End-to-End Planning for LTE Backhaul Networks
By Satish Gujar
With the accelerated rollout of LTE in their networks, mobile operators face challenges in sizing their backhaul to meet not just higher data rates but also increased signaling traffic. Traditional planning and capacity-forecasting methods have relied on techniques such as trending based on historical data, but given the vastly different bandwidth and signaling requirements for an LTE network, these techniques have limited value in predicting future capacity requirements.
The flat architecture of LTE networks also presents new challenges in backhaul design. Some of these challenges are outlined below.
In order to meet the raw capacity and increased throughput requirements of LTE traffic and deliver it at the lowest cost per bit, operators have to now also consider additional constraints of delay, loss and latency of data in their network plans. Differentiated traffic such as signaling, conversational voice and streaming media and video are all included in the LTE traffic mix. Operators also need to consider aggregated traffic, which includes user-plane traffic, control-plane traffic, X2 and Operations Administration and Management (OAM) messages. Operators will have to confront the following issues:
Operators will continue to have to support multiple technology types (2G/3G and LTE) in the backhaul until the older technology is phased out. Newer backhaul technologies are typically packet-based, so strict delay and loss constraints must be maintained to preserve quality of service and the user experience. Additionally, the varied capacity requirements of each coexistent technology continues to be a challenge.
Backhaul distances in 3G networks have ranged from tens to 100 kilometers, whereas in LTE networks they can range above hundreds of kilometers. The longer distances impose additional constraints on meeting delay requirements.
LTE architecture defines X2 interfaces between ENodeBs to facilitate handovers, which increases both signaling and data traffic. Rather than implement a full physical interconnected mesh, operators will need to implement access and aggregation nodes to minimize the connectivity requirements.
User behavior on mobile networks has changed from that of voice-centric to more data- and app-driven. Some smartphone apps in particular put a disproportionate demand on signaling capacity by dipping into the network with small payload messages.
Operators will increasingly deploy small cells to increase capacity, however the deployment scenarios become increasingly complex with multiple nontraditional access technologies such as Non-line of Sight Radio.
Network operators need planning solutions that enable them to integrate mobile-network planning across the access and core networks along with backhaul planning. Planning systems must offer functionality and flexibility to model Key Performance Indicators (KPIs) that are relevant to the unique challenges of LTE network planning. In addition to the traditional capacity planning KPIs of data throughput and bandwidth utilization, planning systems must also be able to calculate complex KPIs like CPU utilization, which are a true indicator of signaling capacity.
Rapid changes in technology within operator networks, such as the adoption of LTE, require a planning solution that is adaptable and flexible enough to deal with these changes. Operators need solutions that are designed to meet the challenges of a multi-technology, multi-vendor network allowing operators to perform end-to-end planning on all parts of the LTE network, including backhaul. VPIsystems’ OnePlan, for instance, performs all of these functions for mobile operators.
Satish Gujar is senior manager, sales engineering, at VPIsystems , which provides network analytics for capacity planning, network profitability, business metrics and demand analysis . He has 17 years of enterprise software experience in large-scale, high availability systems, working with globally distributed development groups and customers. Before joining VPIsystems, Satish was a project lead and senior software engineer at Telcordia. Contact him at satish.gujar@VPIsystems.com.