Telecommunication networks deliver a massive volume of data to individuals and institutions day after day. Network operations also generate valuable data about the network's own health and status. Similarly, records of the activities they support convey much about the subscribers and devices they serve. In each context, there are opportunities to drive information up the value chain, transforming raw data into knowledge. Innovative use of data analytic methodologies can therefore be central to enhancements of telecom networks' capabilities and value.

Success stories from high profile Internet companies in recent years have highlighted the potential of the methods and tools that can deal with data on massive scales (so called Big Data). Data from telecom networks are of similar scales, and they often require real-time processing. Indeed useful insights from telecom data need to be uncovered within a short timeframe so actions in the network can be taken without delay. In many scenarios, the demand for real-time processing means that the analysis must be done locally on geographically distributed platforms. Another challenge is that the sensitive nature of certain telecom data demands treatments that can preserve user privacy and ensure information security, while allowing useful inferences to be made.

This issue of the Bell Labs Technical Journal focuses on innovative and practical techniques and application case studies relevant to data analysis in the telecom context. The papers selected for the issue jointly represent the ongoing work related to this theme in the research and development (R&D) communities within Alcatel-Lucent.

The first group of papers addresses the need to understand the impact of network performance on customer experience and the privacy concerns in mining and utilizing subscribers' network usage.

“Using Big Data to Improve Customer Experience and Business Performance” by Spiess et al. discusses concepts from the Big Data theme and the opportunities and challenges such technologies present in understanding telecom customers and improving their satisfaction. “A Multi-Layer Dynamic Model for Customer Experience Analytics” by Chen et al. presents a statistical model for correlating observations across multiple layers that enables inference of a subscriber's unobservable, subjective perception of network performance. It also touches on how anonymization techniques may serve to encourage better sharing of experience-revealing service data by the service providers. A deeper dive into privacy-preservation techniques is given in “Exploring the Impact of LSH Parameters in Privacy-Preserving Personalization” by Aghasaryan et al. Finally, “QoE Model for Video Delivered Over an LTE Network Using HTTP Adaptive Streaming” by De Vriendt et al. discusses quality of experience for mobile video, a type of traffic that is particularly sensitive to network performance.

The second group of papers describes specific use cases where data analysis technologies are developed and applied to extract valuable information from the signal streams transmitted by telecom networks.

Continuing on the video theme, “Surveillance Video Analysis Using Compressive Sensing With Low Latency” by Jiang et al. presents a novel technique for extracting moving objects from a surveillance video. “Joint Sequence Complexity Analysis: Application to Social Networks Information Flow” by Milioris and Jacquet explores the notion of joint sequence complexity for Markov sources, and discusses its uses in discriminating in real time without overhead between text streams from different sources. “Enterprise Social Networking Data Analytics Within Alcatel-Lucent” by Friedman et al. presents an analysis of activities in our enterprise social network, which reveals interesting usage patterns that could be of good reference value for managing employee communications. For either people or devices, localization indoors has been a challenge unaddressed by the widespread GPS technology. “Probabilistic Radio-Frequency Fingerprinting and Localization on the Run” by Mirowski et al. describes a set of methods for mapping indoor wireless signals and using such maps for localization purposes. Another example of monitoring and modeling physical signals on a network is given in “Demand Forecasting in Smart Grids” by Mirowski et al.

The last two papers in the issue deal with practical implementations of data analysis techniques in the context of monitoring network operations. “Measuring and Simulating Cellular Switching System IP Traffic” by Del Signore describes a method to summarize observed sequences of connection events associated with a base station, which produces a representative model that can be used to simulate traffic load for performance studies. “Towards the Optimization of a Parallel Streaming Engine for Telco Applications” by Theeten et al. discusses challenges in finding an optimal configuration for a streaming analysis engine that ensures rapid responses to performance issues in real time. This is one step in our progress to bring Big Data platforms onto the center stage in telecom data processing.

(Manuscript approved October 2013)

Biographical Information

  1. Top of page
  2. Biographical Information
Thumbnail image of

TIN KAM HO leads the Statistics of Communication Systems Research Activity in Bell Labs at Murray Hill. She pioneered research in multiple classifier systems, random decision forests, and data complexity analysis, and pursued applications of automatic learning in many areas of science and engineering. She also led major efforts on modeling and monitoring large-scale optical transmission systems. Recently she worked on wireless geo-location, video surveillance, smart grid data mining, and customer experience modeling. Her contributions were recognized by a Bell Labs President's Gold Award and two Bell Labs Teamwork Awards, a Young Scientist Award in 1999, and the 2008 Pierre Devijver Award for Statistical Pattern Recognition. She is an elected Fellow of IAPR (International Association for Pattern Recognition) and IEEE, and served as editor-in-chief of the journal Pattern Recognition Letters in 2004-2010. She received a Ph.D. in computer science from State University of New York (SUNY), Buffalo.

Thumbnail image of

PHILIPPE JACQUET is ingénieur général de Mines and director of the Mathematics of Dynamic Networks and Information Department at Alcatel-Lucent Bell Labs in Nozay, France. He received the engineering diploma from the École Polytechnique (X78), a Ph.D. from Paris XI University, and an HDR from the University of Versailles Saint-Quentin-en-Yvelines. Prior to joining Bell Labs, he was scientific head of the Hipercom team located within two member institutions, INRIA and École Polytechnique ParisTech, where he also serves as a part time professor. The Hipercom Team is known worldwide as the creator of the Optimized Link State Routing protocol (OLSR) for Mobile Ad hoc NETwork (MANET). This protocol and associated protocol suite is now the basis of many wireless networks for civil and military applications. Dr. Jacquet has authored more than 300 papers, which collectively have garnered over 9,000 citations. He has published in mobile networking, pattern matching algorithms, information theory and analysis of algorithms. In 2004 Philippe Jacquet and Paul Mühlethaler received the French prize “Prix Science et Défense” for their contribution to mobile networking.