Three short stories on today’s Mobile Networks Performance

Ensuring the quality of the networks for an optimal end user experience is often a challenging task for mobile network operators. While the carriers’ engineers adjust the systems for getting the most efficient usage according to the load required, you might be affecting the quality of the subscribers’ service in particular conditions, subject to the applications being used by them, the coverage and access technologies available in determined locations, or even the non-always optimal policies used for access technology selection.

Evolved QoE – Application Performance who?

Nowadays delivering quality services to the mobile subscribers has evolved beyond the traditional network availability and quality. Today’s users are demanding sufficient performance for each type of application used, leading to profile based modelling of the traffic and increasing the complexity of the Quality of Experience (QoE) evaluation for the carriers. For the operators evaluating the QoE is hard, as published by the GSA and Ericsson this month (here) “A 2012 study from the University of Massachusetts Amherst and Akamai Technologies found that internet users start abandoning attempts to view online videos if they do not load properly within two seconds. As time goes on, the rate at which viewers give up on a given video increases”, “with the rise of mobile-broadband and smartphone usage over the past few years, the meaning of user experience has changed dramatically”.


What used to be measured with coverage and bandwidth capacity is now extended to performance per application and end user experience, involving signal coverage maps, latency analysis, QoS, security features, loading speed for web pages or online multimedia content (e.g. HD video) and apps, among others. As explained and exemplified in a recent Ericsson Whitepaper on Network Performance (here) “Network performance cannot be generalized because the only true measurement of performance is the experience of those who use it.”, “App coverage is one way we describe this performance. It refers to the proportion of a network’s coverage that has sufficient ability to run a particular app at an acceptable quality level. For example, an area that has 95 percent coverage for voice calls, may have 70 percent coverage for streaming music and only 20 percent coverage for streaming HD video. A consumer’s usage patterns will determine their preferred type of coverage”


Indoor small cells – Please mind the gap between the macro and small cells platforms

Evolved small cells for indoor installations are coming to fill the coverage gap between the macro networks (i.e. 4G/LTE, 3G, 2G, etc.) and the small cells technologies (i.e. Pico and Femto cells, etc.). A new solution was recently announced by Ericsson called Radio Dot System (here), which is according to them “The most cost-effective, no-compromise solution to indoor coverage challenges”. It is well known the operators have challenges for covering indoor areas and buildings on a cost effective manner, while more than 70% of the traffic is generated in this domain. The solution is ultra-small, light, scalable, with fast deployment, and relies on Ethernet connection for integrating with the existing mobile network.

Although Ericsson’ solution should not be available before next year, we would expect to see other similar solutions in the market in the near future. This trend would potentially look to take over part of the current usage being done on WiFi technologies, preferred by most of the users for indoor communications.


Smart access network selection – The seamless cellular and WiFi access marriage

A recent report from 4G Americas (here) analyses the role of the WiFi technology in current mobile data services, and the methods for overcoming the challenges appearing as a result of the integration and mobility between cellular technologies and the WiFi. As stated by them “with smartphone adoption continuing to rise and the increasing prevalence of bandwidth-intensive services such as streaming video, the limited licensed spectrum resources of existing cellular networks are as constrained as ever. Wi-Fi, and its associated unlicensed spectrum, presents an attractive option for mobile operators – but improved Wi-Fi/cellular network interworking is needed for carriers to make optimal use of Wi-Fi.”


The so-called interworking between traditional mobile access technologies and the WiFi networks must be seamless and transparent to the end users. In such way, the service continuity must be assured when a subscriber moves in example from 4G/LTE coverage to WiFi covered zones and back, using methods like an automatic offload policy. Different methods are currently used for this interworking like session continuity, or client-based mobility, or network-based mobility. One of the most popular and accepted, also standardized by the 3GPP, is the network-based Access Network Discovery and Selection Function (ANDSF), which is already supported by most of the WiFi devices and network elements, including Policy Managers and specific network gateways. Other innovations have been made available for addressing the seamless interworking issues, in standards like the Hotspots 2.0, or the seamless SIM-based authentication.


As it was commented in my previous post “The top 10 fast facts you should know about LTE today”, the 5G will be a combination of access technologies for jointly fulfilling the requirements of the future. In these scenarios the seamless network selection and mobility becomes even more important beyond the classical offload scenarios, and some particular issues for these are commented by 4G Americas and vendors like Ericsson. These issues include: Premature WiFi selection (access technology shifted when coverage is still too weak due to distance), Unhealthy choices (traffic offloaded to systems overloaded), Lower capabilities (offload to alternative technologies having less performing networks), or Ping-pong effects (frequent access technology shifting due to mobility affecting the QoE).

A. Rodriguez