By Michelle Amodio
The explosive growth of smart mobile devices capable of making video calls is placing intense pressures on service providers to configure their networks to assure users have experiences that meet or exceed their expectations. However, providing a superior quality of experience (QoE) is a complex challenge.
As part of its support of mobile service providers around the world, Alcatel Lucent’s Bell Labs recently looked at the factors that influence QoE) to help service providers full understand all of the issues and what needs to be done to resolve them.
The questions the Bell Labs researchers explored were:
- What are the key factors that influence QoE for mobile video calls?
- Are residential broadband networks ready for them?
There was good news on both fronts. The challenges could be identified and categorized in terms of importance, and viable solutions can help operators meet consumer expectations in a timely and cost effective manner, as well as lay the foundation for new profitable services.
The main finding of the researchers was that service providers require not bigger (more bandwidth) networks but faster ones (in terms of response time). They validated that speed, combined with the proper provision of a consistent quality of service (QoS), are the critical determinants in mobile video calling QoE.
Users bring expectations to all types of mobile video. They are based consuming and interacting with video on wired networks, particularly at home on their televisions, PCs and now smartphones and tablet computers. The trick for mobile operators will be in replicating (at a minimum) today’s home video calling wired experience on services like Skype, and hopefully exceeding them as a current Apple iPhone ad implies.
However, to date there have been a number of limiting factors that have made real-time video calling on mobile devices problematic. This is especially true on older mobile networks which were optimized for voice communications and not packetized multimedia ones which things like 4G LTE and Super WiFi are architected for from the start.
Bell Labs identifies the pain points
As noted in a posting by Kyung Mun, Corporate Technology Strategist, Alcatel-Lucent, there are a variety of factors that impact mobile video calling experiences which include: the type of wireless network to which the user is attached, use of the best effort Internet as the medium which introduces latency at various points, and the video “engine” in the client application.
What the researchers at Bell Labs found put simply is that broadband wireless network throughput is not the issue for creating a great video calling experience, proper packet prioritization and network engineering that puts a premium on consistent, guaranteed “fast” responses are.
Knowing, feeling and conquering the pain
What are the major pain points that slow things up?
- Link bandwidth— video needs high throughput both downstream and upstream which is constraint where 4G LTE actually out performs wired DSL.
- Video adaptation to changes in link bandwidth— today’s video calling apps distinguish QoE through video adaption algorithms that unfortunately react to different network congestion and random packet loss situations and hence lead to varying quality user experience. This is why consistency is so critical.
- Packet latency — the bane of packet communications from its earliest years for things like VoIP calling, the combination of propagation, transmission and queuing delays, can be lethal on QoE. Without going into the details, it is best to say going back to the link challenges that any queuing delays, especially upstream ones will create an undesirable mobile video calling session.
- Competing traffic flows — network congestion, particularly that created by other video apps and services like streamed video, will impair QoE of video calling if traffic is not properly prioritized and groomed.
And those are the major ones. Bell Labs also identified the following minor ones:
- Video and audio codec performance —there are lots of codecs and they perform differently. Making them transparent to the user is a challenge.
- Packet loss — something to be aware of but not an issue thus far in broadband wireless implementations.
- Video processing power — impact on battery life is an issue here.
The end results of the identification of all of these pain points is that to quote Mun, “For the wireless network, the mobile video calling service will likely require mapping to a quality of service (QoS) Class Identifier (QCI)...It will also require priority weighting or guaranteed bit rates (GBRs) in base stations.”
Why is there good news?
Despite all of the issues raised in the study, the answer to the second question as to whether residential networks are ready to provide the required QoE.
How can carriers prepare themselves to take advantage of the mobile video calling opportunity, especially in regards to having the ability to differentiate their service on the basis of providing a superior user experience? The Alcatel-Lucent Open API QoS solution provides a solution to the challenge of how to maximize on this opportunity. It gives user-controlled prioritization by providing application programming interfaces (APIs) for the ACP client or server to specify which flows get priority through the network.
Getting ready to be ready when you are
The mobile video explosion is at hand. Juniper Research, for instance, predicts that mobile VoIP and mobile video calling services are going to develop significantly faster in developed markets. They cite not just the enticement of packetized real-time visual interaction over broadband wireless networks with real quality of experience (friends and family) but also due to the video enablement of capabilities that enhance things like telemedicine and distance-based education.
However, developing nations will not be far behind. New network construction will be next generation network construction. Coupled with smartphone adaption as the communications device of choice and the growing size of those who can afford advanced services, usage of video calling services in developing countries cannot be under-estimated especially in regards to how business usage in place that have had historically poor wired services are leveraged. In addition, user demand for improved QoE around the world is already driving the possible offerings of a premium Facebook capability, and others are waiting in the wings.
The end result of the Bell Labs research was that while wireless networks face challenges for getting ready to provide pervasive QoE that will allow mobile service providers to meet or exceed user video calling expectations while they may be varied they are not insurmountable. In fact, solutions are available that not only will enable users to enjoy the services they wish to use today, but through network engineering that prioritizes speeds and packet optimization operators will be well positioned to offer new services with sufficient QoE going forward.