Self organizing networks in LTE

BANGALORE, INDIA: In the last few months with various LTE launches, there is a lot of hype on Self-organizing networks (SON) in the media. 

SON is not a new idea and has been used in optical and Ethernet networks, however, this is of interest now to the wireless LTE world with the explosion of new radio nodes such as eNodeB, femto cells and pico cells.

Also Read: Will your new iPad work on Indias 4G?

Why is the topic of SON getting important now?

Most wireless operators have an enviable position of managing three very different networks, 2G, 3G and the new LTE networks. It is expected that the new roll-out of LTE will require a parallel network of radio nodes to be configured and managed.

Some experts say that nearly 40 per cent of operator CAPEX and OPEX is spent on engineering, installation, commissioning and maintenance of these equipment. The interest can also be stemmed to the fact that world over Network Infrastructure Providers (NEP) such as Ericsson, Nokia Siemens Networks have the responsibility of roll-outs due to the nature of managed services contracts.

They hope that SON as a technology will give them some operational respite (10-20 per cent reduction of OPEX) and better gross margins.

SON offers operators the flexibility to manage their radio network more optimally through self-configuration, self-optimizing and self-healing nodes. These functionalities have been standardized through 3GPP Rel 8.TS 32.500 and Rel 9.

NGMN is an industry body, defining SON standards, which looks more of industry need from the standards. There is hope that SON implementations will be vendor neutral.

Most implementations of SON tend to be OSS or network management centric still making the solution centralized to certain degree. The later architecture proposes the algorithms to get embedded in to the nodes (network elements ) providing possibility of automatic optimization and healing decisions when required at the node.

The embedded implementations become very essential with mesh networks that will form whether its LTE radio or Metro-Wifi networks.

Large Telco operators such as Deutsche Telecom, TeliaSonera and KDDI have announced their intention in using SON in their LTE trials and roll-outs. Major LTE NEPs such as  Ericsson, Nokia Siemens Networks, Huawei, Alcatel Lucent, ZTE and NEC have announced their solutions and roadmaps. Actix, Aircom and Airhop are some of the smaller players for optimization and niche solutions.

SON promises a automated network with many benefits. Usually the real world implementations are not accurate in the first cycle and tend to be more complex. One of the key success factors is the accuracy of the optimizing and healing algorithms. Expertise and research is required to evolve the machine learning techniques and the predictive algorithms to get the effectiveness of such automation.

This could be an opportunity for academicians and entrepreneurs to come out with quick time to market solutions.

SON as a concept is also applicable for the other emerging mesh networks like Mobile Ad-hoc networks (MANETS), Metro WiFi, Wireless Sensor Networks and Wireless Vehicular networks. Its very early days, it will be interesting to wait and watch for new developments and news in this space in the next years.

The author is a consultant at Waves & Cloud Consultants.