Updates on my Final Year Project

How’s my final year project going? I know it’s crazy, 2 weeks away from the submission deadline, I am writing a blog post about it. But I am travelling, so I have no Internet access and I felt like writing.

Screen Shot 2016-04-08 at 10.21.21

My report already exceeds 20,000 words and I am extremely happy with the literature review section. I am discussing  the current state of higher education in the sub-Saharan Africa region, as well as initiatives that support its development. I also go over the state of Internet availability, costs and the overseas Internet infrastructure that links Africa with the rest of the world. A Kenyan start-up initiative stands out on this topic, called Brck, which created a portable Internet router and the Brck Education platform. I have been surprised to see the great interest, both local and international, in supporting the development and access to technology in this region. The most notable initiatives would clearly be the annual e-learning Africa Conference, USAID and the Association for Affordable Internet.

Here is also an amazing TED talk from Juliana Rotich on Internet connectivity in Africa. It’s was presented in 2013 in Edinburgh.

Recent reports show that the most used Internet connection now is 2G, with a few countries such as Nigeria or Kenya adopting 3G. Also costs are not to be neglected, as it may be the case that sometimes the fee for an Internet subscription can reach 90 to 100 percent of a family’s monthly income.

I have also discussed the use of online learning technologies used in African Universities and other forms of higher education. It was clear to see that African students benefit widely from this type of new connectivity, which helps them communicate with other students or lecturers beyond their local area, explore opportunities, exchange ideas and explore new ways of accessing content.

My research then goes into online video streaming for slow Internet connections. An overview of the media streaming process is included to offer a bit of context. Then the report goes into more detail on adaptive bitrate streaming over HTTP and how it is designed to react to adverse conditions. I have been lucky to find a wide range of research papers on the topic, with experts discussing all sorts of new methods and algorithms to make the video segmentation and streaming process as seamless as possible. It is no news that in a decade, video streaming will make up approximately 80% of all Internet usage. So interest is high from all companies that revolve around Internet use. We can see how services like Netflix and YouTube, as well as Facebook’s live video and video content gain more users day by day.

The protocol I selected for my project, DASH (Dynamic Adaptive Streaming over HTTP), is an open-standard developed by the MPEG consortium with the support of major technology companies, such as Microsoft and Adobe. They all form the DASH Industry Forum. The need for a standard was long due. It is based on HTML5 and the Media Source Extension, which makes it available in the vast majority of web browsers. We all remember the Adobe Flash Player, needed a couple of years ago (and still used on some websites, I won’t name them!) in order to view video content. Oh and let’s not forget Microsoft’s Silverlight. Many alternatives which only annoy the end-user and offer more downsides than advantages.

I also did a bit of research into CDN and video encoding providers and I decided to go with the Amazon Web Services Cloudfront and Bitcodin. They work really well together, as all my encodings are directly transferred to my S3 bucket. I wrote more on this topic in this post.

The development stage of my project’s prototype is now complete and I am currently testing it with University students, including some from the Nigerian student society at my University. I am super happy to receive their feedback and see how well I’ve managed to address the issues. Oh, I’ll also borrow some devices this week from our technical support desk so I could test the streaming implementation on real feature phones and tablets with slow CPUs.

Overall, I couldn’t be happier with the topic I have chosen and I can see this from the amount of work I am able to produce for this project, compared to other modules. As I have said, just 2 more weeks until I submit my report and then I’ll have to present it by mid-May in front of my supervisor and examiner.