Managing LTE Core Network Signaling Traffic
- The volume of core network signaling will increase with LTE, particularly on the Evolved Packet Core (EPC) Mobility Management Entity (MME)
- The MME can reduce signaling traffic by implementing smart signaling management techniques
- Improved network analytics help identify the cause of signaling problems
LTE signaling must be managed
Packet core signaling volumes in the early deployments of large-scale LTE networks are significantly higher than in existing 2G/3G core networks. This is partly due to the flatter, all-IP architecture of LTE where the macro and metro cell is directly connected to the MME – the dedicated control plane element in the EPC. Analysis of field data from several large LTE network deployments found that a MME can experience a sustained signaling load of over 500-800 messages per user equipment (UE) during the normal peak busy hours and up to 1500 messages per user per hour under adverse conditions.
The rise in core signaling can also be attributed to an overall increase in network usage by LTE subscribers. In some large US metropolitan markets where LTE is available, network peak usage is as high as 45 service requests per UE per hour in peak busy hours. As LTE grows in popularity, signaling in the EPC will continue to rise, which increases the potential for control plane congestion and signaling storms if not properly managed.
As a result, Mobile Network Operators (MNOs) must take steps when moving to LTE to ensure that their core control plane network can support the expected increase in signaling volume. Specifically, MNOs need to deploy a carrier-grade, next-generation MME/Serving GPRS Support Nodes (SGSNs) platform that not only has the capacity, scalability and CPU processing performance, but also the capability to intelligently manage this traffic to reduce overall core signaling. Two areas where signaling efficiencies can be gained are in MME paging and Tracking Area (TA) management procedures.
Paging and TA management challenges
Paging procedures are signaling messages between the MME, the eNodeB and the UE. Paging is required to locate a UE in the network when it is in an Idle state, making its exact location in the network unknown. Paging procedures are used by the network to:
- Request establishment of a non-access stratum (NAS) signaling connection between the MME and the UE to support a network service request.
- Prompt the UE to reattach itself to the network after a network failure.
- Initiate a mobile Circuit Switched Fallback (CSFB) procedure.
In LTE, the signaling traffic generated by MME paging the UE is significant. Figure 1 illustrates field data from a US LTE service provider in a large metropolitan market. It shows that paging is more than 28% of the total signaling load on the MME. Thus, finding methods to reduce the paging on the MME will help MNOs lower the overall network signaling load and manage their MME capital costs.
A TA represents a group of contiguous cells within the Evolved Universal Terrestrial Radio Access Network (E-UTRAN). It is used by the MME to track and locate the UE when in Idle mode as it moves through the network. A TA list is a group of adjacent TAs that is managed by the MME and periodically sent to the UE. The MME sends paging messages to the cells that are included in either the TA or TA list in which the UE is registered. A Tracking Area Update (TAU) procedure is initiated only when the UE crosses the TA boundary into another TA that is not in its TA list or when the periodic TAU timer expires.
TAU procedures can generate a lot of signaling if a UE is moving along the border between TAs that are not all part of its TA list, especially when the TA size is large (e.g. 50-100 eNodeBs in one TA). This is known as “toggling” effect because of the multiple registrations with the MME that occurs as the UE moves in and out of TA boundaries which generates additional TAU signaling.
In addition to high generating signaling volumes, both paging and TAU procedures are a significant power drain the UE battery. It is therefore important to carefully design TAs so they are neither too large to minimize the volume of paging, nor too small to prevent frequent TAUs and avoiding toggling at TA borders.
Smart paging and TA management to reduce signaling
Next-generation MME /SGSNs, such as the Alcatel-Lucent 9471 Wireless Mobility Manager (WMM), allow MNOs to tailor the MME paging policies for each type of service. This helps MNOs lower the volume of paging while still meeting end-user quality of experience expectations for each type of service.
The 9471 WMM paging policy supports the following configurable parameters:
- Paging by service type
- Number of page attempts
- Timing between page attempts
- Paging method used in each attempt
The paging policies are fully compliant with 3rd Generation Partnership Project (3GPP) standards and deployable in networks with any vendors’ eNodeBs.
As illustrated in Figure 2, smart paging and TA features can reduce signaling messaging over basic TA and TA list paging methods by as much as 80% depending on the TA size.
Dynamic TA list management techniques can further reduce MME signaling loads. These techniques optimize the number of TAs in the UE TA list as the UE moves through the network. When cyclic patterns in the UE movement are detected, the TAs are automatically added from the UE’s registered TA list as it crosses the boundaries of either 2 or 3 TAs (Figure 3). By constantly updating and optimizing the UE TA list, fewer TAU requests are generated, reducing the toggling effect at TA boundaries.
TAU procedures also drain UE battery life – consuming an estimated 10 mW of UE power per procedure in a current-generation smartphone. Table 1 shows the UE power and battery life savings for a smartphone using dynamic TA management features as compared to a smartphone using basic TA list management capabilities. This TAU procedure power estimate includes the TAU messaging as well as the scanning effort and eNodeB attachment. The comparison is for an LTE smartphone with a power consumption profile of:
- 800 mW for transmit
- 500 mW for receive
- 5.5 mW for idle
Table 1 shows that dynamic TA list management techniques consume less UE battery power and can extend smartphone battery life up to several hours depending on TA size and the subscriber’s smartphone usage.
Analytics to optimize LTE radio frequency and core networks
With LTE and its all-IP architecture, a new set of analytical tools is necessary to maintain the network. Using traditional Radio Access Network (RAN) management tools is time consuming, resource intensive and has a number of limitations:
- Drive testing provides only a sample of the network coverage at ground level. It is also time constrained and reactive.
- Service measurements aggregate data across all users. This aggregation is useful for detecting hardware faults but not for detecting individual customer performance issues.
- Using call tracing to solve specific customer complaints can require the subscriber to make test calls while troubleshooting takes place. This is not always practical. It is also time consuming and may still result in the need for a costly field visit to replicate the user behavior.
Diagnostic tools, such as the Alcatel-Lucent Per Call Measurement Data (PCMD) software capability for LTE networks, provides an integrated record of call procedures involving the UE, eNodeB and the MME. These tools are different from traditional RAN performance tools in that they utilize subscribers’ mobile devices to collect a rich set of signaling and bearer data about the device and the network. Unlike call traces, PCMD call procedures for all the UEs are captured. This provides a more accurate view of the end user’s experience with the network. In addition, because the end-user mobile devices are used to collect the data, the need for drive testers is significantly reduced.
Improved network analytics bring benefits across the MSP organization:
- Customer care: Personnel can quickly identify coverage and device issues to provide rapid responses that can improve customer satisfaction and lower call hold times.
- Operations: Personnel have the data analysis and reports needed for continuous improvement and quality management. For example, with data from end-user devices, coverage holes, dropped calls and cell performance issues can be identified and key performance indicator (KPI) targets for improvement can be set.
- Engineering and network planning: Personnel can use traffic pattern analysis with accurate radio frequency (RF) measurements to predict and verify coverage requirements. This information also assists in RF coverage and capacity planning and utilization, helping to lower capital costs with more targeted RAN investments.
New technologies demand new network capabilities
As smartphones and tablets with always-on applications and services proliferate and the competitive push to move to LTE increases, MSPs must consider how they will address signaling challenges.
A high-performance and scalable SGSN and MME that supports expected signaling volumes and provides smart paging and TA management techniques helps MSPs address LTE signaling challenges:
- Reducing signaling volumes enables MSPs to support more subscribers per MME and to defer additional MME capital investments.
- Dynamic TA management ensures that the UE always has an optimized TA list that minimizes TAUs and extends UE battery life.
In addition, advanced diagnostic and troubleshooting tools help MSPs manage and support the LTE network. MSPs can resolve customer issues faster and improve network performance to increase customer satisfaction and reduce churn.
Moving to LTE is a significant investment for MSPs. Those who make it a priority to address the signaling challenges that come with this move will be in a better position to achieve a timely return on their investment.
To contact the author or request additional information, please send an email to firstname.lastname@example.org