LTE-Capable Mobile Backhaul

Transcription

1 APPLICATION WHITE PAPER Author: Michael Ritter ADVA Optical Networking ADVA Optical Networking All rights reserved. With rising demand for mobile broadband services, operators are seeing a sharp increase in bandwidth requirements. To keep pace with demand, operators must evolve to new packet backhaul networks that offer increased capacity at lower cost while providing the necessary service reliability and quality of experience that users expect. This white paper focuses on the challenges operators face when migrating to LTE and LTE Advanced radio access and the solutions they need to profitably benefit from packet backhaul. Introduction The rising tide of data traffic experienced in mobile networks is putting the backhaul infrastructure under more pressure than ever before. Data intensive applications on powerful smartphone and tablet devices are popular with many users and the arrival of LTE and LTE Advanced will only accelerate this process. Infonetics Research reports that the number of mobile broadband subscribers passed fixed broadband subscriptions in 2010 and is estimated to reach 2.1 billion by Source: Infonetics Research 2011 Figure 1: Mobile broadband subscriber growth The introduction of LTE and LTE Advanced also referred to as 4G radio access technology promises a whole new mobile broadband experience for private and business users, with short latency and data rates beyond 100 Mbit/s. At the same time, service differentiation and multiple quality-of-service profiles

2 will enable mobile network operators to efficiently use available spectrum while offering differentiated services with a superior quality of experience to their customers. However, this new, fixed network-like performance can only be experienced when supported by the backhaul network. There is general consensus in the industry that only packet-based Carrier Ethernet backhaul will be able to meet the challenges. Carrier Ethernet networks provide the bandwidth and flexibility required to dynamically adapt to capacity and connectivity demand originating from mobile services at cost points attractive to network operators. Reliability of the mobile backhaul network is essential for efficient network operations and providing a superior user experience. While efficiency and reduced cost per bit are important metrics, reliability of the mobile backhaul network is essential for efficient network operations and providing a superior user experience. With the introduction of LTE and LTE Advanced, the architecture of the backhaul network becomes more diverse and has many more dimensions. Connectivity between the mobile core and the base stations is no longer strictly hub-and-spoke as with 2G and 3G radio access technology. Base stations now communicate directly with each other, exchanging signaling and user data without involving the mobile core. They also use different anchoring points for signaling and data traffic in the mobile core. Data plane and signaling plane are now completely separated. Furthermore, the concept of small cells introduces another level of complexity. Small cells are an important component of LTE to provide substantially increased access capacity to a large number of users and enable a more efficient utilization of the available spectrum. Backhaul Fundamentals Mobile networks are growing. In many countries, radio access network installations have evolved from 2G to 3G and are now evolving to 4G while maintaining a large portion of the legacy radio equipment. The diversity of radio equipment installed at cell sites poses a challenge especially to the backhaul network. While the IP-based architecture of the Evolved Packet Core (EPC) is designed to replace former 2G and 3G core networks, this migration is a slow process for many operators. Seamless handovers for both voice and data to cell towers with older network technology such as GSM, UMTS and CDMA2000 Figure 2: Mobile backhaul in the context of the EPC 2

3 therefore requires a careful design of the backhaul network in addition to the mobile infrastructure itself. Transmission delays have to be kept at a minimum across the entire backhaul infrastructure while legacy TDM and packet-based traffic must be transported simultaneously. In addition, the backhaul network needs to provide the flexibility to migrate to the anticipated long-term solution. A mobile backhaul network based on physical fiber infrastructure clearly is the ideal solution from a capacity, reliability and operational perspective. However, many cell sites will be microwave- and copper-fed for years to come. While larger cell sites and those acting as aggregation hubs can only provide the required user experience when connected over fiber, there are many sites especially in rural areas where new fiber deployment is not justified from a commercial standpoint. Migration to packet-based microwave and Ethernet-over-Copper is the alternative solution. For many of the small cells that are expected to be deployed in metro areas during the coming years, microwave and the physical infrastructure already in place will play a dominant role when designing the backhaul network. Nevertheless, the share of fiber-fed cell sites is expected to grow with copper losing its attractiveness due to bandwidth constraints and microwave remaining at a stable share, cf. Figure 3. Source: Infonetics Research 2011 Figure 3: Installed backhaul connections by physical medium The architecture of packet-based mobile backhaul networks is not consistent for all network operators. There are topological and operational differences depending on whether the backhaul network is operated by the mobile service provider or leased from a fixed-line network operator. While a single-operator environment provides advantages in terms of simplicity and efficiency, the multi-operator environment illustrated below is the typical case for fiber- and copper-based backhaul. Fixed-line mobile backhaul services are often provided by a third-party operator or a separate organization within the same operator. In a multi-operator environment, mobile backhaul services are typically offered over a converged, multi-service backhaul and aggregation infrastructure. Network resources are then shared with other traffic originating, for example, from DSL services and business Ethernet connections for enterprises. 3

4 These different scenarios result in a number of different challenges and implementations when it comes to delivering mobile backhaul services in realworld deployments. In a single-operator environment, the complete network infrastructure including radio access, backhaul and mobile core network is controlled by one organization. The backhaul network can therefore be designed and optimized according to the requirements of the mobile network. Service Level Agreements (SLA) are typically not defined explicitly at intermediate nodes. Figure 4: Multi-operator mobile backhaul environment In a multi-operator environment, the backhaul network operator provides an independent service interconnecting the radio access network with the mobile core. Quality of Service (QoS) is defined at the User Network Interface (UNI) and must be met and reported by the backhaul network operator according to the SLA agreed between both parties. Accurate SLA measurement, assurance and reporting play a critical role in this context. Challenges in Mobile Backhaul for LTE The first challenge is providing differentiated QoS while keeping the transmission latency at a minimum. QoS differentiation enables mobile operators to manage the performance of different streams of traffic. Even though Carrier Ethernet backhaul provides significantly more capacity compared to legacy TDM, dimensioning at peak rates is not practical and cost prohibitive. Backhaul networks will therefore be oversubscribed in many cases, making sophisticated QoS management a necessity and a powerful tool for managing user experience and cost. The optimum solution will essentially balance user satisfaction with economical and technical feasibility. In this context, transmission latency becomes a critical design factor, especially for delay-sensitive applications such as packetbased voice and online-gaming. Also, seamless call handover between cell sites requires keeping transmission latency at a minimum. With LTE and LTE Advanced, seamless handover can only be achieved when guaranteeing lowest latency on the X2 interface, which directly interconnects base stations with each other. Typical latency requirements in LTE are summarized in Table 1. The importance of QoS management across packet-based mobile backhaul networks mandate powerful tools for service assurance and simplified network 4

5 The QoS provided by the backhaul network must be constantly measured and reported. operations. The QoS provided by the backhaul network must be constantly measured and reported. Parameters such as packet delay, delay variation and packet loss are important characteristics ultimately defining user experience. The individual performance requirements must be met for each traffic stream and immediate measures need to be taken when the network can no longer assure the anticipated QoS. The potentially large amount of traffic streams transported over Carrier Ethernet mobile backhaul networks additionally requires efficient procedures for performance verification testing at service turn-up. LTE Interface S1 User Plane S1 Control Plane X2 User Plane X2 Control Plane Delay Budget ms 10 ms 1 ms (recommended) 10 ms Table 1: Delay budgets in LTE Mobile services are dependent on timing and base stations need a stable reference to support mobility. As operators replace their TDM-based backhaul with Carrier Ethernet backhaul, they face a major challenge: how to provide precise timing reference or synchronization for base station clocks and do so in a cost-effective way. Mobile services are dependent on timing and base stations need a stable frequency reference to support mobility. Actually, operators are confronted with a broader, two-part challenge. Firstly, they must replace their TDMbased clock function with a suitable packet clock. Secondly, as they deploy advanced LTE technologies incorporating Time Division Duplex (TDD) multiplexing, they must eventually expand that packet-clock capability so that it distributes not just the frequency reference but also phase and time-ofday information. The timing requirement for different LTE air interface standards is summarized in Table 2. Air Interface Frequency Time/Phase LTE (FDD) 50 ppb - LTE (TDD) 50 ppb 3 µs LTE MBMS 50 ppb 5 µs Table 2: Air interface stability needs Solutions for QoS and Latency Management QoS differentiation helps to manage and allocate network resources during times of congestion, adapted to the actual need of applications. It is a tool that guarantees that traffic generated by certain applications e.g., voice and 5

6 control plane signaling is prioritized over traffic from applications that are less sensitive to delay or loss performance. Carrier Ethernet allows prioritizing services by assigning to each service a specific QoS class, which is based on a number of parameters. These parameters include packet delay, delay variation and packet loss and are specified for the service across the entire backhaul network. The QoS defined for the LTE radio interface has to be aligned with the QoS experienced across the backhaul network. QoS must be managed consistently end-to-end. The QoS defined for the LTE radio interface has to be aligned with the QoS experienced across the backhaul network. Classification and tagging is therefore carried out by the base stations and the gateways in the mobile core, based on the information collected from policy servers. The 3GPP collaboration has defined a number of QoS Class Identifiers (QCI) for LTE, each referring to a certain type of application. The identifier is used as a reference for controlling packet forwarding and treatment across the radio access network and is translated into a packet priority marking to control packet forwarding across the Carrier Ethernet backhaul network. To meet the required QoS levels and simultaneously maintain cost efficiency, Carrier Ethernet supports sophisticated traffic management capabilities. QoS management in Carrier Ethernet networks enables better service to certain selected flows, therefore significantly reducing overall bandwidth requirements while still maintaining the QoS required for each individual flow. Figure 5 illustrates the architecture and the main building blocks of a generic traffic management implementation that is compliant to Metro Ethernet Forum (MEF) recommendations. The main functionalities include traffic classification, policing, queuing and scheduling. Figure 5: Carrier Ethernet traffic management architecture In order to meet the level of scalability and flexibility imposed on the backhaul network by LTE and LTE Advanced in particular, the Carrier Ethernet mobile backhaul must support QoS management for a large number of traffic streams also called Ethernet Virtual Connections (EVC) in a hierarchical queuing architecture. The combination of multiple QoS profiles and the potentially large number of connections between individual base stations referred to as 6

7 X2 interface and between base stations and the ECP referred to as S1 interface creates a challenging environment. Only a solid traffic management implementation with enhanced classification capabilities can assure efficient usage of network resources while meeting strict QoS requirements. Due to the increased autonomy of base stations and the improved user experience aimed for, LTE networks are in general more sensitive to latency accumulated across the backhaul network in comparison to 2G and 3G mobile networks. The availability of strict priority queuing in the traffic management architecture is therefore a must to meet the challenging latency limitations for signaling and control plane traffic as well as for time-critical applications. Operational Simplicity and Service Assurance Continuous and standards-compliant performance monitoring and automatic fault resolution are the foundation for SLA assurance and accurate reporting. The capability to cost-effectively provision intelligent services with differentiated QoS metrics across the mobile backhaul network makes the availability of powerful tools for installation, commissioning, performance management and SLA reporting inevitable. Manual configuration and test procedures do not provide operational efficiency and therefore limit scalability. Furthermore, continuous and standards-compliant performance monitoring and automatic fault resolution per traffic flow are the foundation for SLA assurance and accurate reporting. With Y.1564, the Telecommunication Standardization Sector of the International Telecommunication Union (ITU-T) has defined a standard for turn-up, installation and troubleshooting of services across Carrier Ethernet networks. The test methodology allows for fast and complete validation of Ethernet SLAs in a single test and with the highest level of accuracy. Services that will run across the network are simulated during the turn-up phase and all important SLA parameters are qualified simultaneously. Y.1564-compliant testing also validates the QoS mechanisms provisioned in the network to prioritize the different service types. It results in more accurate validation and much faster deployment and troubleshooting compared to manual procedures. To keep mobile networks alive and maintain the quality of experience users expect, mobile operators are particularly interested in continuously understanding the status of their packet backhaul services so they can localize faults and trigger corrective actions from remote locations. The Ethernet Operations, Administration and Maintenance (OAM) standards 802.1ag and Y.1731 defined by the Institute of Electrical and Electronics Engineers (IEEE) and the ITU-T, respectively, provide mechanisms for connection monitoring and performance measurement on an end-to-end service level. Based on the hierarchical concept shown in Figure 6, 802.1ag defines the following OAM tools: Connectivity Check Loopback Link Trace. These tools make use of specific Ethernet frames that are following the same path as the frames belonging to the monitored service. This has the additional advantage that no explicit interworking with service protection and restoration procedures or other dynamic network changes are required for compatibility. 7

8 Figure 6: 802.1ag and Y.1731 connection monitoring Y.1731 builds on 802.1ag to add in performance monitoring features on an endto-end service basis. Fault management and indication are supported by alarm indication signaling and remote defect indication. The mechanisms defined in Y.1731 enable backhaul network operators to measure and report one-way and round-trip service parameters for Frame Delay Measurement Frame Delay Variation Measurement Frame Loss Measurement. Service demarcation and aggregation units deployed in the backhaul network must be engineered with a hardware processing architecture to provide the required attributes. Carrier Ethernet OAM functions for connectivity fault management and performance monitoring provide backhaul network operators with a complete set of tools for assurance and reporting of SLAs to the mobile network operator. While backhaul network operators typically use these mechanisms on a constant basis, mobile operators may choose to only verify the service quality periodically. All Carrier Ethernet OAM functions must be implemented with special diligence to guarantee superior user experience. Scalability, flexibility and, last but not least, high measurement accuracy are elementary when designing mobile backhaul networks for a large number of EVCs. Consequently, service demarcation and aggregation units deployed in the backhaul network must be engineered with a hardware processing architecture to provide the required attributes. Scalability and high measurement accuracy in particular can only be achieved by a hardware-centric design. Radio Access Network Synchronization Both Carrier Ethernet system vendors and the timing community worked on methods to deliver synchronization information over packet networks. The obvious goals were to keep it simple, cost-effective, predictable and reliable. Two practical mechanisms for providing synchronization via packet-based networks have emerged: Synchronous Ethernet (SyncE) and 1588v2. Both standards are the result of efforts by international standards bodies, notably the ITU-T and the IEEE. 8

9 Two practical mechanisms for providing synchronization via packet-based networks have emerged: Synchronous Ethernet (SyncE) and 1588v2. SyncE uses the Ethernet physical layer to synchronize neighboring nodes. It is attractive to many network operators because it closely resembles the familiar SONET/SDH model and its timing quality is completely independent of the network load. However, SyncE only provides frequency synchronization and requires that each node in the hierarchy supports it. If a single network element in the chain does not support SyncE, all nodes lower in the hierarchy do not receive accurate timing information. 1588v2, in contrast, specifies a master-slave exchange of packets that carry time stamps for recovering frequency, phase and time-of-day information. Operators can use 1588v2 to provide synchronization directly across any packet network. However, operators must ensure that the synchronization flow is not distorted by packet loss, delay or delay variation beyond the filtering capabilities of the slave clock. The draft ITU-T Telecom Profile for 1588v2 requires that all nodes in the network must support 1588v2 boundary clock functionality for the high accuracy network phase synchronization required by LTE Advanced and other TDD air interfaces. Both mechanisms provide additional information about the delay conditions in the network and therefore support increased clock accuracy. Table 3 summarizes the key differences. Attribute SyncE IEEE 1588v2 Capability Frequency Frequency, Time, Phase Layer Physical Ethernet, UDP Distribution Physical Layer In-Band Packets Sensitivity Asynchronous Switches Delay, Jitter, Loss Table 3: SyncE / 1588v2 comparison SyncE and 1588v2 are complementary technologies that can co-exist in the network. SyncE and 1588v2 are complementary technologies that can co-exist in the network and can be used on the same path. Both technologies have distinct advantages and disadvantages over each other. SyncE is deterministic and the performance is independent of the network load. 1588v2 can function over asynchronous switches and additionally distributes phase and timeof-day information. Slaves that support both can converge on accurate timing information quickly by using the SyncE frequency to discipline the 1588v2 local oscillator. SyncE in conjunction with 1588v2 also provides an alternative holdover capability in case of failure at the packet layer. A combined implementation promises to deliver the best overall performance. Assured delivery with guaranteed QoS metrics is a necessity not only for data traffic streams but also for timing services. The ability to consistently monitor and accurately test and troubleshoot the synchronization infrastructure when delivering timing information via SyncE and 1588v2 is mandatory for assuring clock accuracy and therefore the quality of the delivered timing service. Assured delivery with guaranteed QoS metrics is a necessity not only for data traffic streams but also for timing services. As 1588v2 packet flows potentially traverse different technologies and operator networks, service assurance mechanisms as implemented in Carrier Ethernet OAM are required. 9

10 Network timing behavior is not a stationary process. It is subject to dynamic conditions and changes over the short and longer term. Appropriate tools are required for cost-effective and time-efficient end-to-end management of the synchronization domain during all phases of the network lifecycle installation, turn-up testing, monitoring and troubleshooting. Figure 7: Synchronization service assurance support tools The Right Solution for The explosive growth of video and data services on mobile devices has created a challenge as mobile network operators look to provide them to an expanding base of subscribers while simultaneously reducing the cost of transporting this increased traffic load across the mobile backhaul network. While efficiency and reduced cost per bit are important metrics, reliability of the mobile backhaul network is essential for efficient network operations and providing a superior user experience. Our Etherjack and Syncjack suite enable mobile backhaul network operators to deliver reliable, highperformance data and synchronization services. ADVA Optical Networking has a comprehensive FSP 150 Carrier Ethernet access and backhaul portfolio that offers a complete solution including scalable QoS management, end-to-end service assurance and accurate delivery of timing information for mobile backhaul networks of any size. Our Etherjack and Syncjack suite, which are fully integrated into the FSP 150 platform, enable mobile backhaul network operators to deliver reliable, high-performance data and synchronization services supported by a rich and complete set of tools for end-to-end service monitoring and assurance. Our FSP 150 Carrier Ethernet solution provides operators with the capability to evolve their mobile backhaul network without constraints and supports seamless migration of radio access networks to LTE and later LTE Advanced. It is architected to deliver % availability, supports end-to-end SLA management per traffic flow and scales with your radio access network: a complete and uniform solution for demarcation and aggregation applications in mobile backhaul networks. 10

11 About ADVA Optical Networking ADVA Optical Networking is a global provider of intelligent telecommunications infrastructure solutions. With software-automated Optical+Ethernet transmission technology, the Company builds the foundation for high-speed, next-generation networks. The Company s FSP product family adds scalability and intelligence to customers networks while removing complexity and cost. Thanks to reliable performance for more than 15 years, the Company has become a trusted partner for more than 250 carriers and 10,000 enterprises across the globe. Product FSP 150 ADVA Optical Networking s family of intelligent Ethernet access products provides devices for Carrier Ethernet service demarcation, extension and aggregation. It supports delivery of intelligent Ethernet services both in-region and out-ofregion. Incorporating an MEF-certified UNI and the latest OAM and advanced Etherjack demarcation capabilities, the FSP 150 products enable delivery of SLA-based services with full end-to-end assurance. Its comprehensive Syncjack technology for timing distribution, monitoring and timing service assurance opens new revenue opportunities from the delivery of synchronization services. ADVA Optical Networking North America, Inc Peachtree Industrial Blvd. Norcross, Georgia USA For more information visit us at ADVA Optical Networking SE Campus Martinsried Fraunhoferstrasse 9 a Martinsried / Munich Germany ADVA Optical Networking Singapore Pte. Ltd. 25 International Business Park # German Centre Singapore Version 07 /