- Portfolio A to Z
- Customer Experience Management (CEM) for Liquid Net
- Device Management
- Enterprise Connectivity over Mobile Broadband
- Heterogeneous Networks
- Mobile Backhaul
- Mobile Site Connectivity
- Multiservice IP Backbone
- Network Sharing
- Operator Content Delivery Network (CDN)
- Policy Integration Package
- Quality of Service Differentiation
- Self Organizing Networks
- Service Operations and Management
- Smart Labs
- Telco Cloud
- Voice over LTE
- WCDMA Refarming
- Expert advantage
- Capacity Optimization
- Customer Experience Optimization
- Global Service Delivery
- Managed Service Operations & Assurance
- Mobile Broadband Performance service
- Network Implementation
- Network Modernization
- Professional Services
- Services for CEM
- Services for Heterogeneous Networks
- Services for Long Term Evolution
- Services for OTT management
- Latest launches
- Liquid Net
- News & Events
- About us
- Global presence
- Solution Experience Center
- Government relations and policies
- Contact us
- Directory of contacts
Solving the mystery of network signaling
How one research analyst found that network signaling was the root cause of rising smartphone performance issues, and the most effective solution available today.
Mobile broadband’s runaway success is due, partly anyway, to its versatility. With a USB dongle plugged into their laptop, users typically stay connected for long periods, downloading lots of data. With a smartphone, they connect for many short sessions to check email or update their social network status, consuming relatively little data.
So, when smartphones rose spectacularly in popularity, it was widely assumed that they would have little impact on network capacity. Few in the industry could have predicted the network congestion and turmoil that was to arise. Increasing numbers of users in the US and Europe started complaining about poor quality voice and data services, as well as the short battery life of their smartphones.
“Michael Thelander, founder of Signals Research Group, a respected technology-focused analyst that conducts independent research, often using advanced test equipment to investigate the performance of mobile technologies. “I provide a unique, first-hand perspective of how stuff works that helps companies develop their product roadmaps,” claims Thelander.”
But with the average smartphone generating only about a sixth of the traffic of a laptop, how could they be degrading the quality of an entire 3G network? While several theories were put forward, it took a January 2010 report from the US-based research consultancy Signals Research Group to pinpoint the problem.
"I had seen the press reports about network congestion. Some people in the industry suspected the cause, others thought the issue was in backhaul or lay elsewhere. I decided to investigate using some new test equipment I had access to," explains Michael Thelander of Signals Research Group.
The hidden signaling load
The problem identified by Signals Research Group lies in the high volume of signaling that smartphones generate. Smartphones make many connections to the network, carrying small amounts of data each time. Some email applications, for example, can be set to look for new mails as often as every 30 seconds. Each time any device connects to the network, no matter how much data is involved, there is background signaling traffic that opens and closes the data session.
Many always-on smartphone applications generate eight times as much signaling traffic as laptops generate. And when the network elements that handle signaling traffic overload, they are no longer able to support additional data or voice calls, leading to the degradation of quality seen in many smartphone-heavy networks.
"I didn’t believe the magnitude of the problem until I saw it for myself. I didn’t expect to see the degree to which applications generate enormous amounts of signaling, I had no idea that was going on. It was a surprising result and even after I published the report some in the industry still didn’t believe it," says Thelander.
To further investigate the issue, Nokia Siemens Networks commissioned Signals Research Group to undertake a second study examining the differences in smartphone-generated signaling traffic between a Nokia Siemens Networks commercial 3G network in which a feature called Cell_PCH was enabled, and one in which it was not. The research also looked at the impact on handset battery life.
Keeping a smartphone in active data transmission mode for long periods eats battery life. This is the challenge facing operators today: To ensure the quality of experience for all their customers by managing smartphone signaling traffic volume, while simultaneously ensuring the longest possible handset battery lifetime.
Published in May 2010, the new Signals Research Group report found that the Nokia Siemens Networks network with Cell_PCH enabled generated an average of 40 percent fewer signals from highsignaling applications than the other supplier’s network, with a simultaneous battery life increase of up to 30 percent.
Managing signaling differently
Conventional networks keep an unused handset in Idle mode, which uses little power. When the handset connects to the network, 30 signaling messages are needed to activate the handset to transmit data and to return it to Idle again. Doing all this makes the user wait an average of two seconds for the initial connection, so the network is set to instruct the device to stay in active mode for a little longer, to give a fast response if the user requests more data.
However, active mode drains the handset battery. Therefore, some handset manufacturers implement ‘fast dormancy’ that actively returns the handset to the battery-saving Idle state quickly after data transmission. While this lengthens battery life, it also generates more signals that the network must manage.
Nokia Siemens Networks handles signaling differently. To begin with, handsets not in use are not kept in Idle state, but in another state called Cell_PCH which uses no more battery power than the Idle state. However, between three and 12 signals only are needed to move the handset into the Active state for sending data. Because there are fewer signals, the delay for the user is only 0.5 second on average. This is a more acceptable delay than two seconds, enabling networks with shorter timer settings, taking handsets back down to Cell_PCH quickly and providing the low power consumption that other networks achieve with fast dormancy.
Fewer signals and a shorter active-state timer setting mean that Nokia Siemens Networks, alone among all network suppliers today, offers operators less signaling and reduced battery consumption simultaneously.
"Of all the solutions to the problem that exist, this has the most ‘bang for the buck,’ in particular when dealing with chatty smartphone applications that frequently transmit and receive relatively small amounts of data, without any associated ill consequences," concludes Thelander in his report.