Nokia Siemens to tackle overload with base-station smarts

Nokia Siemens Networks has developed software that can shift users among cells to prevent spot slowdowns

Nokia Siemens Networks says it can prevent LTE base stations from getting overloaded while extra capacity in nearby cells goes to waste, even if the cells use different types of spectrum.

At Mobile World Congress this week in Barcelona, the company is demonstrating a patented algorithm for load-balancing among LTE base stations. The algorithm is part of the intelligence that NSN gear can use to analyze and act on information about the load on each base station.

Carriers place cells to meet their best guess about where the most people will use their phones and other devices, but network demand can flare up suddenly in certain areas. NSN's new algorithm can look at what cells are heavily loaded or barely used and find subscribers who are close to more than one base station. Then it can make smart decisions about moving some of those users to a less-loaded cell, according to Hua Xu, a senior specialist in NSN's Technology & Strategy Office, who developed the algorithm with a team of researchers.

Load-balancing among cells is a fairly new phenomenon, and NSN's new software goes a step further by doing it between cells that use two different types of LTE spectrum, FDD (Frequency-Division Duplexing) and TD (Time-Division). While FDD is used in almost all LTE networks today, TD is now emerging as a mainstream approach to the fast network technology. It uses one block of frequencies rather than separate blocks for upstream and downstream data. The allocation of TD spectrum for LTE networks in China is expected to create a big volume market for TD-LTE cells and phones, and other carriers and countries are also gearing up to use that system.

For consumers, the load-balancing system could prevent the kinds of slowdowns that often hit in crowded areas with a lot of other mobile users nearby. For mobile operators, it could cut the cost of building networks, because some base-station upgrades to add needed capacity could be put off until a later time. Other nearby cells would deliver that capacity in the meantime, said Lasse Makinen, a TD-LTE sales manager at NSN.

The system can provide systemwide optimization across multiple cells, including both full-size macro cells and small cells, Xu said. While its ability to work with both FDD and TD base stations is a highlight, it will also work on networks that use only one of the systems. Base stations themselves can run the load-balancing software without having to send the work to servers in the core of the network.

There are limits to how much such a system can ease capacity crunches, Xu said. Even though the software can analyze loads across many cells, it can only move users among adjacent cells that are within signal range. In networks where TD-LTE base stations are added to an FDD network for extra capacity, only users with dual-mode devices will be able to take advantage of load-balancing between those technologies.

The 3GPP, which oversees the LTE standard, defines how base stations should communicate information about their current load to other cells. The value of NSN's software is in how it analyzes that data to draw conclusions about when to shift users' devices between cells, Xu said. The team that she leads from Arlington Heights, Illinois, near Chicago, developed the new algorithm last year. It will probably go into an official software release next year, Makinen said.

It's too early for most carriers to worry about load-balancing on LTE networks because the user loads on them aren't that high yet, Makinen said. However, as more consumers buy LTE-capable devices and start taking advantage of the technology's throughput, it's like that carriers will start running into capacity issues soon, he said.

Stephen Lawson covers mobile, storage and networking technologies for The IDG News Service. Follow Stephen on Twitter at @sdlawsonmedia. Stephen's e-mail address is stephen_lawson@idg.com

What’s wrong? The new clean desk test
Join the discussion
Be the first to comment on this article. Our Commenting Policies