No abstract

In this paper, we quantify reordering and connection interruption while downloading data to mobile phones in modern HSPA/HSPA+ mobile networks based on end-to-end measurements conducted in commercial operator networks. We show how these events are affected by phone model, operator, time of the day, user movement and signal strength. We find that phone model, time of day and user movement affect the number of connection interruptions, but not their length. Data loss during interruptions is influenced by operator and time of day. The number of reordering events is impacted by the operator and the time of day, the operator additionally affects the reordering extent. This improves the understanding of HSPA/HSPA+ networks and enables further research to improve the data transfer and methods for delay spike prevention.

A significant part of the data in mobile networks is transferred as bulk data with transmission control protocol (TCP) as apps or video downloads. When the video takes too long to start, users are more prone to abandon watching, which eventually leads to decreased revenue for the content provider. While it is widely known that TCP has performance issues in mobile networks, end-to-end measurement studies, especially based on real data, should be studied further.In this paper, we measure the efficiency of TCP in long term evolution (LTE) networks and provide an analysis on the reasons of unoptimal performance based on 235 000 measurements from deployed mobile networks. For this purpose, we propose an algorithm for LTE networks that detects the periods during a TCP connection where the path is saturated, ie, the network is the limiting factor. From this data, we find that TCP is a source for unoptimal performance, and assumed reasons from other papers are partly confirmed and partly refuted. Most importantly, we find that the amount of queueing on the path has significant impact on the achieved protocol performance. Lastly, we learn from the LTE experience and put the findings into a 5G context.

No abstract

Data transfer in mobile networks has increased significantly over the past years, and the capacity of these networks has grown accordingly. Due to their different set of characteristics compared to wired networks, they have also received attention from end-to-end developers on transport and application level. However, there exists no research on packet reordering in modern mobile networks, and how a transport layer should be designed to cope with it. In this paper we discuss how to measure packet reordering in mobile networks with the help of TCP. Our analysis in Finnish mobile networks shows that reordering characteristics depend heavily on the operator, but is present in all tested networks, and negatively impacts TCP when not handled. We find that both number of reorderings in a connection and the extent of the reordering is influenced by the sending rate. Additionally, we present the framework for a reordering prevention algorithm which takes into account the findings by automatically adapting to the current sending rate.