Sanjay Nayak Issue: India 2011
Article no.: 6
Topic: Mobile Backhaul – the key to 3G in India
Author: Sanjay Nayak
Title: CEO & MD
Organisation: Tejas Networks
PDF size: 218KB

About author

Sanjay Nayak, is the Co-founder and CEO of Tejas Networks, Bangalore. Mr Nayak is a technocrat with over 23 years of experience ranging across the telecom, networking and Semiconductor industries in India and the USA. Prior to founding Tejas Networks, Mr Nayak was Managing Director of Synopys India. Mr Nayak is a frequent speaker on the subjects of telecom industry, innovation and entrepreneurship from India. He is on the governing council of Government of India’s Telecom Export Promotion Council (TEPC), Executive Council of Indian Semiconductor Association (ISA), CII National Committee on ICTE and a Trustee of Deshpande Foundation for Social Entrepreneurship (India). Sanjay Nayak earned a M.S. degree in ECE from North Carolina State University and B.E in ECE from the Birla Institute of Technology, Mesra.

Article abstract

The success of 3G in India depends upon the ability of the backhaul networks to deal with the expected increase in traffic. Using existing infrastructure to keep costs low, while adopting next-generation Carrier Ethernet packet transport technologies that can drastically reduce the cost per bit without compromising reliability or performance is the best available strategy. For now, operators can use Carrier Ethernet networks for best-effort Internet traffic and existing mobile backhaul networks for more quality sensitive services including voice.

Full Article

Operators in India have to focus on two main areas to ride the 3G wave profitably: packaging and pricing their 3G offerings to convert their high-paying 2G/2.5G customers to 3G and building their mobile backhaul network to handle the imminent 3G data explosion. If, AT&T’s five thousand per cent iPhone-driven traffic increase is reproduced in India 3G broadband service demand will explode and backhaul traffic will tax available network resources. Mobile backhaul Backhaul has grown over the years from being a mere traffic delivery pipe to an application enabler. So far, a significant portion of the investments by Indian operators’ went to build basic wireless infrastructure, but going forward the focus will be upon scalable, flexible and cost-effective mobile backhaul infrastructure based on optical fibre or microwave radios according to circumstances. With 3G, the share of data services in Indian operator’s revenues will increase, although voice will continue to be the major contributor. Operators will need to strike a balance between providing such new-age data services as high-speed Internet access, mobile TV, video calling etc., and the need to continue providing basic voice service and data services such as SMS, GPRS etc. all on the same backhaul network. An intelligent mobile backhaul solution can help them strike this balance. Mobile backhaul can enable Indian 3G operators to: • Achieve ROI quickly – Time to market is a key determinant of how quickly an operator can recover his investment in spectrum license fees. • Ensure quality 3G services – Operators can give their users bandwidth guarantees and assure superior QoS enabled by supporting features such as HQoS (hierarchical quality of service) in the backhaul equipment • Prevent disruption to 2G business – We expect that even in 2015, 60 per cent of mobile connections will still use 2.5G, so upgrading the networks to 3G should not disrupt 2G operations • Reuse existing network assets – Existing Indian telecom networks, built predominantly for voice traffic, will require innovative solutions to extend their lifespan by reusing such infrastructure components as microwave radios, optical transmission equipments and fibre on the ground to support data, VAS and content services. • Achieve network convergence – Traditionally, Indian telecom networks were built and operated in silos with separate overlays to transport mobile services, fixed-line voice and leased line traffic. The growth in packet traffic on 3G networks will strain this deployment model and operators will look to synergistically combine these disparate networks to maximize network utilization. Backhaul needs 3G operators in India will use a variety of approaches to deploy 3G backhaul networks. The differences will be based on their existing network infrastructure. The few large operators own end-to-end networks, but some of the newer operators lease bandwidth from third parties. One category of operators, known as greenfield operators, do not have legacy 2G backhaul networks to support during 3G rollouts, so they can rollout a pure packet Carrier Ethernet network. Another category, the brownfield operators, have existing 2G backhaul networks and voice customers that need continuous support. They need sophisticated packet processing capabilities they can use on existing infrastructure to minimize disruption of their current revenue generating services. A third category will keep their 2G backhaul and 3G backhaul networks separate. Both voice and data contribute significantly to the revenues of these operators and they have enough voice and data traffic to justify two separate networks or because their mobile voice services and broadband are separate business units. Backhaul approaches 3G operators would take a three phase approach to build backhaul capacity – optimize-offload-overhaul. A brownfield operator would start with the first phase, while a greenfield operator and an operator that wants separate overlay networks would jump in at the third phase. Optimize – The current backhaul infrastructure of most carriers in India makes extensive use of PDH and SDH microwave in the access layer. 3G/LTE is expected to strain these networks given the high-bandwidth requirements of multimedia traffic, but the business models to help monetize packet transmission are not yet sufficiently mature. Hence operators are reluctant to make significant investments in new transport equipments or in new fibre to meet the surge in 3G/LTE traffic. Another option is to buy additional frequency spots to expand available microwave capacity. However, congestion in the 6-38 GHz microwave band and the burden of annually recurring spectrum charges for new frequency blocks are dissuading operators from taking this path. A backhaul optimization of three new technologies to help carriers maximize reuse of this PDH/SDH infrastructure while minimizing upgrade costs is necessary. Some methods such as unlocking hidden bandwidth in protection channels of current SDH Microwave rings, aggregating PDH and SDH into Ethernet and open ring ERPS, that provides sub-50ms recovery even without reserving protection bandwidth can be used to free up access bandwidth for 3G. Packet Optical transport equipment helps build this optimization by supporting Voice (TDM) and Data (Packet) processing natively. Offload – When packet traffic has increased sufficiently and there is greater predictability in data service usage, carriers can start offloading data to an overlay network. The overlay can use a physically separated, different, fibre strand or a different microwave frequency or it could be virtually created on the existing network. There are a variety of ways to create overlays depending on the specific situation of the operator and his network constraints, such as: i) deploying IP microwave equipments on an additional microwave carrier, ii) using Ethernet modules on existing hybrid microwave radios and sharing the air interface with TDM, or iii) partial rollout of new fibre in locations with high 3G usage and installing Gigabit Ethernet rings using Carrier Ethernet equipment. The advantage of option iii) is that the operator can use advanced packet handling capabilities in these products to start up-selling higher-margin services and tailored 3G plans to the existing subscriber base. A key feature of Carrier Ethernet standards i.e. Hierarchical QoS which enables different quality of service parameters to be set for different types of data traffic, will be useful for option iii) of this phase. Overhaul – In Phase II of the backhaul migration, voice and TDM leased line services have native support. This is reasonable since mobile voice on TDM will continue to be an important revenue driver for the next several years and cannot be compromised. However, with voice itself using packet interfaces (Voice over IP) in newer 2G BTS equipments, the carrier can choose to retire its legacy network at an appropriate time. The residual TDM traffic can then be transmitted on emulated circuits over pure packet backhaul networks using special mapping techniques that also maintain accurate synchronization information for the emulated TDM circuits. In the absence of a circuit-based network, timing delivery and transport to the LTE eNodeBs (Evolved Node B supports LTE air interfaces and handles radio resource management) will be critical. TDM requires low latency and jitter for reliable transmission and this requires guaranteeing adequate bandwidth, prioritization and buffering on ‘engineered’ and protected paths that minimize packet losses, delays and reordering. TDM emulation in LTE also impose additional expectations typical of circuit services such as performance monitoring, service restoration, and service quality. Using Carrier Ethernet, circuit emulation allows voice to be carried over a data network. Packet synchronization assures voice users a high level QoE (quality of experience); ‘protected tunnels’ created in the data network replicate the resilience of voice networks. ___________________ In the near and medium term, operators will continue to witness a dramatic increase in traffic with 3G services like video applications (over-the-top, video calling, mobile TV) and high-speed Internet access. The rapid divergence between network traffic volumes and costs associated with transporting this content will present significant operational and business challenges to operators in the coming years. Operators have to prepare for this surge by adopting next-generation packet transport technologies based on Carrier Ethernet that can drastically reduce the cost per bit without compromising on fundamental transport requirements like service reliability, determinism, fault management and performance assurance. The recommended approach would be to start this transition by offloading best-effort Internet traffic to a more cost-effective Carrier Ethernet network layer and retaining the more high-value QoS sensitive services including voice on the existing mobile backhaul network. At an appropriate time, when the business models for the new packet-based services are more mature, the legacy infrastructure may be retired and completely replaced by an end-to-end Carrier Ethernet transport.