GEOG2147 Lecture 3a Enabling Technologies for Smart City 2024 PDF

GEOG2147 Building Smart Cities with GIS Prof. Winnie Tang Mr. Kelvin Shum Adjunct Professor Adjunct Assistant Professor Department of Geography Lecture 3 Enabling Technologies for Smart City  Leverage on Technologies For Better City Ma...

GEOG2147 Building Smart Cities with GIS Prof. Winnie Tang Mr. Kelvin Shum Adjunct Professor Adjunct Assistant Professor Department of Geography Lecture 3 Enabling Technologies for Smart City  Leverage on Technologies For Better City Management ICT is the backbone of any society that wants to achieve “smart” status GEOG2147 | By Prof. Winnie Tang GIS 2  What is Internet of Things ( IoT ) ? IoT , Internet of Things Everything is connected Tying together the physical, digital and analytic worlds GEOG2147 | By Prof. Winnie Tang 3  Connectivity and the “ Any ” Dimension IoT allows people & things to be connected Anytime at Anyplace with Anything and Anyone using Any Network and Any Service: GEOG2147 | By Prof. Winnie Tang 4  A Comprehensive Definition from IERC - (European Research Cluster on the Internet of Things) The IERC is actively involved in ITU-T Study Group 13, which leads the work of the International Telecommunications Union (ITU) on standards for next generation networks (NGN) and future networks and has been part of the team which has formulated the following definition: “Internet of things (IoT): A global infrastructure for the information society, enabling advanced services by interconnecting (physical & virtual) things based on existing & evolving interoperable information & communication technologies. – NOTE 1 – Through the exploitation of identification, data capture, processing & communication capabilities, the IoT makes full use of things to offer services to all kinds of applications, whilst ensuring that security and privacy requirements are fulfilled. – NOTE 2 – From a broader perspective, the IoT can be perceived as a vision with technological and societal implications." The IERC definition states that IoT is "A dynamic global network infrastructure with self-configuring capabilities based on standard & interoperable communication protocols where physical & virtual “things” have identities, physical attributes, & virtual personalities & use intelligent interfaces, & are seamlessly integrated into the information network". (Source: http://www.internet-of-things-research.eu/about_iot.htm) GEOG2147 | By Prof. Winnie Tang 5  IoT Evolution Starts with only network & evolves into everything that can be connected with a network GEOG2147 | By Prof. Winnie Tang 6  Characteristics of IoT ❖ Dynamic Changes ❖ Heterogeneity ❖ Things-Related Services Characteristics of IoT ❖ Inter-Connectivity ❖ Enormous Scale GEOG2147 | By Prof. Winnie Tang 7  Architecture of IoT Information Application Integrated Application Layer Information Processing Management Layer Information Transmission Network Construction Layer Information Generation Sensor and Identification Layer GEOG2147 | By Prof. Winnie Tang 8  Sensor & Identification Layer Lowest Abstraction Layer With sensors we are creating digital nervous system. Incorporated to measure physical quantities Interconnects the physical and digital world Collects and process the real time information GEOG2147 | By Prof. Winnie Tang 9  Barcode & QR Code Low cost No technological difficulties Several devices can read a barcode Starting point for more complex systems GEOG2147 | By Prof. Winnie Tang 10  RFID The reduction in terms of size, weight, energy consumption, & cost of the radio takes us to a new era. – This allows us to integrate radios in almost all objects and thus, to add the world ‘‘anything” to the above vision which leads to the IoT concept Composed of one or more readers and tags RFID tag is a small microchip attached to an antenna It can be seen as one of the main, smallest components of IoT, that collects data Widely used in Transport and Logistics Easy to deploy: RFID tags and RFID readers The communication range and the frequency depends on the type of technology GEOG2147 | By Prof. Winnie Tang 11  RFID Applications Library Management Luggage Tracking Sports Tracking GEOG2147 | By Prof. Winnie Tang Asset Tracking IVU Inventory Management 12  Free-Flow Tolling System (FFTS) - RFID Applications of Toll Tag Free-Flow Tolling Toll Tag Congestion Charging Conceptual Design Carpark Parking Fee Electronic Road Pricing 13 GEOG2147 | By Prof. Winnie Tang https://www.ffts.hk/en  Smartphone Sensor In the near future almost everybody will probably have a smartphone A smartphone isn't just a mobile phone that has access to the Internet The smartphone can have a lot of different types of sensors GEOG2147 | By Prof. Winnie Tang 14  Network Construction Layer Robust and High performance network infrastructure Supports the communication requirements for latency, bandwidth or security Allows multiple organizations to share and use the same network independently GEOG2147 | By Prof. Winnie Tang 15  Wireless Network Telecommunication systems – Initial/primary service: mobile voice telephony – Large coverage per access point (100s of meters – 10s of kilometers) – Low/moderate data rate (10s of kbit/s – 10s of Mbits/s) – Examples: GSM, UMTS, LTE WLAN – Initial service: Wireless Ethernet extension – Moderate coverage per access point (10s – 100s meters) – Moderate/high data rate (Mbits/s – 100s) – Very common – Widely used both in indoor and outdoor environments – General purpose – Low cost – Highly interoperable – Maybe not a good solution in some special conditions – Examples: IEEE 802.11(a-n) GEOG2147 | By Prof. Winnie Tang 16  Wireless Network Wireless Network Short range: – Direct connection between devices – sensor networks – Typical low power usage – Low cost – Very long battery life – Easy to deploy – Large number of nodes (up to 64770) – Can be used globally – Secure – Ideal for WPAN and mesh networks – Support for multiple network topologies – Examples: Bluetooth, Zigbee, Z-wave (house products), NB-IoT GEOG2147 | By Prof. Winnie Tang 17  NB-IoT NB-IoT = Narrowband Internet of Things A new global standard based on public and spectrum-licensed mobile operator networks. It is a wide area wireless technology which enables ubiquitous connections. Anything can be connected to the network from anywhere, regardless of its distance. Part of the global 3GPP Release 13 standard published in June 2016 optimized for Low Power Wide Area (LPWA) applications. Advantages: – Low Power: Battery life up to 10 years – Low Cost: Cost-effective and affordable enough for mass-scale deployment – Enhanced Coverage: Covers heavily shadowed environments – Massive Deployment: Billions of connections (source: https://www.astri.org/tdprojects/narrowband-internet-of-things-nb-iot/) GEOG2147 | By Prof. Winnie Tang 18  LoRa LoRa = Long Range Radio Long range, low power consumption connection Provide secure data transmission The physical layer or the wireless modulation scheme utilized to create long distance communication link LoRaWAN: a protocol specification built on top of the LoRa technology developed by the LoRa Alliance. Use unlicensed radio spectrum in the Industrial, Scientific and Medical (ISM) bands to enable low power, wide area communication between remote sensors and gateways connected to the network. Source: http://www.rfwireless-world.com/Terminology/LoRa-technology-basics.html GEOG2147 | By Prof. Winnie Tang (Source: https://www.semtech.com/technology/lora/what-is-lora) 19  LoRa vs NB-IoT GEOG2147 | By Prof. Winnie Tang (source: https://www.linkedin.com/pulse/nb-iot-vs-lora-its-ecosystem-race-art-reed) 20  Wireless Network Other examples: – Satellite systems Global coverage Applications: audio / TV broadcast, positioning, personal communications – Broadcast systems Satellite/terrestrial Support for high speed mobiles – Fixed wireless access Several technologies including DECT, WLAN, IEEE802.16, etc. GEOG2147 | By Prof. Winnie Tang 22  5G - More Data from Billions of Devices The fifth generation of cellular networking Early products available as soon as in 2018 Dramatically increase: – Speed of data transfer – Response time – Capacity for billions of devices to be connected GEOG2147 | By Prof. Winnie Tang 23  Management Layer Capturing of periodic sensory data Data Analytics (Extracts relevant information from massive amount of raw data) Streaming Analytics (Process real time data) Ensures security and privacy of data. GEOG2147 | By Prof. Winnie Tang 26  Integrated Application Layer Provides a user interface for using IoT Different applications for various sectors like Transportation, Healthcare, Agriculture , Supply chains, Government, Retail etc. GEOG2147 | By Prof. Winnie Tang 28  Smart Home Scenario Remote Control of Smart household appliances Remote monitoring of Smart House GEOG2147 | By Prof. Winnie Tang 29  Transportation Scenario vehicle can interact with its surroundings provide valuable feedback on local roads, weather and traffic conditions to the driver condition & event detection sensors can maintain driver & passenger comfort and safety sensors for fatigue and mood monitoring, driver behavior and facial indicators can ensure safe driving GEOG2147 | By Prof. Winnie Tang 30  Healthcare Scenario - Wearables Can IoT help us to know more about ourselves? GEOG2147 | By Prof. Winnie Tang 31  Smart Waste Management Scenario Source : http://www.cs.umanitoba.ca/~comp7570/assets/media/0316Seo.pdf GEOG2147 | By Prof. Winnie Tang 32  Smart Barrier System – GIS and IoT Integration Smart Barrier allows real-time monitoring of falling debris and build up It uses a web platform and mobile app to issue alerts to the authority IoT sensors includes two depth gauge sensors, three impact switches, and two cameras Base station Digital Depth gauge Wireless camera impact switch GEOG2147 | By Prof. Winnie Tang 35  IoT Applications & Compound Applications GEOG2147 | By Prof. Winnie Tang 37  IoT Challenges Device Level Privacy Energy Issues Issues Lack of Standardization New Network Traffic Patterns Security Concerns Addressing To Handle Issues 1 2 3 4 5 6 38 GEOG2147 | By Prof. Winnie Tang  Home CCTV Camera Users Urged to Step Up Security By Consumer Council on 15 March 2017 40 GEOG2147 | By Prof. Winnie Tang  With billion of devices connected, vast amount of data can be generated Data sharing / exchange among the devices will generate new data as well Data Analytics is necessary to analyze the data to acquire insights and trends GEOG2147 | By Prof. Winnie Tang 41  What is Big Data? There is no single definition, let’s take a look from Wikipedia: – Big data is a term for data sets that are so voluminous and complex that traditional data processing application software are inadequate to deal with them. Challenges include capture, storage, analysis, data curation, search, sharing, transfer, visualization, querying, updating and information privacy. – The term "big data" tends to refer to the use of predictive analytics, user behavior analytics, or certain other advanced data analytics methods that extract value from data, and seldom to a particular size of data set. – Analysis of data sets can find new correlations to "spot business trends, prevent diseases, combat crime and so on. GEOG2147 | By Prof. Winnie Tang 42 Data Never Sleeps ! What Happens in an Internet Minute ? 43  Data Growth Exponential growth of data requires new methods of collecting, storing & processing data GEOG2147 | By Prof. Winnie Tang 44  What’s Driving the Growth of Data? 45 GEOG2147 | By Prof. Winnie Tang 45  Sources of Big Data Mobile Devices Users Applications Microphones Readers/Scanners Science facilities Programs/ Software Systems Social Media Sensors Cameras GEOG2147 | By Prof. Winnie Tang 46  Types of Big Data Activity Data Conversation Data Photo & Video Sensors Data from Image Data IoT Devices Real Time Data Spatial Data Spatiotemporal Data GEOG2147 | By Prof. Winnie Tang 47  Structure of Big Data GEOG2147 | By Prof. Winnie Tang 48  The Characteristics of Big Data 1. Volume The vast amounts of data generated every second Volume Terabytes Record/Arch 2. Velocity Transactions The speed at which new data is generated and Tables Files Value Velocity the speed at which data moves around Statistical Batch Events Real/near-time 3. Veracity Correlations 5 Vs of Processes The messiness or trustworthiness of the data Hypothetical Streams Big Data 4. Variety Veracity Variety Trustworthiness The different types of data can now be used Structured Authenticity Unstructured Origin, Reputation 5. Value Multi-factor Availability Having access to big data is no good unless we can Probabilistic Accountability turn it into value GEOG2147 | By Prof. Winnie Tang 49  How is Big Data Different? 50 GEOG2147 | By Prof. Winnie Tang 50  Traditional Analytics vs Big Data Analytics GEOG2147 | By Prof. Winnie Tang 60  Big Data & Location All IoT sensors have locations The most common IoT sensors in Smartphones – GPS receiver All posts, photos and messages are tagged with phone or IP locations in social media GEOG2147 | By Prof. Winnie Tang 62  Find Spatial Relationships Seeing spatially enabled big data on a map allows users to answer questions & ask new ones – Where are disease outbreaks occurring? – Where is insurance risk greatest given recently updated population shifts? Geographic thinking adds a new dimension to big data problem solving & helps users make sense of big data GEOG2147 | By Prof. Winnie Tang 65  Perform Predictive Modeling Predictive modeling using spatially enabled big data helps users develop strategies from if/then scenarios – Governments can use it to design disaster response plans – Natural resource managers can analyze recovery of wetlands after a disaster – Health service organizations can identify the spread of disease and ways to contain it GEOG2147 | By Prof. Winnie Tang 66  Issues of Big Data Data Data Data Discrimination Existence Security Data Data Accuracy Privacy GEOG2147 | By Prof. Winnie Tang 69  Can Aggregate Information Help? GEOG2147 | By Prof. Winnie Tang 70  Artificial Intelligence ( AI ) GEOG2147 | By Prof. Winnie Tang 71  IoT is the “Senses” , Big Data is the “Fuel” , AI is the “Brain” Data-based learning, analytics, automation AI Capture, storage, analysis of data BIG DATA Connect devices and collect data IOT GEOG2147 | By Prof. Winnie Tang 72  Artificial Intelligence (AI) we can’t survive without data GEOG2147 | By Prof. Winnie Tang https://www.aiforeducation.io/ai-resources/generative-ai-explainer 73  Evolution of AI Generative AI ChatGPT is available to public Microsoft introduces testing Cortana – intelligent personal assistant Paper about Uber pilots self- machine’s Predator UAV Apple introduces Google self-driving driving car human used by US DoD in Siri – intelligent cars-cross the 1- program in ability war personal assistant million mile mark Pittsburgh, PA autonomously 1950 1956 1995 1997 2011 2012 2014 2015 2016 2017 2022 IBM Facebook develops Term “AI” IBM Deep Watson DeepFace – near- coined at Blue defeats defeats human accuracy Google Dartmouth Garry Ken DeepMind’s by John Kasparov at Jennings AlphaGo McCarthy Chess Amazon introduces on defeats Lee Alexa – intelligent Jeopardy Sedol at Go personal assistant 1950 – 1970 1980 – 2000 2005 onwards AI as a concept, no Military & academia begin to Large tech companies invest in commercial applications real application show interest in AI of AI & ML GEOG2147 | By Prof. Winnie Tang 74  What is Generative AI ? Generative AI is a type of Artificial Intelligence (AI) that uses machine learning algorithms https://www.visualcapitalist.com/generative-ai-explained-by-ai/ to create new content like images, videos, text, and audio. https://www.visualcapitalist.com/generative-ai-explained-by-ai/ 75  “Generative AI” Fever The landscape of the generative AI market is diverse, encompassing both established tech giants and startups. ChatGPT by OpenAI (Microsoft) Bard & Gemimi by Google Azure by Microsoft Hunyuan 混元 by Tencent Tongyi Qianwen 通義千問 by Alibaba ERNIE bot 文心一言 by Baidu SenseChat 商量 by SenseTime …MORE GEOG2147 | By Prof. Winnie Tang 77  OpenAI Shows off Lifelike Videos Generated by Sora a Text-to-Video Deep Learning Model Sora : all elements mentioned in the prompt are present without any missing parts and are presented well. 60 sec 3 sec 4 sec 4 sec The Videos you are about to watch are not real 78 GEOG2147 | By Prof. Winnie Tang  Generative AI Use Cases Healthcare Medical Image Synthesis Drug Discovery Personalized Medicine for Diagnostics and Development and Treatment Plans Finance & Trading Algorithmic Trading Predictive Market High-Frequency Trading Strategies Analysis Algorithms Content Creation Art and Music Automated Content Virtual Assistants Generation Creation for Marketing and Chatbots Natural Language Processing (NLP) …MORE Text Generation for Language Translation GEOG2147 | By Prof. Winnie Tang Sentiment Analysis and Sentiment Analysis Content Creation https://www.solulab.com/top-generative-ai-use-cases/ 79  Challenges of Generative AI GEOG2147 | By Prof. Winnie Tang 81  GeoAI - Building Next-Generation Location Intelligence Tool Object Detection GeoAI Studio Geospatial Artificial Intelligence See Learn See Point Cloud Classification Object Tracking Land Cover Classification AutoML & AutoDL Read Entity Extraction GeoAI Toolbox arcgis.learn Analyze Read AI Assisted Labelling Learn Prediction Text Classification Create Clustering Classification Text to Shape Analyze Super-resolution 3D Reconstruction Forecasting Create Regression Generative Translation Model Explainability modelling Cloud removal GEOG2147 | By Prof. Winnie Tang 82  GeoAI - Helping Address Our Most Complex Challenges Making AI applications easy with ready-to-use tools & models Natural Disasters Energy 70+ Pretrained GeoAI Models for Transportation Natural Resources Elephant Detection Road Extraction (Global) Imagery, 3D and Text Utilities Ships Climate Change Public Safety Shipwrecks Cars Oil Spills Cloud Mask License Plate Blurring Cloud Masking Generation Parking Lots Water Bodies Parking Spots Pavement Cracks Humans Arctic Seals Crowd Counting Elephants Face Blurring Seabirds Land Cover Mangroves Buildings Palm Trees Roads Trees Wind Turbine Building Parcels Plant Leaf Disease Detection Footprints Ag Fields Common Object Detection Swimming Pools Text Parsing from Photo Well Pads Object Tracking Power Lines Segment Anything Model (SAM) Transmission Towers Insulator Defects... and Many More Land-Cover Wind Turbines Classification Solar Arrays Solar Park Classification Solar Panels Parking Spot Detection GEOG2147 | By Prof. Winnie Tang 83  Swimming Pool Detection using Aerial Photos Detected pools within residential parcels Tools for labelling and exporting training data Purposes - property tax assessment - maintenance companies - mosquito control GEOG2147 | By Prof. Winnie Tang Automatically find neglected pools 84  Parking Lot Vehicle Detection Visualize Results on Operations Dashboard Operations Dashboard 85 GEOG2147 | By Prof. Winnie Tang 85  GeoAI Deep Learning Model for Sunken Ships Detection in NY after storm Sandy Jamaica Bay, New York 3 100+ sunken boats By applying the same deep learning process, it can help us to identify more than 100 sunken boats in the bathymetric survey 1 data taken after storm Sandy Teach the model what we want to look for – ships in the water not 4 those that sitting on land Using the result to 2 update the NOAA Navigational Detect boats on Charts to indicate the water using the sunken ships the Mask-RCNN deep learning model GEOG2147 | By Prof. Winnie Tang 86 GeoAI Vehicle Identification from Real-time CCTVs for Traffic Monitoring Traffic Analysis from CCTVs Detect Vehicles by Type Detect Pedestrians Detect Accidents, Sudden Stops, or Other Anomalies Understand Traffic Patterns Deal with Incidents Quickly YOLO (You Only Look Once) from the arcgis.learn library Traffic flow detection using Transport Department real-time CCTV feed 87  GeoAIDetection Objects Road Features Detection for Real-time TrafficRoad Monitoring via CCTV Crack Detection feed Analysis Detect Vehicles by Type Detect Pedestrians Detect Accidents, Sudden Stops, or Other Anomalies Road Crack Detection Automate Road Crack Detection for road maintenance and Public Works Determine type of crack and potentially length and width 88  Landslide Detection with Satellite Imagery by Deep Learning Automatic identification of recent natural terrain landslides from satellite images using deep learning Landsat 8 Extract spectrum band 3, 4, 5 (green, red, & near infrared) Export Training Data for Deep Learning using GIS for model training geoprocessing Model Training Post-classification processing, converting 2D to 3D Generate the results of recent natural landslides from AI detection with Landsat data 89 GEOG2147 | By Prof. Winnie Tang  GeoAI-infused App for Tree Idenification & Inventorying in Georgia Deep Learning Citizen Engagement Collect Both Tree Locations and Species Build a Comprehensive Tree Inventory City of John’s Creek, Georgia, USA GEOG2147 | By Prof. Winnie Tang 90  Smart Cities In the AI Era Healthcare Cross Industry Applications Cyber Security Commerce AD & Marketing FinTech & Insurance Sales & CRM HR Tech Personal Assistants News, Media & Entertainment Education Agriculture Travel Legal Regtech Real Estate Sports GEOG2147 | By Prof. Winnie Tang …MORE 91  8 Ways AI will Transform our Cities Smarter by 2030 Transportation Education Healthcare Public Safety Home & Service Employment & Entertainment Low -Resource Robots Workplace Communities https://singularityhub.com/2017/11/25/8-ways-ai-will-transform-our-cities-by-2030/#sm.001crgkj01ds2exlv6g2n3h00vf5v GEOG2147 | By Prof. Winnie Tang 92  Will This Happen in The Future? GEOG2147 | By Prof. Winnie Tang 93  Or AI Will Save Us? (An example from Facebook) GEOG2147 | By Prof. Winnie Tang 94  Discussion What is the relationships between IoT, Big Data and AI? Will AI kill us or save us? GEOG2147 | By Prof. Winnie Tang 95