Bachelor Thesis
Analysis of Transmission Speeds Across Different Link Types
With the rapid increase in internet connected devices it's become more and more important to use the best link type for the application. The network link types Ethernet, Wi-Fi and 5G have their own characteristic attributes which gives them their different uses and qualities. The aims of this report is to analyze the download speeds that these link types enable in a network composed of IoT devices, and to see how well the link types scale as the network grows larger.
Experimentation with the devices occurred in incremental steps by first testing 4 devices, then 16 devices, then 32 devices and so forth up to 73 devices. The result showed that using Ethernet to transfer data in a network would be the fastest option and that 5G was the best option if wireless connectivity is needed. This result confirms the notion that using Ethernet over other link types for connectivity in a demanding network is the preferred option. However, due to limits regarding the technology used it was not possible to test mmWave-5G or a pure 5G network. The work developed during this thesis can be expanded to a distributed computing setting and can allow for the establishment of computational heavy applications such as Federated Learning algorithms.
