Affordable, faster, and reliable. Those are the three main features that customers expect with an Internet provider. Following the lockdown caused by the pandemic, those features became a basic requirement for the world to keep turning round. The South African mobile operator Rain, which wasn’t able to sustain its rapid growth, has given a disappointing quality of service to their customers, and is now facing a major customers’ backlash.
The lockdown was a challenging situation for most mobile operators. As people stayed home, their internet usage peaked. MyBroadband pointed out that most mobile carriers didn’t face issues with the quality of their internet service thanks to an additional spectrum temporarily provided by the Independent Communications Authority Of South Africa (ICASA), except for one mobile provider: Rain.
Founded in 2017, Rain started business in South Africa with a data-only mobile service, in a country where the mobile market is highly-concentrated.
Instead of focusing on the technical challenges, Rain’s team focused on stretching its marketing muscle to “democratize” mobile connectivity in South Africa and quickly acquire new market shares. Rain had hired the entrepreneur Will Roos to lead the company that promised uncapped services to its new users.
Unfortunately, Rain’s service has deteriorated year on year. Despite the fact they were the first carrier to bring 5G to South African homes, they failed to optimize their QoS along the way, paving the way to a widely clogged mobile network.
Rain has been ranked the worst Internet provider in South Africa by MyBroadband. In the 2020 Mobile Network Quality Report, the mobile network analyst company pointed out that Rain has been the only company whose QoS went in a downward spiral since its launch.
For MyBroadband, Rain’s main issue is oversubscription. The company tried to provide service to more customers than its capacity allowed. Rain quadrupled sign-ups after the lockdown. As a consequence, data demand increased multifold and Rain’s network performance dropped dramatically. Promising an uncapped service definitely contributed to the crash.
Specialists in QoS know this kind of issue. A collaborative research published in the International Journal of Scientific Research in Computer Science, Engineering and Information Technology (IJSRCSEIT) pointed out that poor QoS perceived by mobile users is a consequence of the rapid growth of network links and network services. The paper pointed out that
“Data network quality of service (QoS) degrades over time when networks cannot keep up with the growing demand for the network resources.”
Researchers Folasade Ayankoya, Olubukola Ajayi and Blaise Ohwo
Rain’s quality issues are common to South African telecom companies. Sentech tried a similar approach to Rain with an uncapped service delivered by its subsidiary MyWireless in 2009. The service was a big success at first, but it quickly led to its death. Network congestion rapidly increased and consumers bitterly complained. In February 2010, Sentech discontinued its service, only one year after it launched.
Network congestion is a common problem that companies can handle with a proper QoS optimizer. QoS level is crucial in mobile broadband multimedia networks. Requirements of multimedia applications are strict and different issues can influence QoS in the network such as latency, jitter, bandwidth, and packet loss. Companies need to incorporate in their working process tools that allow them to identify or prevent problems in their network connections, and allow them to take more advantage of their mobile broadband.
Recent studies show promising results. A team of researchers in computer science of Babcock University (Nigeria) reviewed existing content delivery network models to understand their architecture and operation to develop an optimized model that would integrate into a Long-Term Evolution (LTE) network. The team evaluated the model with a Network Simulator and QoS metrics, showing that the optimized model can improve the Quality of Service of mobile broadband. According to the study, a cloud-based content delivery network can act as a QoS optimizer improving the experience of their subscribers.
About QoS optimizer, a study of the Department of Computer Science & Engineering at Chungnam National University found that the right QoS policy can provide various communication capabilities. However, this requires a proper setup, a rather complicated task, even for the specialists. The study suggested that a QoS Optimizer can find the best combination of QoS policies after collecting information of Data Distribution Service (DDS) nodes. This way, the optimal QoS Profile can be quickly sketched, allowing for improvement in the performance of data communication
That’s the biggest challenge for internet providers: identify the solution that best resolves its problem, a task that requires data collection and testing. For instance, a multinational study led by teams from Pakistan, France, China, India, and South Korea optimized QoS for improving mobile edge computing-based healthcare applications, comparing different methods in a 5G network. For the specific case of telesurgery, the window-based Rate Control Algorithm (w-RCA) was the best QoS optimizer, outperforming other solutions.
However, for multimedia networks, researchers found that the best solution depends on the case. In a study published in the journal of the Institute of Electrical and Electronics Engineers (the world’s largest technical professional non-profit organization), researchers conducted performance measurements involving several scenarios to identify the most suitable scenario to provide qualitative services and to efficiently utilize resources given the conditions imposed by each traffic type. In some cases, Custom Queuing underperformed; in other cases, Priority Queuing was the winner in the tests
All the studies highlighted the importance of higher QoS management for delivering more reliable connections. These kinds of services can be scaled up for optimizing QoS of mobile operators in South Africa, leading to a qualitative improvement in the network performance of mobile operators.