One lesson learned from the global rollout of 3G is that when you have a problem with your first deployment it takes a long time and a lot of money to recover the market. LTE has arrived – with the first deployment by TeliaSonera in December 2009 and a trickle of arrivals announced for later this year. A lot depends on these early deployments delivering on the technology’s promise, says Nigel Wright of Spirent Communications
High expectations is the first challenge that must be met. The initial 3G market was relatively forgiving. The productivity increase attributable to the transition from fixed-line modem to mobile Internet connection was so significant that mobile executives made allowances for any unmet expectations. But today’s user is becoming addicted to high-speed mobile broadband and their natural instinct is to blame service faults on the network rather than the device. The expectation is that the next generation of mobile services will be a quantum leap beyond the current 3G user experience.
The second challenge is the complexity of LTE deployments. Although LTE makes use of a simpler all-IP architecture with smarter network nodes, there are still plenty of challenges that can impact a user’s Quality of Experience (QoE). Inter-technology (2G/3G/LTE) mobility, support for voice services and greater air interface complexity with MIMO, to name but a few. So it is vital to learn from past mistakes that everybody – operators, equipment and device manufacturers – must take every measure to ensure that subscribers have a great experience from day one.
How long have we got?
While there is no doubt that mobile communication’s future is LTE, there are technical challenges to address, as well as business reasons for delaying LTE deployment.
The deployment of 3G taught us that it is one thing to launch a few pilot services, but quite another to allow time for the whole ecosystem – including the radio access network, the packet core network and handsets and devices – to reach critical mass. This is especially true for LTE, where there is still churn in the specifications.
There is a strong business case for not rushing prematurely into LTE. The majority of operators planning to deploy LTE already have robust and extensive HSPA networks. Purchasing new spectrum, in addition to brand new infrastructure, will cost billions of euros and the operator must determine over what time period these hefty up-front costs will be recouped by their LTE services.
According to the Dell’Oro Group, many operators are planning to migrate to HSPA+ as an interim step. It’s no match for expected LTE performance, but as a cost-effective interim measure for the massive existing 3G user base, HSPA+ is a sound business decision for many.
As a result, analysts suggest that the real take off point for LTE will not be until 2012 when Europe and China begin significant deployments. All three Chinese operators have committed to deploying LTE by 2013.
Two more years – that deadline might not seem that near for operators, until we look a little more closely at what lies ahead…
Key QoE challenges to address
Delays in initial adoption simply reinforce the obvious fact that carriers will need to operate both 3G and 4G networks. Multi-mode devices will continue to be required, negotiating the highest level of service where available and falling back to legacy services when necessary. Although wireless access can pose some of the more visible QoE issues, the more fundamental challenges lie in LTE’s All IP Network (AIPN) architecture, replacing legacy TDM/ATM voice networks, as well as the need to upgrade existing backhaul networks to support the Evolved Packet Core (EPC).
There are five key areas where QoE issues can arise:
– User authentication and mobility. These functions depend on three devices in the LTE system – the MME (Mobility Management Entity), SGW (Serving Gateway), HSS/AAA (Home Subscriber Server / Authentication, Authorization and Accounting) server – and their interaction between wireless and core networks.
– Multi-platform interoperability – the mobile user must move transparently between LTE and legacy technology cells without call loss or other QoE degradation.
– Timing and synchronization – this was a standard feature of legacy TDM backhaul technology, but LTE architecture specifies an All-IP network based on Ethernet, which lacks these native features. While the IEEE 1588v2 standard promises the precise levels of synchronization needed for real-time applications, thorough testing of its implementation is necessary.
– Operations, Administration and Maintenance (OAM) is another area where enhancements were needed to qualify Ethernet for mobile backhaul. Link OAM is for a single link between two devices, and Service OAM is for fault management, performance monitoring and protection switching for a circuit that traverses multiple devices over one or more network operators.
– Backhaul throughput and latency – once Carrier Ethernet was formalized to ensure features such as predictable latency, minimal jitter, guaranteed bandwidth and traffic engineering, it became the obvious candidate for All-IP mobile backhaul.
The way forward
Cost-effective backhaul throughput is clearly going to be a major issue once LTE puts its anticipated data rates of up to 100Mbps or more into users’ hands. But haven’t we shown that we still have two years leeway before that challenge begins?
Forget it! Backhaul networks are already creaking at the seams from the mass of data traffic and adding leased lines is an expensive temporary solution. The smartphone has moved from a niche business solution to a consumer must-have.
This converged mobile communications device is putting an enormous data throughput load on mobile networks, not to mention the added signalling burden as smartphones conserve their batteries by constantly signing on and off the network during use. As a result, there is already a lot of pressure to upgrade to Carrier Ethernet backhaul simply to meet the scalability and performance needs of today’s traffic, let alone the data tsunami in years to come.
So, although the rollout of LTE wireless networks may not be an urgent priority, the transition to next generation backhaul is already high on the agenda. And it must happen without disrupting or degrading existing services as well as being ready to support superb performance and QoE when LTE comes online. We cannot allow existing users to be treated as guinea pigs, we need to get the service right from day one. And that calls for thorough pre-testing of network upgrades prior to roll-out.
Although an All-IP core network offers a simpler and more efficient basis for data traffic, as the previous section shows, there is more complexity involved in testing the interface between the evolved packet core and LTE radio network that delivers services to the user. Testing to ensure a seamless upgrade to Ethernet backhaul is already an urgent priority and the need to fully test the LTE EPC is increasing rapidly, especially when inter-technology mobility and impact on QoE is considered.
Solutions for LTE testing
The lessons from 3G need to be remembered. Even if LTE roll out is not an immediate prospect, gearing up the infrastructure is already a vital priority. Today’s user of smartphones, data cards and netbooks already has sky-high expectations and infrastructure upgrades must only be apparent to a user in terms of improved QoE. It may take a long time to recover from the business implications of user disappointment with networks and services that are rolled out before they are thoroughly tested and ready.
Spirent Communications is a global leader in integrated performance analysis and service assurance solutions for IT systems and networks, with well over a decade of experience in pre-testing 2G and 3G technologies, as well as providing test and monitoring to ensure continued service quality. Building on this experience, Spirent is now geared up to offer the same support for the next-generation mobile network and devices and is actively supporting mobile operators, core network infrastructure manufacturers and device manufacturers with their LTE evolution product testing.
Already the world’s leading provider of 2G/3G mobile packet core and Ethernet network test solutions, Spirent is ahead of the game in providing and supporting Ethernet backhaul test and monitoring solutions, and solutions for testing mobile packet core performance, scalability and mobility. Spirent also offers solutions for LTE device testing.
Spirent test solutions are used by operators for acceptance testing, user device and network equipment interoperability testing, as well as validation of equipment manufacturers’ performance claims. Network equipment manufacturers use Spirent solutions during development, QA, installation, service and sales cycles to verify and demonstrate that their equipment meets operator requirements.
As the one-stop provider of choice for next generation test solutions, Spirent provides operators and manufacturers with innovative and industry-standard tools to test from the device through the backhaul and mobile packet core to the Internet.
Spirent has the test solutions, experience and support you need to make sure your LTE network will be ready.
About the Author:
Nigel Wright is Vice President of Product Marketing at Spirent Communications. In this role he oversees the marketing communication and industry engagement around all of Spirent’s wireless and positioning product lines.
Spirent Communications plc
Tel: +44 (0)1293 767676
www.spirent.com