (LTE/4G) is a rapidly evolving sector. Apple is the only technology with a live service now and limited to SOS SMS only via a app. As the technology evolves, so will the regulatory challenges. The global mobile sector is massive and well-financed and would object to satellite operators joining the business as competition. Some mobile networks are owned by national government who would impose bans on this new tech if it were to threaten revenues. Currently the emerging business models seem to be based on partnerships where national telecoms providers work with the satellite operators to provide roaming options to extend terrestrial service into remote and rural areas. My gut feeling is that StarLink has the funding and capacity to deliver services at scale and most likely to emerge as the leading direct to satellite LTE/4G provider within the next three years.

Disrupting the Sat-coms Sector

Over the past few weeks, we have seen two major announcements about how we may be able to access satellite communications with a standard mobile telephone. Currently, the four main operators providing satellite telephone services are Inmarsat, Iridium, Thuraya and GlobalStar. To access these networks, a dedicated satellite telephone is required. As these networks are proprietary, it is not possible to use a satellite telephone from one network on any of the other networks.

The big announcements: Recently, Elon Musk announced that the next generation of StarLink Satellites would allow people to make calls using a standard mobile telephone (4G). The service in partnership with T-Mobile where mobile phones could roam onto the satellite service when outside of standard terrestrial services.

The second announcement is from Apple where they stated that the new iPhone 14 would have the capability to send text alerts via its satellite partner GlobalStar. This satellite operator covers around 80% of the global land mass and already has a range of small products to send text alerts for adventurers walking of climbing in remote areas. The Apple/GlobalStar partnership does not provide full voice or data services, but it is sufficient for people to get help in remote areas.

AST Space Mobile:  The concept of using a standard mobile phone with Satellites is not new. AST & Science and its Space Mobile technology came to my attention in 2020 when they announced that they would be launching satellites to provide 4G and 5G services from space. According to AST & Science, they plan to launch a new satellite this month (Sep 2022).

The announcement from StarLink is very significant and with a fleet of a few thousand satellites already in orbit, we know that Elon Musk has the funding to deliver his next generation of satellites at scale. Since the StarLink/T-Mobile announcement, shares have fallen at AST SpaceMobile, but with a large market of over 15 Billion mobiles phones in the world, there is likely to be enough business for all operators in this sector.

How will these new services affect Humanitarian Aid workers? Let me deal with the Apple/GlobalStar partnership first. As a Aid Worker, two way communications is essential so the iPhone 14 text back up service is not going to work for me as the service is limited to sending emergency text messages. Secondly, Apple technology is expensive and we normally use cheaper Android models from a range of manufacturers. Whilst there is definitely a market for the iPhone 14, due to its limited function, I doubt it will gain any form of support in the aid world. Secondly, coverage is poor in some parts of the global south.

I feel more positive about the solutions from StarLink and AST SpaceMobile. Both of these organisations plan to use the 4G and 5G spectrum as used by most standard mobile phones. For aid workers, this will make telecommunications more accessible through standard handsets. However, will this technology work?

Challenges: The traditional satellite operators use different parts of the radio spectrum to provide a service. International agreements enable the four main satellite networks to provide service into most countries. There are some countries where satellite phones are illegal.

Where these new services are permitted, some of us in the aid sector have some question we need to see answers to;

  1. How well with these new services perform when they become available?  How well will an standard handset cope with a satellite a few hundred KM away when its designed to communicate with a base station a few KM from the user?

  2. 4G and 5G spectrum will be crowded. So expect to see some regulatory challenges and perhaps some lawsuits between some of the terrestrial operators. The T-Mobile/Starlink model is probably the best way to success where service users have a contract with a local network provider and the satellite provider fills in the gaps with a roaming service.

I can see a problem with this. What happens when a user is away from his/her home country?  Does the mobile phone connect to the satellite or roam on another terrestrial network?

Conclusion: Whilst the Apple/GlobalStar is not the solution I will buy in to, AST SpaceMobile and StarLink are two interesting technologies to watch closely. These technologies have the potential to disrupt parts of the satellite communications market, especially on land. At sea, its less likely as traditional satellite operators such as Inmarsat and now Iridium form an essential part of the Global Maritime Distress and Safety System as used by ships at sea.  

How to build a Covid-19 clinic in the Global South

In this article, I want to share a design I have created for a “Pop up Covid-19 Vaccination Clinic” Its based on the practices as used by the NHS in England. The design is flexible and can be modified to suit local clinical regulations. My focus is on technical infrastructure. If anyone uses the draft information below, its essential to use these plans as inspiration and involve clinical experts in the final design to be deployed.

The Site Plan
The site has been designed to facilitate the smooth flow of people through a one way system. Markings will be placed on the floor to remind patients to maintain the correct social distance. A waiting area is provided outside where patients can queue and have their temperatures checked. With good planning, appointment slots can be given to patients so that they arrive at specific times to prevent overcrowding at the entrance. Before entry, the temperature of each patient will be checked.

This is a “pop-up” clinic which means that it needs to be built quickly and brought into service. The modules in the plan can either be tents or temporary structures made from local materials such as plastic sheeting and timber.

Once inside the clinic, the patient is registered on the appropriate IT system which is defined by the country government where the vaccinations are taking place. Some vaccines will  require a second future dose, so accurate record keeping and recording of patient contact details will be essential. Screening can also be managed at the registration post. Any patients who fail screening for reasons such as previous reactions to vaccines can be taken out of the clinic via a side entrance.

The next stage is vaccination. This site has been designed to support 8 clinical bays, so it is possible to have a daily throughput of 600+ patients if the clinic is open for 12 hours a day. Good HR planning is essential as there should be sufficient staff to allow for breaks. In hot countries larger teams may be needed as the time staff can spend in full PPE will need to be limited.

After immunisation, an observation area is provided if its needed for the vaccine being administered. In the UK, the Pfizer vaccine is using new techniques and as a precaution, patients will stay in the observation zone for 15 minutes. If there has been no reactions to the vaccine, the patient will be free to leave. Should there be a severe reaction such as anaphylactic shock, the resuscitation module is set up where the patient can be managed.

The whole clinic is secured with fencing. Inside the clinic, there are two restricted areas where access is limited to staff only. A main service area is used to host the pharmacy and staff office / rest room. Another secure area is set up to host power generation and waste management.

Other modules can be added to the staff zone such as wash rooms and PPE storage.

The basic IT will consist of an internet connection such as 3G or satellite for remote areas. Secure Wi-Fi hotspots will be set up for the computers and if resources permit, public Wi-Fi can be used to provide patients with information. The software used to manage patient information will need to be determined locally in each country. Its likely to be a government system.

Clinical Waste Management
Waste from the clinic needs to be handle carefully and responsibly. Firstly items of PPE may need to be incinerated so that the risk of contamination is removed. The empty containers which held the vaccine must either be returned to a formal system to recycle the containers or they must be destroyed. The containers must not fall into public circulation as they may be used by criminal gangs to make money from fake vaccines.

Cold Chain
Vaccines must be stored in a medical standard fridge. The specific model of fridge will be determined by the sort of vaccine in use and its environment requirements. As part of the clinic design, there needs to be stable power available for the fridge with back-ups. If power fails, this might result in temperature levels rising which will destroy the vaccine.  The following design concept should be sufficient to mitigate this risk.

The power source will either come from the local grid or generator. An Inverter/Charger is provided and will charge batteries while power is available. Should the power fail, the battery will take over and power the fridge via the inverter. The battery bank will be sized to provide power for at least 24 hours. The Inverter/Charger has built in IOT technology and will send an alert to let management know that power has failed. (Note – core IT infrastructure will also be connected to the same back up power supply).

As a further protection, a smart temperature sensor can be added to the fridge to monitor temperature and send alerts when temperatures are close to becoming too high or too low.

This is a very high level design. There will be various clinical factors to be added. Other modifications may be needed to make the site accessible to disabled patients. This design is a good starting point for a team of experts to begin work.

Home Worker – Solar Energy Bench Test

In the last edition, I highlighted the challenges faced by home workers who live in locations where the power supply is either intermittent or is not available at all. Since the article was published, I have carried out rigorous testing of two systems. Both performed well. One system was “Out of the box” and the other system was built from separate components I was able to source from suppliers in the UK.

The quickest solution is to buy a solution which has been pre-built. But with the world in lockdown and many flights cancelled it may not be possible to import solutions, so the “Build your own” option may be the only way to provide power. In addition to the home solar designs below, I was recently asked to look at design for 20 staff.

Voltaic Arc 20W Solar Charger Kit: This system is supplied by Planson International. Over the course of four days, I was able to use a Lenovo X390 laptop without needing to connect to the mains power to top up.

The Lenovo X90 uses around 65 Watts when charging which is a little more than the capacity provided from the solar panel, however it only takes just over an hour to charge. Testing was conducted in the UK during sunny weather (early April). During the testing enough energy was created to produce two full laptop charges a day.

It is also possible to get more from a charge by doing the following:

  • Use battery save functions
  • Do not play music – it consumes power
  • Reduce the screen light power (try to work in a shaded place
  • During conference calling, avoid video if possible.

The Voltaic kit also comes with a LED light which is more than sufficient for a desk and a mains charger which can be used to charge the battery in the places where intermittent power is needed.

 100W Home kit (Built from separate components: This system can be built from components purchased locally. In the UK, the following components were purchased from RS components (

  • 100W Solar Panel             $160                                                 
  • 100 A/H Lead acid battery                     $280
  • 100W Inverter                                        $70                                          
  • Controller (10A)                                          $80
  • Cables and plugs                                        $80

For a total of approximately $680, this system is sufficient to power a few LED lights, a laptop, mobile phone and a printer. The purpose of the controller is to manage the power from the solar panel so that the batteries is now overcharged. The controller also has a display which indicates how much charge is being generated, and how much energy is stored in the battery.

This system can also be scaled up. With the same controller, two further 100W panels can be added (Bringing the capacity to 300W or 9A). Additional batteries can also be added to the system to increase storage capacity. Important Note: Energy loads must always be connected to the output terminals on the controller and not to the battery directly. This will prevent the battery from being completely drained (which can cause damage to the battery).

The inverter in this design provides a maximum power of 100W. Larger inverters can be used, but till consume power more quickly. Given that it often takes just over an hour to charge many laptops, I think a single invertor will be sufficient for charging laptops.

For lighting, I would recommend 12V LED lights rather than using 220V lights connected to an inverter. This is more energy efficient as there are often energy losses within most inverters.

Small office Design for 20 users

When designing a system for a small office, it’s important to work out what the overall load is. This will help to identify what components will need to be purchased to build the system. Once the system has been set up, it must only be used for the power load it was designed for. If more items are loaded (e.g. more computers) the system may not have the capacity to support the increased load without upgrading the solar system first.

Many years ago, I developed a tool to estimate the size of a solar energy system. This first screenshot lists the items we need to power.

The load list does not contain items like fridges and security lights as I would recommend buying dedicated stand-alone solutions which have their own panels built-in. Please note that in addition to power load information, it’s important to include information about how long each item is used each day.

The next table calculates the amount of panels and batteries needed to provide the power for the load listed in the first table. The battery bank has been specified to store enough energy for three days. The solar panel has been designed to produce enough energy to service the load and provide a little extra power which can be used to store up energy for three days.

So for this load, if we use 100W solar panels, and 100A/H batteries, we will need 29 Panels and a bank of 50 Batteries. In addition to this, we will need a fairly large controller which can handle 150 Amps. Larger controllers also require bigger cables to transfer the power. In some places higher capacity cables and controllers can be difficult to source, but it’s possible to achieve the same result by building three smaller 50A systems.

As a very rough guide, here is an estimate for building the system above.

  • 150 Amp controller:        $2300
  • 50 Batteries:                      $1250
  • 29 Panels                             $4640
  • Inverter 2KW                     $1000
  • Cables etc                            $1500
    Total                                    $10,690

This is a rough estimate which does not include shipping and taxes. Whilst the initial set up cost is high, over time, the overall return on investment is good as no fuel is used. The only component which needs to be changed from time to time is the batteries.

Conclusion: Solar energy can be a great solution for remote locations where power is not available. It is also a cleaner method of providing energy. With some good planning, selection of low powered office systems, smaller power systems can be purchased. Good discipline is also important as it’s easy for a system to fail to deliver if extra load is added which was not planned for.  

Solving the energy problem for the Covid-19 homeworker in an off-grid enviorment

Over the next few weeks, I am going to be focusing on technical topics which can may assist the aid sector in our fight against COVID-19. So let me kick off this series of articles by addressing one important issue that affects aid workers living in the “Global South”  – lack of stable electricity at home.  There are still vast sectors within cities where reliable power is absent. Over the past week, I have had heard that staff need to send laptops back to the office to be charged so that they can continue working. This requires staff to move around and mix, which goes against our practice of trying to stop this virus from spreading.

So here are a few ideas to overcome the problem.

  1. Power consumption: Before we explore any power solutions, it’s important to have an understanding about how much power is needed to keep a staff member productive. Effectively we need to create a power budget. The idea here is to strip consumption back to the absolute minimum so we only need to build the smallest power solution and keep costs down.

    So basic needs are:
    1. Laptop computer (Without the external monitor as this uses extra power!)
    1. Mobile phone and or a 3G/4G hotspot
    1. LED light for the workplace.

Mobile phones and most 3 or 4G hotspots / dongles can be powered directly from the computer (but will reduce computer battery life). Portable solar powered lanterns can also be a great solution for the home office. If possible, try to avoid using additional technology like printers, scanners and external monitors if you can.  The camera on a mobile phone can be used as a very effective scanner with apps like Microsoft Office Lens. Online electronic signature software should also be used by organisations so that there is no need to print hard copies for signature.

Interesting Fact: During the Ebola Crisis, medical staff did everything online – including prescriptions as paper is a great medium to pass a dangerous virus between people!

The laptop is the main consumer of power so we need to understand how much power the laptop will use so that we can size up the correct power solution. Our target should be to provide enough power so that the computer will run for at least 8 hours. Ideally I would like to cap it at 8 hours in the interest of work-life balance, but some people may need to work for longer periods, especially if they are supporting the COVID-19 response.

The basic rule of physics which applies here is this:  Larger power needs require larger power systems, hence more costs!  So one quick win here is to look at the office computer estate and if possible, carry out some temporary redistribution so that the more power thirsty computers are allocated to staff that live in places where there is stable power. This will leave the more efficient computers for the people who live in the off grid areas.

So here are some power budgets for two popular computers used by NGOs

Model Lenovo X390 Lenovo T470p
Power adapter 65W 90W
Battery Capacity 48 WH 48 WH
Battery Life 3.8 hours 2.5 hours
  • Battery life is subjective to how the computer is being used.

The X series computer would be a better computer for a power starved setting it needs less energy (65W) to charge, and the battery last longer.

TIP: Computer battery life can be extended by using power save function on the laptop, and by installing larger battery packs (But larger batteries take longer to charge!)

  • Unreliable power scenario: In this scenario, we will look at what solution we could use to keep a laptop running in a location where power supplies are intermittent. The following solution is fairly cost effective.

This system is designed to use the grid power when its available  to charge up  battery. The charger should be high power, at least 30 Amps or more charging capacity. The cheaper low power charges would take too long to fully charge! The battery should be around 120 AH or more so that it can run an inverter for up to 10 hours. A 100W inverter should be sufficient to charge a laptop. Many inverters have a USB charging point built in which can power a phone.

Safety first: When a battery is charging, it can produce hydrogen gas which is explosive. So lead acid batteries should be placed in well-ventilated and away from any naked fames (such as a cooker).  

As this system supplies 220V AC, be careful that others in the house do not plug things into the system and steal your electricity!

  • No Power: For locations which is completely off grid, here is a design which should be sufficient to keep the technology running for a homeworker.

The 120W panel at peak will produce 120W energy, but it could be less when the sun is weaker earlier or later in the day. So the idea is to locate the panel in a place where it will get the maximum sunlight. The controller uses the power from the panel to charge the battery. The 20A specification means that the system can be scaled up to two panels if more capacity is needed.  

During the day, the system should produce  600-800 w/h of energy which is more than sufficient to run a computer, LED light and a charge a mobile phone.

Safety first: When a battery is charging, it can produce hydrogen gas which is explosive. So lead acid batteries should be placed in well-ventilated and away from any naked fames (such as a cooker).  

On cloudy days, solar panels will still produce some electricity but not as much as on a sunny day. As a temporary solution, this design with connecting cables should cost no more than $650. For longer term use, I would recommend doubling up on the panel and battery as it will store more power and will keep a laptop working for more than a couple of days during bad weather.

TIP: The inverter I have specified can be plugged into a car, so this is a good back up plan should energy stored in the two systems above run out of energy. Make sure that the engine is running (out in the open!) when an inverter is in use.

Conclusion: These two solutions are a “Quick and Dirty” design. It utilises components which are readily available from online retailers or hardware stores. Components can be sourced easily in global south countries. The design can also be fine-tuned to match specific power requirements by people who have experience in this field.

Solving the energy storage problem

Around the world, the use of wind and solar farms is increasing as the efficiency of panels and wind generators increase and production costs fall. In the global north, the large renewable energy plants feed power into national grid systems which means that we can reduce our carbon impact by using the gas, coal and oil fired power stations less. But, there is an issue. Whilst we can produce plenty of energy from energy farms, we have not yet really figured out how to store the megawatts of energy efficiently. During a cycle ride from London to Gibraltar last year, I passed through many wind farms in France and Spain. On windy days, there were times when only have of the turbines were not turning. So whilst we have all of this infrastructure, the maximum benefit cannot be realised as we do not have the technology to facilitate the mass storage of electricity.

For smaller solar systems like the ones we would use in the aid sector to power a school or a clinic, the same power storage problem exists. In most countries we store energy in lead acid batteries or some other variant. The use of Solar systems can reduce the carbon footprint, but at the same time we are producing a lot of environmental waste which is not good for the communities where the aid sector is meant to be “doing no harm”

The Battery Problem: The most widely used component in a small solar system is the lead acid battery. The same battery as we use in vehicles. In a vehicle, the lead acid battery normally lasts up to three years which is its expected life. This is because the engine will keep the battery charged. In a solar system the lead acid battery may last two years if we are lucky. Unlike in a vehicle, the battery will frequently be drained to less than 50% of its capacity. If this happens often, longer term damage occurs in the battery cells. The hot climates where NGOs operate also has a negative effect on the battery shortening its life further.

Given the short life span of lead acid batteries, the by-product of the green solar systems is a lot of toxic waste. This is can be a massive issue in developing countries where they are not geared up for recycling.

Lithium Ion batteries are often seen as a good alternative to lead acid batteries because they can hold more charge. But there are some major disadvantages. Lithium Ion batteries contain some very toxic chemicals and have a troubled history of catching on fire. Most airlines will not transport larger lithium batteries due to the fire risk.

An alternative approach:  There are better technologies and in the future we will see innovation come from the automotive industry.  Tesla and other manufactures of electric cars are working hard to develop battery technology which will allow electric vehicles to go much further than they can today. As vehicles move from petrol and diesel to electric, the mass production of new battery technologies will bring the cost of energy storage down. Where the automotive industry produces answers for energy storage, in the aid sector, we will be able to take advantage of new battery technology for our solar systems.

We have been using electric vehicles for many years and there are already battery technology we can use now to make our solar systems more sustainable. Schneider Electric uses Nickel Sodium batteries to store energy in its Vilaya range of solar systems.  

The Nickel Sodium battery (This example made by FZSoNICK) works in an interesting way. The battery need to be warmed up so that the salt inside melts. Once the battery is at its operating temperature, energy can be stored and discharged as needed. The lifetime of the battery is 13-15 years. At the end of life, the waste product is a block of salt and some associated electronics.

The FZSoNICK has the ability to store 10 KWH of energy in this single unit. The cost for one battery is roughly $10,000. It’s a big upfront costs, but there is a return on investment over time. So let’s take a closer look at the numbers.

A good quality deep cycle 90AH battery will cost around $(US)250 and can store 1KWH of energy. Taking in account that we don’t want to discharge a lead acid battery than 40%, then we need to buy more batteries than the stated capacity to ensure that we can store and use the 10KW without damaging the battery bank. So for this example, we would need to buy 14 lead acid batteries at a cost of  $3500. As the lifetime of the battery is likely to be two years or less, over the course of 12 years, we would need to change the batteries 6 times, which comes to a total of $21,000 or more. This does not include other costs such as installation and shipping.  

So whilst there is a higher start-up costs, the return on investment is significant. But there are other advantages which you will see in the following summary:

  • Cheaper to run over a long period
  • Batteries take up much less space which means less cables are needed for installation.
  • No fire risk from gasses such as hydrogen
  • End of life waste is smaller as this technology does not use as many materials as other batteries. The main waste product is a block of salt.
  • Battery is stable and safe to transport
  • Good return on investment

Deployment example: Nickel Sodium batteries are built into complete systems such as the Villaya solar system from Schneider Electric. The solar plant is transportable with electronics installed and fixed to the walls of the an ISO container. This approach is great for disaster preparedness due to its mobility.  

This system can produce enough power to run a small office. In addition to the power circuits, communications technology can also be fitted inside the container so that internet connectivity can be provided in addition to electricity.

The Villaya system is designed with the appropriate systems to protect the circuits from lightning, which means that this system is very well suitable for topical and sub-tropical locations such as Africa.

For disaster response, where it may be difficult to move a container quickly, it’s possible to design the same system into other  formats which can be broken down to smaller shipping units for future assembly at a disaster site.

Sustainability: We know that solar panels have a long life if looked after. Nickel Sodium Batteries also have a long life which means that a system built on this technology will be sustainable, but technology is not the only area we need to make sustainable. We need to build peoples capacity. As we adopt new sustainable technology, with built in monitoring systems, these systems will become more complex. We need to initiate a training programme to build the skills of the people who will source, install and maintain solar energy systems.  NGOs will have a massive role to play here as they work in very remote locations. If they can adopt green energy systems instead of generators, other sectors might do the same.

In my next article, I will discuss how we in the aid sector should establish teams within our organisations to take ownership of environmental affairs and build the skills in house to help reduce the carbon footprint.