Networking and Customer Experience focuses on inventing technologies that revolutionize how networks and services are architected, transforming network efficiency and enabling the rapid creation of innovative new services.
Service Quality Management (SQM)
Using machine learning, we created a Service Quality Management (SQM) program focused on ensuring that customer experience is at the forefront of how we design and operate networks and services. The suite of SQM innovations mine vast amounts of data. They identify, quantify, and predict customer impact while prioritizing and automatically resolving issues. Technologies include Argus (event detection), Achilles (isolation of service issues), TONA (impact quantification) and service to network correlation. AT&T’s newest capability, ECOMP (Enhanced, Control, Orchestration and Management Policy), has created a “closed loop control” that detects issues and automatically repairs them within minutes.
Network Performance, Reliability and Optimization
Performance, reliability and optimization are at the heart of AT&T’s services. Using the power of SDN and Network Virtualization, we apply data science and optimization techniques to ensure that AT&T delivers highly reliable services with optimal resource efficiency. We design 5 9’s services on commercial-grade clouds, and use the power of SDN to create self-optimizing, reconfigurable data networks that quickly adapt to failures and surges with minimal additional capacity. We accurately forecast network demands so that the optimal amount of capacity can be deployed to the places that need it, and we calibrate the location of our network resources to meet our stringent reliability and performance requirements.
Data Powered Customer Experience
Our customers are always top of mind at AT&T, and our data-science experts focus on improving customer experience through data collection and analysis. We analyze the largest, most comprehensive integrated surveys using a segmented analytic approach across all ways in which customers interact with our services, along with objective data for modeling and metrics (e.g., CQI). This end-to-end integrated approach allows us to evaluate perceptions and behavior to anticipate future needs and impacts, continuously improving our customers’ experience.
Predictive Impact Analysis for Designing a Resilient Cellular Backhaul Network
Sen Yang, Yan He, Zihui Ge, Dongmei Wang, Jun Xu
ACM SIGMETRICS, 2018
Coordinating Rolling Software Upgrades for Cellular Networks
Mubashir Adnan Qureshi, Ajay Mahimkar, Lili Qiu, Zihui Ge, Max Zhang, Ioannis Broustis
IEEE ICNP, 2017, Best paper award
ParaBox: Exploiting Parallelism for Virtual Network Functions in Service Chaining
Yang Zhang, Bilal Anwer, Vijay Gopalakrishnan, Bo Han, Joshua Reich, Aman Shaikh, Zhi-Li Zhang
Symposium on SDN Research (SOSR), 2017
AESOP: Automatic Policy Learning for Predicting and Mitigating Network Service Impairments
Supratim Deb, Zihui Ge, Sastry Isukapalli, Sarat Puthenpura, Shobha Venkataraman, He Yan, Jennifer Yates
ACM KDD, 2017
LiveJack: Elastic CDNs-Edge Clouds Integration for Live Content Broadcasts
Shu Shi, Rittwik Jana, Bo Yan, Yong Liu, Haoqin He, Weizhe Yuan, Yang Xu, H. Jonathan Chao
ACM Multimedia, 2017
Rapid Detection of Maintenance Induced Changes in Service Performance
Ajay Mahimkar, Zihui Ge, Jia Wang, Jennifer Yates, Yin Zhang, Joanne Emmons, Brian Huntley, Mark Stockert
ACM CoNEXT, 2011, Best paper award
Communication Networks: A Concise Introduction, Second Edition
Jean Walrand, Shyam Parekh
Morgan and Claypool, 2018