By John Giere
One of the primary reasons that more and more mobile operators are rolling out their 4G networks is to accommodate the growing demand for video content, which is expected to comprise 75 percent of all mobile network traffic within the next few years. Even the most bullish of predictions about mobile data and smartphone growth have fallen short of today’s reality. So here’s my prediction: Mobile data, especially video, will outgrow 4G networks sooner than anyone would care to admit. According to a Yankee Group report, by 2014, there will be more than 1.4 billion smartphone subscribers, up from over 600 million in 2010.
This new 4G world will be different from previous technology upgrade cycles because of the abundance of over-the-top (non-operator) services that will use 4G networks for only the cost of access. Building more infrastructure capacity cannot be a complete solution if operators are to manage this growth and expect to turn a profit. So how do operators build a sustainable business model in this new 4G world?
If operators do not manage all the different types of traffic at various points of network control – from video optimization, to operational and marketing analytics, to policy management – their ability to offer subscribers a glitch-free video experience could be seriously compromised. Throwing money at the problem, such as costly network infrastructure expansion, is not necessarily the answer either. Not only can few operators do this and still remain in business, ultimately they will be left with the same problem as before because bigger waves of data are coming (HD video, Machine-to-Machine communication to name a few).
To better meet the demand for a quality mobile video viewing experience, a more immediate approach would be to take a combined software and hardware approach to manage a variety of traffic types in real time. Quality mobile video viewing experiences can be enabled through “self-optimizing" techniques: solutions that automatically trigger optimization only when the network conditions reach certain congestion thresholds that cause video quality to degrade. Rather than compressing all of the traffic all of the time, which is expensive and can stifle networks, optimization is triggered only when there is a video quality issue caused by too much data traffic.
Operators all over the world are looking for smarter optimization techniques that enhance the user experience. They do not want to touch data traffic unless there is a problem, and for many, their problems pertain to a specific region, a specific user type, or a specific type of long-form video traffic. Understanding these patterns allows for strategic deployments of congestion-aware optimization at the busiest cell sites and activated at peak times. Intelligent caching – a technique that prevents additional bandwidth consumption every time a popular video is requested – is another strategy that could work for many operators. Caching the most frequently viewed videos closer to the network edge results in faster download and playback, and prevents wasted bandwidth that occurs from retrieving the same video repeatedly from the server of origin.
