YOU ARE AT:Network InfrastructureReader Forum: Transforming the ‘Layer 2’ connected mobile cell site

Reader Forum: Transforming the ‘Layer 2’ connected mobile cell site

Editor’s Note: Welcome to our weekly Reader Forum section. In an attempt to broaden our interaction with our readers we have created this forum for those with something meaningful to say to the wireless industry. We want to keep this as open as possible, but we maintain some editorial control to keep it free of commercials or attacks. Please send along submissions for this section to our editors at: [email protected].

Mobile cell sites are transitioning from simple base stations for voice and low-speed mobile data access to multi-functional hubs for delivery of new services, multimedia-rich content and broadband data access. Existing mobile backhaul solutions make this transition difficult as more capacity and smarter devices complicate cell site configurations, increase cost and add management overhead. This paper presents a new integrated cell site architecture that promises to transform the cell site and restructure the mobile backhaul cost equation – making the new service-delivery cell-site vision possible with fewer boxes and simplifying tomorrow’s mobile network.

The expanding role of the mobile network

In the past, mobile networks were built to deliver voice services with data connectivity services as a secondary consideration. Today’s mobile networks are being relied on to support mission-critical business objectives, which are far beyond the realm of basic mobile connectivity. New enterprise, government, videoconferencing, social media and mobile commerce services place growing demands on the network in terms of capacity, security and reliability. The mobile network infrastructure must be flexible enough to deliver a growing number of new services easily and cost effectively.

Moreover, capacity demands are driving new levels of network densification by leveraging small cell, Wi-Fi and cloud-based RAN architectures. While the densification “end game” is hard to pin down, what is certain is the mobile network needs to be flexible to support a considerable amount of future uncertainty. To meet subscriber coverage goals, and considering consumer demand for more mobile applications and that the mobile network operators desire to enrich their service offerings, lower costs and grow average revenue per user, the cell site is undergoing a transformation.

The mobile cell site as a service delivery hub

Operators need to enrich their offerings to attract and retain high-value customers to offset the low ARPU for traditional consumer-level mobile users. To support new business initiatives, fast and flexible delivery of new services is necessary. This fact, combined with the pending network densification, means the cell site must evolve.

The mobile cell site needs to evolve from simply housing a base station to supporting a network architecture that accelerates the delivery of new services to achieve business goals – the service delivery hub. IP is the only proven, scalable, future-proof technology upon which to meet these goals completely.

Cell site routers: the required ingredient for cell site transformation

Traditional carrier Ethernet (i.e., Layer 2) advocates believe cell sites should be maintained at a L2 implementation, citing a lower cost and simpler network infrastructure compared to L3. IP (i.e., L3) intelligence at the cell site gives the operator flexibility to support new scalable, high-quality services and the pending mobile network densification. Based on this, below are some reasons why routers at the cell site are poised to become the norm in future mobile network architectures.

New service delivery: Many enterprise, government, videoconferencing and banking services can be delivered much more efficiently and at lower cost over a routed network. For example, L3 VPN services often provide the most flexible solution for enterprise access and connectivity services – especially as service requirements expand. Many of these transport connections will terminate directly at the cell site. On-site routed infrastructure can provide optimum flexibility to support the full range of access services including IP VPN, L2 VPN, VPLS, MPLS VPN, etc. to ensure overall business objectives are met.

Lower cost: L2 is often believed to be a lower capital expense solution, but many times requires higher operating expense – especially as the network changes and grows and new services are offered. Traditionally, IP routers have commanded a higher price tag due to the higher cost of the silicon chips they required. However, newer technologies coupled with a highly software driven paradigm and the fact that cell sites do not need the same level of performance found in traditional routers makes routing at the cell site cost effective.

Network densification: The network edge is constantly evolving, resulting in today’s edge sites becoming aggregation nodes – where IP and multi-protocol label switching are now commonplace. In addition, the network needs to be flexible to support the evolving network topology. L3 intelligence makes it easier to change the network topology (e.g., add a site, mesh to rings). With a mesh of L2 connections, for example, operational complexities can result in huge costs and slow the pace of the network evolution – meaning delays in new service offerings or improving network performance.

Multi-service connectivity: New services, and network densification means the emergence of a multitude of devices at the cell site. Many of these devices are multi-service interfaces and many can even include legacy interfaces such as TDM, ATM or even FR/X25. Routers are the only proven solution to cost effectively converge multi-services interfaces onto a single low cost IP transport. Also, with the migration to LTE/HSPA+, native IP is the default transport technology.

Security: L3 intelligence at the cell site provides many security benefits. Because mobile security requirements continue to evolve, routers offer optimum flexibility (e.g., having the ability to initiate and terminate IPSec) and the ability to support ever-changing security requirements. Ultimately, ensuring both control and bearer channels are carried in the most secure fashion possible.

Higher capacity and network scalability: IP routers at the cell site enable the network to better deal with capacity challenges. By having the intelligence to route around issues, routed intelligence at the cell site can have a great effect on capacity. When compared to flat L2 networks, routers scale to vast numbers of nodes, thus, giving flexibility to grow the network seamlessly with a lower total cost of ownership.

X2 interface: The interface between ENodeBs in LTE architecture, X2 creates significant challenges for network architects. Carrying up to 10% of an ENodeB’s total traffic, the X2 has extremely low latency requirements and is a key connection to sup-port mobile hand off. When ENodeBs are on different subnets, a router is required to provide X2 connectivity and avoid storms of broadcast traffic. A more scalable X2 infrastructure is achieved by supporting router functionality at the cell site.
Cell site routers are an essential investment to support the expanding role of the mobile network.

Microwave becoming more complex, less integrated

Despite all the benefits, the addition of the router adds a completely new level of complexity to the transport network of microwave connected cell sites.

Routers can be deployed alongside a microwave indoor unit in a split-mount microwave scenario or connected directly to an outdoor unit in an all-outdoor Ethernet scenario. In either case, routers are not well integrated into the microwave transport network, which can cause a host of challenges for operators.

Five key challenges are listed below:

Management complexity: Separate devices, with different management interfaces, create a multitude of operational challenges. End-to-end configuration, provisioning, fault and performance management can be difficult especially in IP MPLS
configuration.

Lack of dynamic bandwidth awareness: The router does not have awareness into the microwave radio’s link bandwidth. Without this knowledge, traffic engineering and L2/L3 routing become very difficult. In addition, adaptive coding and modulation, which results in dynamic reduction of link bandwidth, makes matters worse. Network performance is drastically affected in the above scenarios where routers are “flying blind” and not aware of the actual bandwidth available in the network.

Poor failure detection and recovery: In the above scenarios, the router generally does not know what the microwave radio is doing in relation to failure recovery. Separate boxes require complex signaling/inter-working that is not well coordinated. Also, there is a lack of integrated failure recovery across all layers including the IP/MPLS layer.

Non-integrated RF: The above options all involve Ethernet connections from the router to the microwave device, either baseband unit or ODU. This is a departure from traditional microwave architectures – which have worked well for years and represent 95% of all microwave deployments today, where a microwave baseband unit is connected to the RF unit with an intermediate frequency connection over the coaxial-cable interface. With the router-plus-Ethernet RF unit options, the modem is in the ODU on the tower, which requires a tower climb for any modem upgrades. The multitude of protection options (e.g., 1+1, N+0, SD, FD) gets extremely complex, and many configurations are not supported. The cabling is also a challenge. Existing coaxial-cable infrastructure cannot be reused, and many new cables are often required (RJ45 Ethernet has limited distances so optical cables need to be run in addition). Power over Ethernet injectors are also generally required often necessitating purchase, deployment and ongoing maintenance of a separate box. In summary, the lack of RF integration on these router-plus-Ethernet radio configurations spells trouble for mobile network operator operations teams.

Lack of integrated synchronization, security and quality of service: Router-plus-Ethernet ODU solutions lack an integrated security policy across all layers and boxes. They also lack automatic alignment of QoS policy across all layers leading to a complex operations effort to manage and maintain end-to-end QoS policy. Also, because of the lack of integration, clock source change on router results in a network wide microwave outage.

The evolution of the cell site will be an important transformation in the future of mobile networks and new solutions will be needed to deal with the growing complexity.

ABOUT AUTHOR