Blog Header
By | March 17, 2020 | 352 views
 

Machine learning: Time-based anomaly detection (part 3)

Anomaly detection is a much-appreciated tool by data scientists. It aims to find data samples that do not conform to the regular distribution of the dataset to which they belong. Finding anomalous samples, also known as distribution outliers, provides valuable insight that often correlates with defects or errors in the data collection process (e.g., faulty or misconfigured equipment). This blog…

 
Read More
By | February 19, 2020 | 487 views
 

Network Performance Score: How to conduct NPS benchmarking campaigns

The Network Performance Score (NPS) reflects the perceived technical performance of a mobile network or one of its subsets such as region, period, or technology. The overall NPS score is the combination of carefully considered and weighted main key performance indicators (KPI) for a wide range of services that are essential and exemplary for service quality. But what needs to…

 
Read More
By | February 4, 2020 | 650 views
 

Network Performance Score: Configurator for step-by-step campaign setup

The Network Performance Score (NPS) is an integral part of our benchmarking and analytics products based on the Smart platform. In previous posts, we have discussed how the score weights the key performance indicators for a wide range of services and combines them into an overall score. With the configuration of the NPS being pertinent to reliable results, we have…

 
Read More
By | August 28, 2019 | 1,838 views
 

Network Performance Score: Templates for easy support in all products

The initial blog post about the standardized Network Performance Score (NPS) discusses the demand for an efficient calculation method of an overall score that reflects the perceived technical performance of a network or one of its subsets such as region, period, or technology. The score considers and weights the main key performance indicators (KPIs) for a wide range of services,…

 
Read More
By | May 2, 2019 | 2,879 views
 

Machine learning use case: Call Stability Score (part 2)

In our previous post about machine learning (ML), we introduced the topic of artificial intelligence (AI) and hinted at the typical deep learning pipeline that is needed to apply ML on data collected with mobile network testing solutions from Rohde & Schwarz. In this post, we will demonstrate how we can leverage ML to extract more value out of the…

 
Read More
By | January 17, 2019 | 2,261 views
 

Network Performance Score: Initiate improvements with a QoE-centric score

There are various scores, as in scoring methods, claiming to measure the performance of mobile networks on the market. Some scores are based on plain technical metrics, others on quality of experience (QoE) measurements or surveys. Which score should a mobile network operator (MNO) choose? The one providing the highest score? The cheapest one? The one based on technical scores…

 
Read More
By | November 7, 2018 | 2,077 views
 

Machine learning in mobile network testing (part 1)

Artificial intelligence (AI) has vast applications in the telecom industry, from anomaly detection to network optimization and utilization; and AI could lead to various new revenue streams. Many telecom leaders believe that machine learning (ML) will be the key to derive and infer user behavior and expectations. It aims to increase the user’s quality of experience (QoE) and reduce customer…

 
Read More
By | April 5, 2018 | 2,147 views
 

Combined engineering and benchmarking measurements add value

Different mobile network measurement tools target different measurement scenarios. One might be the preferred choice for collecting mass data for benchmarking; another might offer extensive possibilities for engineering and troubleshooting. Do we always know how the collected measurement data is being evaluated when we are performing the actual measurement? Most probably not. But wouldn’t it be great if we could…

 
Read More
By | March 19, 2018 | 2,081 views
 

Measuring QoE of mobile data apps for network optimization (part 3 of 3)

How can mobile app testing support network optimization? In previous posts, we explained how you could assess the QoE of data applications on smartphones by measuring the success rate and duration of a series of actions. Based on large-scale real field measurement data, we showed how the duration saturates at higher available network speeds and examined the causes of this…

 
Read More