The massive increase in mobile data usage, driven by video traffic, 3.5G dongles and smartphones, has created a huge network load that just keeps growing. Upgrading to LTE and beefing up the backhaul can provide bandwidth relief but it requires a large investment, both in terms of money and time. Network operators are faced with difficult choices of how to ensure a quality user experience for bandwidth hungry subscribers while keeping expenses for network expansions in line with revenues.
The Broadband Traffic Explosion
Broadband traffic is growing dynamically, led by the growing use of smartphones. A Forrester Research report released earlier this year predicts that more than a third of consumers in Western Europe will access the Internet from their mobile phones by 2014.
Leading the way is video traffic. As reported by Cisco, video is expected to increase by more than 100 times by 2013 and account for roughly 64 percent of all mobile data traffic, around 1.3 million TB per month.
Meanwhile, the increase in network load is taking a toll on the user experience. The Broadband Expert website reported that operators deliver just 24 percent of the advertised download rate, according to a survey examining U.K. mobile broadband speeds.
Heavy load on cells, backhaul and core network on the busy hours creates uncontrolled degradation of user satisfaction. Customers who are paying for high speed Internet connectivity are facing slow web browsing, frozen videos and long download times.
Solutions for managing the data Tsunami
One possible solution to meet data traffic turmoil is to increase network capacity. Network upgrades bring some relief to operators, but they can be expensive and time-consuming. In addition, when a move to a different network technology, such as LTE, is on the horizon, increasing the capacity of tomorrow's legacy network may not be a wise investment.
The adoption of 4G technologies such as LTE (3GPP Long Term Evolution) and WiMAX will enable substantially higher throughput on backhaul and radio. However, deployment of LTE has been delayed due to the complexity of roaming across different LTE networks, and unclear deployment strategies and equipment costs. Most operators prefer to "sit on the fence" and wait while getting the most out of their existing HSPA technologies.
Another solution is to impose limits on subscriber bandwidth consumption, often combined with additional fees for "network hogs" that exceed those limits (typically about 2 percent – 5 percent of all subscribers). Bandwidth caps and throttling can mitigate the problem of bandwidth hogs, but they do not address the huge amount of data generated by average customers who live within the fair-use limitation.
Bandwidth shaping can be used to protect the network from bandwidth demanding applications; however, it can also affect the quality of service for popular applications such as video streaming. A degradation in the quality of the service can deter current and prospective subscribers and can raise criticism from proponents of network neutrality.
A cost-effective and more immediate method for reducing network load is data optimization, which reduces transmission bandwidth requirements on the radio, wireless backhaul and IP transit. These solutions accelerate mobile browsing, with an up to 60 percent reduction in network traffic on the backhaul and radio networks – depending on the level of compression of images and video.
Optimization solutions can be differentiated, personalized and adapted to the changing network conditions, with policies for busy hours and in specific areas of congestion. By following the following five key factors for optimization, operators can achieve optimal results.
• Optimize all major types of traffic and protocols: In order to provide operators with concrete benefits, it is recommended to optimize all major and emerging traffic types, such as video streaming, Flash video, P2P traffic, etc.
• Optimization tuned by peak-hour congestion: The peak-hour represents a challenge to deliver smooth browsing to users of heavy multimedia applications while ensuring that there is sufficient bandwidth to meet all subscribers' needs. Optimization techniques can be based on the level of traffic congestion to maximize network efficiency and avoid service degradation.
• Cell-based data optimization: Cell-Based Data Optimization uses dynamic policies to ensure the most relevant optimization technique is applied to each cell based on local traffic patterns such as busy hours or increased activity due to sporting events or concerts. By varying the level of network optimization, the network is used more efficiently while ensuring the best possible user experience for all subscribers.
• Data optimization with fair use controls and service personalization: The ability to personalize the optimization and traffic management services and align programs with differentiated tariff plan strategies provides subscribers with fair use and controls. The solution should be policy-based and sufficiently flexible to attach both different optimization techniques and different pricing levels based on tariff plans, devices, subscriber preferences, applications and congestion levels.
• Single-proxy optimization solution: Optimization solution should include the necessary flexibility and depth of features. Partial solutions force operators to deploy multiple proxies, increasing latency, management complexity and increasing capital and operating costs.
When implemented using all five factors, optimization can provide the bandwidth operators need with the network availability that subscribers demand, both now and in the future.
Merav Bahat is vice president of marketing for Flash Networks.