Madhav Bhatta, Chief Engineer, Mobile Backhaul Transport, Global Network Solution Department, Huawei

Emerging markets have become an intense battleground for multinational mobile operators with their strategy to enlarge their global footprint. These operators are facing enormous challenges to build the backhaul networks with their TCO efficient and quick time-to-market (TTM) strategy. This article identifies the challenges to build these backhaul networks and recommends the potential solutions from both a technical and business perspective.

Backhaul Challenges and Strategies

TDM-based architecture has traditionally met the requirements of 2G operators. Several major operators are still investing heavily in the expansion of their 2G mobile networks. Thus, the increase in traffic volumes and the pressure of new service deployment demand an exponential growth of bandwidth on backhaul networks.

The gradual migration from 2G to 3G will definitely lead to long term co-existence of the two networks, which in turn poses a challenge in backhaul to support the multi-services transport requirement for a longer duration. This leads to a requirement to bear native TDM services, plus TDM + ethernet services and IP in the future.

Operators in this case, bear a high cost for capacity expansion in SDH networks, leased lines, and microwave equipment. This leads to a need to improve transmission media to accommodate the increased service demands.

Mobile operators are facing considerable operational challenges as they seek to maintain their existing networks with multiple backhaul technologies, including PDH & SDH Microwave, SDH multiplexers, and regional WDM technologies. The evolution to 3G backhaul creates additional hurdles that lead to high TCO in short and long-term.

Challenges in Stacking Many Boxes

Approximately 80 percent of existing backhaul networks comprise legacy PDH, SDH microwave links, and multiplexers stacking many PDH, SDH boxes in coexisting sites. This increases the CAPEX as well as operational difficulties. Existing TDM backhaul modes may meet immediate requirements. However, in the event of adding more subscribers or even migrating to 3G, there will be a need to rebuild or introduce new backhaul technology, leading to more boxes within the network to sustain the increased traffic. It would then become difficult to have an overview of the whole network. In addition, the problem of maintenance as well as efficient media transfer could also arise.

Nevertheless, a series of steps exist to address these problems:

• Reduce complexity via unified solution

Huawei intends to solve operators' complexity arising out of stacking many boxes in their networks with flexible solution equipped with diversified interfaces in an integrated solution. The existing stack of SDH boxes can be easily replaced by a single highly integrated MSTP platform.

• Support all backhaul scenarios to avoid stacking

Given the long-term nature of 2G, 3G, and HSPA co-existence, the backhaul solution should be applicable in diversified scenarios to avoid stacking boxes. To meet the growing demand of 3G services, ATM IMA can be plugged within the same box thus avoiding stacking of ATM switches. Native E1, IMA E1, and FE interfaces at cell sites offer seamless access of cell sites and network aggregation. On the RNC side, single equipment can support STM-N, ATM STM-N, and FE/GE interfaces to cover all the scenarios of RNC-side aggregation.

Each of these features needs to support card level expansion to enable faster deployment and realize the rapid TTM requirement of cellular networks. Directly upgrading from an existing system incurs less CAPEX and is easier to deploy. A single network management system (NMS) operates the whole network with less O&M demands, and thus less manpower.

Daily Network Operation Challenges

In 2007, Huawei had an opportunity to join with FT/Orange to analyze the TCO for emerging markets targeted by 30+ Orange affiliates. The major findings revealed that the excessive OPEX was derived from managing different network platforms.

It is extremely difficult to locate failures across multiple network management platforms that lack both end-to-end (E2E) provisioning and unified performance monitoring mechanism as it makes the troubleshooting extremely complex and causes high OSS integration costs.

In Tier 2 operators' networks, thousands of alarms are generated daily by backhaul equipment. However, alarm correlation techniques to determine faults over multiple network platforms such as microwave, SDH, etc., are absent. This delays rectification of faults. The need to maintain multiple maintenance teams of engineers without a unified NMS will hamper quick fault detection and have adverse effects on the OPEX of the operator.

Operators can take a number of measures in terms of simplifying operations and reducing O&M costs. Some of the measures that could be looked into are:

• One NMS for one backhaul

One unified NMS for each backhaul greatly simplifies daily operations. The solution features a complete point-and-click function that defines traffic routes between BTSs/ Node Bs and RNCs/ BSCs. It avoids multi-segment provisioning to reduce repeated additional link configurations.

The NMS covers up for all equipment, such SDH, MSTP, WDM, microwave, and even packet backhaul.

• Accelerated troubleshooting

Rapid troubleshooting is a key focus for major operators aiming to reduce the number of site-visits, minimize network interruptions, and implement preventive maintenance. Multi-segment configuration is the major factor that increases on-site engineering tasks and hinders failure location.

However, simply deploying a single unified NMS is insufficient; an alarm correlation system is necessary to clarify fault definitions throughout the network. Alarm correlation DB uses aggregation rules to generate root and non-root alarms, and can suppress 90 percent of the non-root alarms to quickly locate the root causes. Ethernet layer OAM also assists in monitoring E2E service connectivity and links status.

• Fewer on-site visits

Cellular networking is growing very fast, requiring quick rollout of mobile base stations. This brings the need for regular capacity adjustment, upgrade, and optimization. Hence, field engineering becomes another important aspect of operational challenges. This situation can be ameliorated by flexibly configuring E1, IMA E1, and FE through card level, software programmable radio (microwaves), and a built-in add-drop mechanism.

Moreover, field visits can be further reduced by using hot patch technology to fix bugs, online software to remotely upgrade NEs, and the NMS to monitor remote optical power.

• Cut inventory and spare parts

Unified hardware and software makes all service cards compatible with the products applied to core sites for access and aggregation. When configured for specified distances, the flexible SFP pluggable optical module reduces the expenditure on spare parts, and the same principle applies to microwave IDUs, ODUs, antennas, and RF couplers. These features can be combined to minimize the TCO network-wide.

Cost of Realty and Future Challenges

2G mobile operators are confronted with the challenges of finding effective ways to evolve to 3G/ HSPA while lowering OPEX and CAPEX. The optimum choice must involve minimum costs and reuse of existing network resources to bear legacy services, provide 3G/ HSPA services, and ensure seamless future expansion capabilities.

For example, increasing investment in legacy service provision with the hope of a delayed migration to all packet platforms in the future is broadly a short-term expedient. On the other hand, rushing to acquire a nascent technology that applies to a separate IP platform will not only incur a high and immediate CAPEX, but also pose operational problems for teams who are currently accustomed to legacy microwave and SDH networks. In terms of emerging markets, the latter approach requires specialized and highly skilled networking experts.

• Evolution not Revolution

Huawei's recommended strategy is based on the unified solution that simultaneously supports TDM and packet to seamlessly transport 2G and 3G or HSPA data services. We recommend the smooth evolution from MSTP to MSTP+ solution instead of the need to build an overlay network.

At the cell site gateway, TDM microwave network can be smoothly upgraded to hybrid microwave to achieve seamless ethernet access. This can be achieved by simply adding RF cards into the existing RTN equipment.

MSTP+ and hybrid microwave can be smoothly evolved by replacing cards that enable rapid deployment and rollout of 3G networks. This saves considerable CAPEX and OPEX, as well as enables the operator to recoup ROI on legacy investments.

E2E Backhaul Solution for Emerging Markets

Huawei has designed its solution to meet current requirements and seamlessly evolve into future pure packet platform architecture.

From the perspective of strategic business models, Huawei's TCO analysis report maintains that operators' primary goal rests in achieving TCO reductions. In this sense, the MSTP-based solution is irrefutably cost-effective, suitable for different application scenarios, and simplifies O&M.

Huawei proposed its mobile IPRAN backhaul solution to respond to the current dominance of TDM traffic in 2G networks, the massive amount of data involved with 3G services, and the inevitability of future All-IP migration. The solution also incorporates legacy TDM strengths'such as OAM, performance monitoring, and so on'to maintain the operational experience of existing operation teams without requiring networking expertise. In the light of the various challenges in the emerging markets, Huawei has considered a multi-layer TCO reduction approach to systematically reduce OPEX and CAPEX.

Following Huawei's technical and economic analysis and discussions with numerous operators regarding emerging markets, FT/Orange, China Mobile, Beijing Unicom, StarHub Singapore, Digi Malaysia, Excelcom Indonesia, Wind Italy, Orascom Group, and MTN plan to implement Huawei's solution in their networks.

Huawei believes that the proposed solution will eventually bring a 30 percent to 40 percent reduction in TCO.