Though vitally important to both wireless and wireline service providers, the mobile backhaul market itself is overflowing with multiple technologies, vendors, service providers and solutions. This volatile mix makes choosing a solution, vendor or technology risky and consequently inhibits investment. The MEF's "Mobile Backhaul Implementation Agreement - Phase 1", approved in January 2009 as MEF 22 is intended to standardize solutions and services and thus reduce the risk of investing in mobile backhaul.
The further success of the wireless services market rests on continued wireless data services growth. For example, AT&T (NYSE: T) and Verizon's (NYSE:VZ) combined results for the most recent quarter showed a 28 percent annual increase in data services revenue contrasted with no growth in voice revenue. This is due to the rapid adoption of wireless data services and the saturation of the voice services market where new subscribers are primarily marginal users. This is seen in a year over year increase in data service ARPU of 17 percent and a drop of 8 percent in voice ARPU.
Most of a "wireless" network is in fact wired-traffic from wireless base stations (the wireless part) is backhauled over the RAN (Radio Access Network) to the IP Mobile Core, and ultimately onto private IP, Internet and PSTN core networks. Backhaul transport media include microwave, copper and fiber while protocols include PDH/TDM, ATM, IP MPLS, and Ethernet. Backhaul may be owned and operated by the wireless operator, incumbent wireline operator, or a third party.
Backhaul economics are daunting. 2G and until recently 3G backhaul services were provided over T1/E1 facilities with a cost per bit of bandwidth about five times that of Ethernet facilities. Clearly, a transition to Ethernet is needed to support the continued growth in wireless data services. However, even Ethernet's economic efficiency will be challenged by the economic requirements of the latest 3G and 4G service offerings. Traffic growth is much faster than revenue growth. For example, a recent MEF document shows HSDPA traffic in Europe, APAC and the Americas growing at an annual rate of over 500 percent. This is more than 17 times the revenue growth rate. Very high rates of productivity measured in terms of cost per bit of bandwidth must be realized to sustain profitable operations.
Wireline operators are best positioned to achieve the dramatic cost per bit of bandwidth reductions needed to satisfy the economic imperatives of mobile backhaul. They alone have the necessary scale and scope of access network facilities needed to drive backhaul costs down to the required operating point. This requires leveraging as much common plant as possible to carry residential, enterprise and mobile backhaul services.
The MEF's standardization efforts are essential to achieving the needed scale and cost efficiency. In particular systems vendors and wireline providers require the assurances provided by complete and widely endorsed Ethernet mobile backhaul specifications in order to invest safely in new technology and new backhaul networks.
The MEF specifications encompass all aspects of mobile backhaul implementation including use cases, legacy network migration, preparing for LTE, accommodating IP base stations, standardized reference points, service requirements, connectivity, class of service (CoS), availability, synchronization, and operation of the RAN. CoS, synchronization and LTE are areas of MEF's current focus.
All existing MEF services--EPL, EVPL, ELAN and EVPL--are addressed by MEF 22. In addition, new services called Ethernet Private Tree and Ethernet Virtual Private Tree are introduced. The tree services recognize that RAN communications tend to be hierarchical with connections between base stations and Base Station Controllers (BSC) or Radio Network Controllers (RNC) rather than peer-to-peer. Backhaul services must maintain consistent performance levels and latency as well. Such time invariant performance runs counter to Ethernet's original connectionless operating mode. Unique mobile backhaul CoS traffic classes are specified to accommodate services, including: synchronization with a very high CoS, through video streaming with a medium CoS, to a low CoS for web browsing.
Synchronization is another area that requires detailed specifications to ensure support for legacy services that were designed for T1/E1 transport as well as new IP-based conversational and streaming services. Two synchronization methods are specified--Synchronous Ethernet (SyncE), a new method that is defined by output requirements and a new UNI attribute--and packet based methods that align with ITU-T recommendations.
LTE backhaul introduces new service requirements. LTE is all IP from the start; it requires new logical interfaces and is multipoint in nature. It employs a new pooling concept and has much greater throughput--up to and beyond 100 Mbps. MEF 22 Phase 2 addresses these new requirements.
MEF 22 Phase 1 addressed most issues needed for current mobile backhaul implementations. Phase 2, which is expected to be approved next year, will address performance recommendations, Service OAM for performance management, SyncE, resiliency, and LTE.
Successful wireless data services require very high bandwidth and low-cost mobile backhaul. MEF Ethernet mobile backhaul standardization efforts are essential to creating market consensus among systems vendors and incumbent wireline operators needed to achieve the low cost objectives. Mobile backhauls expected high growth rates and requirement for converged Carrier Ethernet infrastructure makes it an attractive wholesale service opportunity for wireline operators.
Michael Kennedy is a regular FierceTelecom columnist and is the co-founder and Managing Partner of Network Strategy Partners, LLC (NSP)-www.nspllc.com-management consultants to the networking industry. He can be reached at [email protected].