mindtalks analytics: From analytics to the immune system of the mobile network – Ericsson – picked by mindtalks

Today, your single smartphone generates large data traffic daily – and the amazing amount of data might be transported on a whole mobile network – more than diez Tbit/s. The mobile network operators should not leave the great and continuously moving data property unused, as they can possibly not only get valuable information regarding the operation for this network via those, but they also treatment method or prevent any problems, working with those data.

The Ericsson Expert Analytics (EEA) software family, based upon Machine Learning, has been created for that purpose. It explanations the real-time data of a great deal more than 100 million subscribers together with operators located at different internet sites all over the world for the same; allowing the acquisition of key conclusions or insights about the operation, maintenance, and development involving the network. This is executed without the analysis of the particular content of the user info; the system has no such sight from the content accessed regarding the clients.

Of course, this timely operation could not be discovered without the use of advanced data mining and cloud-based technological innovation, 90% which are designed and developed in Hungary, in the Ericsson building at the Danube commercial lender. Thus, the EEA is indeed a “made in Hungary” product, which became from a research project introduced at Ericsson in Hungary around 2008, to a globally out there service, used all over often the world today. The majority associated with the developments is carried out in Hungary also today; often the company working on the next generation of EEA at the moment, which will relocate the whole of the service to a cloud-based software, terminating bare metal solutions – all the things that is of system performed focusing on machine studying.

Solve, even without human intervention

The program is essentially the immune process of the mobile network; consequently, it can not only help to make tips for the repairing of noticed errors as needed, but can independently intervene and remedy difficulties. Although this functionality is undoubtedly used by some service professional, it is still in any early stage: most locations have points now where the method passes throughout the reviewing and revive tasks of problems to peoples operators.

After an appropriate sum of testing, human intervention can certainly be excluded, and the “immune system” can work individually. The particular algorithms of EEA can study how the professionals associated with a provided with operator manage certain anomalies, trouble – different companies use different practices for similar cases ~ and will later individually choice problems in accordance with the exact samples learned from the company’s procedures. Of course, Ericsson regularly re-measures the automatized repairs, to help ensure that the operation will be smooth.

In the much time run, the industry will probably go that way, and network servicing will become fully automatized.

No mistakes will slip with the net

The company is modular, so that the mobile agency providers can select the experience they need. The analytics package runs on the number of technologies because of very simple checking mechanisms for you to complex machine learning algorithms to be able to analyze the network, predict system failures, search for anomalies, and additionally identify samples of the habits or by using devices. The process tracks in real time the activities of all users and elements of the given mobile networking, and can tell what conditions have been faced by these folks, and even what problems may occur later, quasi predicting long run events on the basis with the past samples.

The solution always tracks network errors, whether agencies like VoLTE, VoWiFi; but 5G and OTT data traffic can be monitored at all times. Professional Analytics promptly notifies the company provider if: a call is certainly disconnected, any cases of speak to fails, poor sound quality, and actually the decrease of quality with streamed videos; and it can easily detect the cases where a good user cannot access an online site. It summarizes the obtained data and makes remedy testimonials towards the service provider, describing around detail where and just how many online subscribers are affected, how serious your problem is, and whether your similar problem has occurred well before or can be expected in the future.

The analysis of typically the above-mentioned OTT data traffic, this sort of as the quality of the streamed videos or the video calls, the number of which will increases as a result of epidemic, poses important challenge for the developers. While high quality fluctuations of IMS-based calls can be relatively easily and exactly detected, in most cases it is far from that easy in case about IP-based data traffic. That is because the companies behind the particular various services usually encrypt his or her data traffic, for which typically the network provider only provides this channel and cannot go into the articles and other content. The EEA tries to resolve this problem through various heuristic algorithms, which allow the diagnosis of samples indicating the degeneration of service quality within often the encrypted data traffic.

In this regard, actually there is a race between your external content providers and the mobile providers: the former types try to hide the results in which are valuable also to them selves, and the latter ones try to extract from them the signals referring to service quality – sometimes though both parties are planning to pursue in the advance of user emotions.

From a social point connected with view, this technology can become used for very useful objectives: in some countries, the service providers used the EEA-based mobility data of the population in addition for the creation of types for the analysis of your growth of the Coronavirus.

One step before problems

Expert Analytics also offers predictive algorithms to customers, which can certainly predict problems expected within some sort of few days – these are beneficial not only for the prevention of failures, but also provide valuable information in case a great expected problem will not occur. All these machine learning algorithms continuously review the operation among the network, hence making a profile of its every day operation, including errors occurring. However, if they later detect behavior trials that caused problems before, they will likely notify the service provider’s individuals; however, in the future many people may also remedy those troubles before the deterioration of this service quality would affect your wider range of subscribers. Often the algorithms were created, tested, and designed into the software environment from a team of technological professionals, mathematicians, data scientists, and experienced software developers.

Meanwhile, the massive analytics does not compromise fragile user data; analysis is fully anonym; the product of Ericsson is wholly managed by the exact service providers and is operate on their own systems; and it does not store any individual data or any data that will can identify any users, it operates with encrypted identifiers never referring to any individual.

Ericsson recommends EEA not only for the operation of existing networks but also for your building of new technologies and devices; and the service is especially relevant now, at the time period of the expansion of 5G networks. Proper analysis of your data traffic and speed connected with new generation networks may not be solved with the traditional technologies, although EEA is prepared for their whole extensive, real-time monitoring, both in regard to commissioning and user satisfaction.

Supply: ericsson. com


mindtalks.ai ™ – mindtalks is a patented non-intrusive survey methodology that delivers immediate insights through non-intrusively posted questions on content websites (web publishers), mobile applications, and advertisements (ads). The conversation is just beginning !, click here to sign-up and connect with other mindtalkers who contribute unique insights and quality answers on this ai-picked talk.

Related Articles


Your email address will not be published. Required fields are marked *