Machine Intelligence for Telecom and Beyond

Size: px
Start display at page:

Download "Machine Intelligence for Telecom and Beyond"

Transcription

1 Machine Intelligence for Telecom and Beyond Elena Fersman Machine Intelligence and Automation Ericsson Research

2 Do you know what s common between these? Cortana Lucida Mika Amelia Alexa Siri Watson

3 Data Lake Cortana Lucida Mika Amelia Alexa Siri Watson

4 Hi, what is Machine Intelligence? Hi Elena, according to Ericsson definition, Machine Intelligence is a combination of MachineLearning and Artificial Intelligence.

5 What s MSDP? MSDP = Multicast Source Discovery Protocol MSDP = Multiservice Delivery Platform MSDP = Managed Services Delivery Platform

6 What are your sources? Ericsson Product Catalogue, CPI Store, Wikipedia, and your calendar

7 Tell me when site X is likely to require a preventative maintenance visit? Judging from the site profile learned over time, and the current site behavior, the maintenance will be needed in Q

8 What actions did you do for Operator A s network last month? I changed parameter configuration in the 5G network nodes. I also sent offers to subscribers with high risk of churn.

9 Why? Because we need to stay competitive towards our enterprise customers while keeping our subscribers happy.

10 The value is in data Cortana Lucida Mika Amelia Alexa Siri Watson

11 From raw data to action

12 Data and its processing

13 Technology landscape in machine learning/ai Machine Learning Artificial Intelligence Driven by massive amounts of data managed by data science Driven by facts and knowledge managed by logic

14 Integrated Management IoT Management IoT Integrated Infrastructures Networks & Cloud Data Events Control Business Support Processes Knowledge Analytics Operations Support Connected Processes & Data Management Network Infrastructure Business Support Industrial AI and Cognitive Technologies Automation Decision support Knowledge management Semantics Machine learning Analytics Data management Data Events Control Operations Support Processes Knowledge Analytics Operator

15 Ericsson at a glance Enabling the full value of connectivity for service providers Business areas: Networks Digital services Managed services Technology and emerging business By the numbers: 180+ countries 201 BSEK in sales 100,700 employees 45,000 patents Image: Ericsson headquarters, Kista, Sweden Sara Mazur Ericsson

16 Ericsson Research innovate together 2G, 3G, 4G, 5G it s all invented here >50% involvement in all of Ericsson patents Ecosystem collaboration with operators, academia and industries >180 countries >100k employees a unique Ericsson footprint Sara Mazur Ericsson

17 Creating the full value of connectivity through world-leading research Radio Network Architecture & Protocols Hardware, Device & EMF Cloud Systems & Platforms Security Machine Intelligence & Automation Digital Representation & Interaction Consumer & Industry Lab Programs Research areas Mission Ericsson Research Sara Mazur Ericsson G for industries Horizontal IoT technologies and solutions

18 5G a foundation for digitalization 1G 2G cutting the cord, adding mobility going digital 3G 4G 5G adding video & data mobile broadband 29 billion devices 5 billion people 1 billion places Telephony and mobile broadband A digital infrastructure for industrial and societal transformation Sara Mazur Ericsson

19 SENSORS EVERYWHERE 5g USE CASES BROADBAND AND MEDIA EVERYWHERE SMART VEHICLES, TRANSPORT INFRASTRUCTURE, MONITOR AND CONTROL CRITICAL CONTROL OF REMOTE DEVICES INTERACTION HUMAN-IOT

20 What is 5G what will it bring A Network for the Networked Society x 1000x 5x 100x Cost dB END-USER DATA RATES MOBILE DATA VOLUMES LOWER LATENCY MORE DEVICES DEVICE COST REDUCTION YEARS BATTERY LIFE BETTER COVERAGE Management Access Applications Cloud Infrastructure Transport One architecture supporting multiple industries

21 Technologies that drive industry changes Machine intelligence Cognitive technologies and deep learning Responsible machine intelligence Capsule networks Automation Model driven Automated life cycle management Autonomous systems Radio evolution mm Wave and massive antenna technologies Multi-purpose, multicharacteristic radio Flexible spectrum assignment and utilization Programmable networks Software defined networking Network abstraction Network slicing Cloud technologies Distributed cloud and edge computing Micro services and DevOps Virtualization, containers

22 ONE DAY IN THE LIFE OF A MEDIUM SIZED NETWORK (~10M CUSTOMERS) Web pages 700,000,000 Internet sessions (PDP) 66,000,000 Videos 40,000,000 Radio sessions (RAB) 120,000, more types of events Handovers (HSDSCH-CC) 300,000,000 Sum data 10à100 TB/day Real-time data rate 100,000à1,000,000 events/second

23 Complexity in the connected world self configuring no admin e.g. transportation systems are increasingly complex and full of sensors, smart devices and steered by multiple services running on both devices and cloud the integration of components like trucks, trailers and containers shall be simple

24

25 Data, Information and Knowledge

26 Types of Knowledge Declarative Knowledge aka Propositional Knowledge = knowing THAT something is the case Formally described as Ontologies Example: Volvo S60 is a car Procedural Knowledge aka Imperative Knowledge = knowing HOW something is done Formally described as State Machines Example: pressing gas pedal makes the car accelerate

27 Machine intelligence for Cyber-Physical SystemS Technologies and Components Examples white-label Enterprise Telco presentation application Applied Context Business Business objectives objectives Maximize Optimize User grid utilization Experience Decision Support and Knowledge Management IoT Resource Management Data Analytics Networks & Devices Sensors & Actuators Declarative and Procedural Knowledge Real World Model Analytics Observation and Control Things and Places Abstraction and Semantics Reasoning and Formal Methods Reasoning and Formal Methods phase network drift capacity is X - increasing is appropriate price by for Y youtube for area Z consumption would reduce Knowledge consumption increasing management, average 5%, stabilizing bandwidth verification grid with alternatively and 10% synthesis, would increase reduce reasoning, hydro churn workflow power by optimization 50% by 2% Semantics Semantics and and context context Smart meter for user Cell XXX high overload reactive power outtake Semantic Device annotation, Substation YYY in Norra Entity CELL 2 XXX of IED 0.1 outage Mbit/s modeling, Processed data feeds Average D/M/Y throughput pattern and of average all devices of Sensor in the 5 City: is: Y 1.5 (curve) Mbit/s Stream Real-time analytics, Anomaly Pattern monitoring detected recognition, of cell in Sensor XXX Anomaly shows 8 YYY detection Raw data feeds Raw data feeds Sensor Cell XXX 5 value = 3.4 = kw22 Data Device2 capture, throughput Sensor Actuator 8 = level 0 commands, kw= 1.1 Mbit/s Cloud1 = Valve 60% 2 processor = 12 load Available Available assets assets Cell Smart XXX, meter, Site 4, 10kV Pico 15, Distribution Site Controller Substation, 7, Antenna Energy 3, Service RBS6000 Interface (ESI), Photovoltaic/Hydro, #2, VM1 (cloud1) Electrical vehicle

28 Data and its processing Expert User Actions Approver Knowledge Data Processing Mission connected things Knowledge Base Generated Knowledge Planning & Reasoning Strategy Online Learning Explanation networks documents Model Repository Data Lake & Streams Predictions &Estimation Deep & Shallow Learning Analysis Property check What-if analysis Actions Raw Data

29 Use case White label GUI 1 3 High-level objective Would you mind A*? 4 Ok! Intelligence Models Analytics (numbers) Analysis (models) AI techniques 2 N Current status of the world 6 Actuation Analysis / Planning Knowledge Vault CORE RAN 5G 4,5G 4G 3G 2G Capilar nws

30 Example: AR assistant Field Service Operations made easy Anyone can maintain an Ericsson Site Augmented-reality based mobile application for guiding engineers to troubleshoot RBS issues Ericsson Site Analytics-powered backend for alarm capture and action recommendations Rich-content knowledge base of CPI documents for fault resolution. Cross-collaboration Ericsson Research MI Technologies BA Managed Services Problem definitions BA Networks RBS Product owners, CPI and operations

31 Telco Knowledge Graph Telco Knowledge Graph Intelligent Applications Product Documentation Digitized Expert Knowledge Insights Machine Learning Data Mobile Digital Assistant Network Information Contextual Knowledge Structured, Linked Data Learning Models, Rules, Procedures Informed Decision- Making Intelligent Network Operations Centre

32 Summary Data-driven and datacentric research Dealing with heterogeneity though semantics Right data at right time and place Keeping the global state together Mix of MI approaches and techniques ML meets Reasoning Declarative meets Procedural Collaborative Intelligence Ensuring safety and trust Modern NW for variety of use cases Mix of industries and telecom Challenges: complexity, isolation, differentiated QoS NW slicing and Automation

33 Questions & Answers

34