PhD Defense: Enabling Smart Homes Using Web Technologies

  • 1,630 views
Uploaded on

New technological advancements allow the Internet to penetrate in embedded computing. IPv6 envisions to merge the physical and the digital world, through the Internet. The Web of Things interconnects …

New technological advancements allow the Internet to penetrate in embedded computing. IPv6 envisions to merge the physical and the digital world, through the Internet. The Web of Things interconnects the expanding ecosystem of Internet-enabled embedded devices, by reusing well-accepted and understood Web principles. In this talk, we will present the development of a Web-based application framework for smart homes, supporting concurrent interaction from multiple family members. By employing intermediate request queues, associated with the physical devices of the smart home, our analysis shows that we can mask transmission failures and faults that occur in the wireless environment, thus enhancing the performance of smart home operations by means of fast retransmissions, load balancing and request priority techniques. In our analysis, we also derive formulas for estimating the response time of requests and for setting the request queue retransmission interval, an important design parameter of the system. In this way, reliability and timely responses from the devices are ensured. We demonstrate that, by using the Web as application layer, flexible applications for smart homes can be built, on top of heterogeneous embedded devices, with little effort. We address many issues related to Web-enabling household devices, from their local discovery and service description to the uniform interaction with them. Our technical evaluation indicates that the process of Web-enabling physical devices offers satisfactory performance, mainly in terms of response time and energy consumption, while modern Web techniques such as Web caching and event-based Web messaging can contribute in facilitating smart home operations. Through various case studies, we demonstrate that Web-based, energy-aware smart homes have the potential to provide flexible solutions to challenges such as energy awareness and conservation, and be smoothly integrated with the smart grid of electricity. Finally, this talk discusses some future research challenges, beyond the home environment, in which Web-based smart homes may constitute crucial elements in order to address them effectively.

More in: Technology
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
No Downloads

Views

Total Views
1,630
On Slideshare
0
From Embeds
0
Number of Embeds
2

Actions

Shares
Downloads
41
Comments
0
Likes
3

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. Enabling Smart Homes Using Web Technologies Andreas Kamilaris PhD Defense Professor: Andreas Pitsillides NETworks Research Laboratory Department of Computer Science University of Cyprus December 2012
  • 2. General Overview University of Cyprus1. Introduction - Problem Statement - Motivation2. Building an Application Framework for Smart Homes3. Web-Enabling Home Devices4. Using Request Queues for Enhanced Performance5. Technical Evaluation6. Blending Smart Homes with Online Social Networking7. Integrating Smart Homes to the Smart Grid8. Beyond the Smart Home Environment9. Conclusion10. Future Work
  • 3. Introduction University of Cyprus• Merging of computing with physical objects.• Physical devices becoming smarter.• Home appliances equipped with embedded microprocessors, wireless transceivers, sensors and actuators.• New automation possibilities in smart homes.• Technology disappears in the background of residents’ lives.
  • 4. Problem Statement University of Cyprus • In an idealized vision of a fully integrated smart home, all the operations are efficiently controlled by a central application. • Proliferation of incompatible standards/protocols used by hardware/software manufacturers. • The smooth integration of appliances from different vendors becomes a very complex process.
  • 5. Motivation: Internet of Things University of Cyprus• The Internet penetrates in embedded computing.• The Internet of Things envisions a network of objects, where all things are uniquely and universally addressable, identified and managed by computers in the same way humans can.
  • 6. Motivation University of Cyprus X10 KNX ZigBee IPv6Network Size: 2^8 2^16 2^16 2^64 per subnetData Rate: 20b/s 9.6kb/s 20-250kb/s 250kb/s...1Gb/sInterface: custom app-level app-level UDP, TCP, solutions gateway gateway RESTful WebCost: low high medium lowInstallation Overhead: low high low lowConnectivity: low medium medium highSecurity: none high medium medium“Internet technology, utilizing IPv6, will become the future standard in home automation.”Matthias Kovatsch et al., Embedding Internet Technology for Home Automation, in Proceedings ofETFA, Bilbao, Spain, September 2010.Carles Gomez and Josep Paradells. Wireless home automation networks: A survey of architectures and technologies. IEEECommunications Magazine, 48(6):92{101, 2010.
  • 7. Motivation: Web of Things University of Cyprus• Interconnecting embedded devices in application level.• The Web of Things reuses Web principles to interconnect embedded devices, built into smart things.• The Web as a pervasive and scalable platform.The WoT practice:1. Connect embedded devices to the Internet, via IPv4 or IPv6.1. Embed Web servers on the devices.2. Model their services in a resource-oriented way (REST).
  • 8. Motivation: REST University of CyprusREST is a lightweight architectural style which defines how toproperly use the HTTP protocol as an application interface.REST Vs Big Web Services (WS-*)A Resource-oriented Architecture is about four concepts:1. Resources.2. Their names (URIs).3. The links between them.4. Their representations (HTML, JSON, XML).Resources can be manipulated with:1. GET to retrieve a representation of a resource.2. POST represents an insert or update.3. PUT to alter the state of a resource.4. DELETE to delete resources.
  • 9. Thesis Objective University of Cyprus Investigate the feasibility of enabling truly Web-based smart homes, following the principles of the Web of Things, for achieving interoperability, flexibility and acceptable performance in home environments.
  • 10. General Overview University of Cyprus1. Introduction - Problem Statement - Motivation2. Building an Application Framework for Smart Homes3. Web-Enabling Home Devices4. Using Request Queues for Enhanced Performance5. Technical Evaluation6. Blending Smart Homes with Online Social Networking7. Integrating Smart Homes to the Smart Grid8. Beyond the Smart Home Environment9. Conclusion10. Future Work
  • 11. Requirements for Future Smart Homes University of Cyprus • Support various interaction types (Ad hoc, pull, push). • Uniform access to heterogeneous home devices. • Multi-hop wireless communication, plug and play functionality. • Reliable operation, masking transmission failures. • Flexible design, light implementation. • Direct access for residents to their home environment. • Concurrent, multi-resident support. • Interoperable programming interfaces for end user development of smart home applications. • Graphical user interfaces. • Small waiting times for simultaneous requests. • Acceptable performance in terms of response times and battery lifetime of devices. • Scalability.
  • 12. Embedded Devices in the Smart Home University of CyprusI. Sensor Motes• Telosb sensor motes.• Equipped with a 250kbps, 2.4GHz, IEEE 802.15.4-compliant Chipcon CC2420 Radio, integrated on-board antenna and a 8MHz TI MSP430 microcontroller with 10 kB RAM.• Equipped with temperature, humidity and light sensors.• Form a wireless sensor network inside the smart home.
  • 13. Embedded Devices in the Smart Home University of CyprusII. Smart Power Outlets• Ploggs• Firmware based on the ZigBee protocol.• Forming a mesh metering network.• High accuracy of electricity measurements.• Control of an electrical appliance remotely (switch it on/off).• External transducers for whole-home measurements.
  • 14. Building an Application Framework for Smart Homes University of Cyprus
  • 15. Building an Application Framework for Smart Homes University of Cyprus• Synchronous/Asynchronous Operation
  • 16. Building a Web-based Smart Home University of Cyprus• A Web cache is an intermediary between Web servers and clients, monitoring incoming requests and saving copies of the responses for itself.• Works only for GET requests using the expiration model for determining freshness of resources. Gateway Web Cache(Web client) (Web cache) (Web server)
  • 17. Building a Web-based Smart Home University of Cyprus• Adding a Graphical User Interface HomeWeb Client Application Restlet-GWT XML JSON Web API Server Application Framework Restlet
  • 18. Building a Web-based Smart Home University of Cyprus
  • 19. Building a Web-based Smart Home University of Cyprus• Energy Awareness through a Web Interface• Device-level energy consumption information.• Historical comparison with previous days, weeks, months, years.• Association of electricity data with costs.
  • 20. Building a Web-based Smart Home University of Cyprus• Smart Rules for Home AutomationIf the illumination in the living room is less than 50% and thetemperature in the kitchen is greater than 25 degrees Celsius,then turn on the red led of sensor7 and the green led of sensor6.
  • 21. Building a Web-based Smart Home University of Cyprus• Smart Rules for Energy Efficiency “If the illumination in the living room is less than 20%, then turn off the television and the DVD player.”
  • 22. Building a Web-based Smart Home University of Cyprus• Web mashups are Web resources that include content and application functionality through composition of existing resources.• Physical mashups exploit real-world Web services, offered by physical devices, combining them using the same tools and techniques of Web mashups. function check { if [ $? -gt 24 ] ; then curl -d "status=OFF" -X PUT [serverAddress]/AirConditioner/Switch/ fi } curl -s -X GET [serverAddress]/Kitchen/Temperature/ $1 check;
  • 23. Building a Web-based Smart Home University of Cyprusfunction getHttpRequest() { var xmlhttp = null; xmlhttp.open(GET, http:// [serverAddress]/ /touchatag/tags, true); xmlhttp.onreadystatechange = function() { if(xmlhttp.readyState == 4 && xmlhttp.status == 200) { var items = eval(( + xmlhttp.responseText + )); var secret_key = "04:BA:4A:B9:23:25:80"; for (var i=0; i<items.length; i = i + 1) { if (items[i].t == secret_key) $(locked).innerHTML = "Door is unlocked!"; } if (!unlocked) $(locked).innerHTML = "Door still locked."; } } xmlhttp.send(null);}
  • 24. Building a Web-based Smart Home University of CyprusUrban mashups, defined as opportunistic physical mashups, validatedwhen the local environmental conditions support the sensor-basedWeb services, defined by the mashups.
  • 25. Building a Web-based Smart Home University of Cyprus• Automatic reasoning and advanced knowledge inference about environmental conditions. Back
  • 26. General Overview University of Cyprus1. Introduction - Problem Statement - Motivation2. Building an Application Framework for Smart Homes3. Web-Enabling Home Devices4. Using Request Queues for Enhanced Performance5. Technical Evaluation6. Blending Smart Homes with Online Social Networking7. Integrating Smart Homes to the Smart Grid8. Beyond the Smart Home Environment9. Conclusion10. Future Work
  • 27. Web-enabling Sensor Devices University of Cyprus1. Local Device Discovery • Multicast Discovery Protocol. • Transmit a single URL instead of a SOAP/XML payload. Similar to WS- Discovery for WS-*.
  • 28. Web-enabling Sensor Devices University of Cyprus2. Service Description • Web Applications Description Language (WADL). • An XML-based language that provides a machine- readable description of HTTP-based applications. Similar to WSDL for WS-*.
  • 29. Web-enabling Sensor Devices University of Cyprus3. Request-Response Interaction • Each device an embedded Web server, offering sensor and actuation services through a RESTful interface. • All interactions are done via standard HTTP calls. • IPv6-based WSN of Telosb sensor motes (6LoWPAN). • Based on blip in TinyOS 2.x • Indirect Web-enablement of Ploggs. Resource URI Parameters REST Method Return Value Temperature: - GET text/plain Humidity: - GET text/plain Illumination: - GET text/plain Leds: Color (Red, PUT Ack Green, Blue) Electricity: - GET JSON Switch: Status (On, Off) PUT Ack
  • 30. Web-enabling Sensor Devices University of CyprusConnecting Ploggs to EnergieVisible• EnergieVisible plots sensor data in real-time.• Pointed EnergieVisible to our application framework, by feeding it with measurements from Ploggs in JSON format.
  • 31. Web-enabling Sensor Devices University of Cyprus4. Web Messaging for Event-driven Scenarios • Pull Vs Push. • Client-server model not appropriate for event-driven scenarios. • RESTful Message System (RMS) is a push-based, lightweight publish/subscribe messaging framework, suitable for embedded devices. • A RESTful API on sensor motes, for managing subscriptions.
  • 32. Web-enabling Sensor Devices University of Cyprus5. Device and Service Sharing • Can be achieved by harnessing social networking sites.
  • 33. Web-enabling Sensor Devices University of Cyprus6. Other Issues • Global Web discovery of devices/services. • Security of devices against misuse. • Privacy of owners and users of the devices. • Semantics of sensor information.
  • 34. General Overview University of Cyprus1. Introduction - Problem Statement - Motivation2. Building an Application Framework for Smart Homes3. Web-Enabling Home Devices4. Using Request Queues for Enhanced Performance5. Technical Evaluation6. Blending Smart Homes with Online Social Networking7. Integrating Smart Homes to the Smart Grid8. Beyond the Smart Home Environment9. Conclusion10. Future Work
  • 35. Using Request Queues University of Cyprus• Transmission failures happen in indoor environments.• Home devices have battery limitations and frequent failures.• No guarantees. High unpredictability.• Better management of the interactions with embedded devices.• Reliability and performance need to be ensured.• Request queues a suitable intermediate data structure for enhancing the performance of pervasive applications that target smart homes.
  • 36. Using Request Queues University of Cyprus• Request queues are defined as FIFO queues.• Installed on middleware applications for smart homes.• Handle the requests coming from the home’s tenants, targeting the embedded devices of the house.
  • 37. Request Queue Operation University of Cyprus
  • 38. Request Queue Analysis University of Cyprus• Incoming requests need to wait in the queue for their turn, in order to be executed.• Waiting time at the queue is a considerable amount of time in increased workload.
  • 39. Experimental Setup University of Cyprus• 6LoWPAN-enabled Telosb sensor motes.• Sensing capabilities exposed as RESTful Web services.• Transmission power at -25 dBm, message sizes 128 bytes.
  • 40. Request Queue Analysis University of Cyprus• Request queue retransmission interval α Average RTT St. Deviation
  • 41. Request Queue Analysis University of Cyprus• Influence of transmission failures and different arrival rates on retransmission interval α.
  • 42. Request Queue Analysis University of Cyprus• Influence of varied response times on retransmission interval α.• RTT times and St. Deviation values learned from the device thread.• Set initially to a larger value, leaving a "safe margin”.• Fine-tune adaptively during the device operation.
  • 43. Potential Benefits of Request Queues University of Cyprus• Multi-Client Support• Multi-hop topology additional delays of around 200 ms.• Heavy workload increases response times by 18-20% in the single- hop case and 14-17% in the multi-hop topology.
  • 44. Potential Benefits of Request Queues University of Cyprus• Avoiding Transmission Failures• In light workload, transmission failures do not affect significantly the response times.• In heavy workloads, transmission failures cause the response times to grow almost exponentially.
  • 45. Potential Benefits of Request Queues University of Cyprus• Estimating Potential Response Times• Average estimation error is 12.38%, and it increases to 14.60% when the request queue size becomes larger than 2.
  • 46. Potential Benefits of Request Queues University of Cyprus• Load Balancing for Serving High Traffic 1. Whenever a new request appears for a service, estimate the potential response time for every device that offers service. 2: Forward the request to the embedded device that offers the least estimated response time.• In low traffic, the improvement of performance is around 4-6%.• In increased traffic, the improvement reaches 11%.
  • 47. Potential Benefits of Request Queues University of Cyprus• Handling Priorities 1. Assign a priority to each request (low, normal or high). 2. Each prioritized request is translated into an integer value by the application framework, according to the following formula: low=1, normal=5 and high=10. 3. The priority heap selects for execution the request with the highest priority number. 4. To avoid starvation of low-priority requests, the priorities of all waiting requests are increased by 1 at every round, i.e. at a successful execution of some request.
  • 48. General Overview University of Cyprus1. Introduction - Problem Statement - Motivation2. Building an Application Framework for Smart Homes3. Web-Enabling Home Devices4. Using Request Queues for Enhanced Performance5. Technical Evaluation6. Blending Smart Homes with Online Social Networking7. Integrating Smart Homes to the Smart Grid8. Beyond the Smart Home Environment9. Conclusion10. Future Work
  • 49. Technical Evaluation University of CyprusPerformance Metrics• Average Request/Response Time (ms)• Energy Consumption (J)• Cache Hits (%)
  • 50. Technical Evaluation University of CyprusA scenario of a family with many members• Multiple family members interact with their home devices through the Web.• Home residents modeled as software agents.• Each agent assigned a unique TCP socket.• Agents select randomly devices/services.• The arrival rate lambda of family members is modelled by the exponential distribution.• Five minutes simulation time.
  • 51. Experimental Setup University of Cyprus A single-hop Topology A multi-hop Topology• A single-hop topology does not cover a typical smart home.• The indoor range of a typical sensor mote is 20-30 m.• A multi-hop 2-3-4 tree topology. 4 leaf IPv6-enabled sensors.
  • 52. Technical Evaluation University of Cyprus A single-hop Topology A multi-hop Topology• Simulated REST Vs 6LoWPAN REST.• Simulated REST message sizes 80 bytes.• The effect of Web caching.
  • 53. Technical Evaluation University of Cyprus• Web Caching• Increase is higher between 10-25 seconds of freshness time.• Cache hits reach 90% with freshness time around a minute.
  • 54. Technical Evaluation University of Cyprus• Web Caching• Small increases in freshness time reduce the request time linearly with a large gradient.• At freshness time of 60 seconds response time only 100 ms.
  • 55. Technical Evaluation University of Cyprus• Energy Consumption Performance• Avrora Simulator for AVR microcontrollers.• Energy consumption of the leaf nodes in the multi-hop 2-3-4 topology of IPv6-enabled sensor motes. Bigger energy consumption due to the 6LoWPAN packet overhead
  • 56. Technical Evaluation University of Cyprus• Event Handling• Push Vs Pull Web Messaging• A fall in illumination levels once every five minutes.• Pull frequency 30 seconds. 23% less energy consumed in Push messaging
  • 57. General Overview University of Cyprus1. Introduction - Problem Statement - Motivation2. Building an Application Framework for Smart Homes3. Web-Enabling Home Devices4. Using Request Queues for Enhanced Performance5. Technical Evaluation6. Blending Smart Homes with Online Social Networking7. Integrating Smart Homes to the Smart Grid8. Beyond the Smart Home Environment9. Conclusion10. Future Work
  • 58. Smart Homes and Online Social Networking University of Cyprus• Two thirds of global Internet Population visit Social Networking Sites (SNS).• Facebook has more than one billion active users!Social networking has become a fundamentalpart of the global online experience.The Web 2.0 is a social Web!
  • 59. Smart Homes and Online Social Networking University of Cyprus• Sharing Home Devices through Online Social Networking
  • 60. Smart Homes and Online Social Networking University of Cyprus• A Pub/Sub Mechanism with Facebook Notifications
  • 61. Smart Homes and Online Social Networking University of Cyprus• A social competition between neighbouring flats towards efficient energy utilization.• Exploring entertainment through a social game and the social influence of the neighbourhood, as parameters for energy conservation.• The influence of the community has the potential to drive residents towards a persistent behavioural change.
  • 62. Smart Homes and Online Social Networking University of Cyprus • Social Norms “People tend to follow what other people do and adapt their behaviour and practices according to the stimuli received by their friends, relatives and neighbours”. “Social norms can motivate people to question their attitude, if they discover it is not ”normal”.
  • 63. Smart Homes and Online Social Networking University of Cyprus• Rules and Conditions• The competition would take place in blocks of flats.• The duration of the competition is one month.• Daily feedback about each flat’s and the whole block’s consumption, flat ranking in the competition.• Feedback through a website, a Facebook application and a notice box located at the main lobby.• The winning flat is the flat reducing most effectively its electrical consumption.• Comparison with electricity bills from previous months.• The award to the winning flat would be an energy monitor.
  • 64. Smart Homes and Online Social Networking University of Cyprus Plogg Smart Power Outlet Flat Mains Meter Flats’ Microsoft SQL Residents Server database Web Server
  • 65. Smart Homes and Online Social Networking University of Cyprus• Case study: Two blocks of flats Suburb Urban Flats: 10 20 Participating Flats: 6 20 Residents: 10 29 Age 18-25: 2 10 Age 26-35: 6 12 Age 36-45: 2 4 Age 46-55: - 3 Age 56+: - -
  • 66. Smart Homes and Online Social Networking Suburban Block of Flats Urban Block of Flats• Comparing with previous months, 11.90% average reduction of energy in the suburban case and 27.74% for the urban block.• Average energy savings in the urban case are 2.4x more.• In the urban block, an increase of temperature by 2 degrees Celsius.
  • 67. Suburban Vs Urban Block University of Cyprus Suburban Block Urban Block• Strong correlation of daily temperature to block’s energy consumption.• People at the suburban block consumed in average 11% more energy.• Residents at the urban block mostly highly educated students.• Demographic analysis based on age, sex, yearly income and number residents per flat.
  • 68. Demographic Analysis: Age University of Cyprus• Older people consume more electricity, as they spend most of their time at home.• Age groups 26-35 and 46-55 are mostly influenced by the competition, reducing their consumption by 32%.• It may be more convenient for people that spend much time at home, to observe and analyze their consumption, taking countermeasures.
  • 69. Demographics Analysis: Sex University of Cyprus• Females tend to consume more electricity as they (usually) spend more time at home, having energy-demanding habits.• Women have contributed more in saving energy, reaching 30% reductions, while men around 20%.• In general, females were more interested in the competition. They found the perspective of protecting the environment appealing.
  • 70. Demographic Analysis: Yearly Income University of Cyprus• Tenants with high income consumed more energy as they were not willing to sacrifice their comfort just for saving money.• Residents with low income consumed less than half the energy of their high-income neighbours.• Low-income residents had the least savings, as they probably had already tried to save energy in the past, to reduce their costs.• High-income residents reduced their consumption by 30%, motivated because of environmental reasons and not to save money.
  • 71. Demographic Analysis: Residents per Flat University of Cyprus• More tenants at each flat implies more consumption.• While this difference is more significant when comparing flats having one or two residents, reaching 44%, it becomes very small between flats of two and three residents, around 4%.• One-tenant flats achieved 30% savings, since it is easier for someone living alone to develop his own energy-efficient practices.• The margin of potential savings is much bigger in one-bedroom flats.
  • 72. Reasons for Increased Consumption University of Cyprus• Flat 302, 27 years old man, suburban block “I am not willing to sacrifice my comfort to save energy and money. I do not encounter financial problems”.• Flat 102, male student, urban block “I have my computer equipment working 24/7, and I can not do much about it. Using more energy-efficient infrastructure is out of my budget”.• Flat 305, 31 years old man, urban block “I want my flat warm the whole day and I earn a good salary to afford that”.
  • 73. General Statistics University of Cyprus• 72% of tenants stated they were actively involved with the competition and this helped them to acquire a more sustainable lifestyle.• 94% believed that this competition will influence them to save energy in the future.• 69% considered that the method of comparing consumption with neighbours is a promising way for saving energy.• 48% used the website for being updated about the competition.• All residents checked the information placed in their notice boxes.• The Facebook application was used by 15% of people.• 89% wanted to be informed in real-time about their consumption.• From them, 88% were willing to buy a product that would show them their consumption in real-time. They would invest at most 70 Euro for such a product.• Some of them were surprised when we explained to them that this is possible at these costs. Some people did not even know that such products exist.
  • 74. Suggestions from Residents University of Cyprus• Feedback through SMS, sent by the utility once per day.• Daily feedback through email.• More detailed electricity bills.• Smart incentives to people to save energy by the government.• Similar competitions with awards from the utilities.• Scalar pricing schemes that reward green flats and houses while punishing energy-wasting buildings.• Grants from the utilities or the government for renewable energy systems and green lighting.• More pervasive and real-time energy feedback techniques.
  • 75. Smart Homes and Online Social Networking University of Cyprus In most European countries, including Cyprus, people receive an electricity bill once every month. It is not easy for them to perceive their electricity footprint, i.e. to understand whether their consumption is low, medium or high.Citizens need an effective way torealize the “semantics” of theirelectrical consumption!
  • 76. Smart Homes and Online Social Networking University of Cyprus• Energy Awareness through Social Comparisons Social comparisons may enable people to perceive the amounts of their consumed electrical energy, by comparing it with their social and local environment! Electrical information is aggregated in neighbourhood level promising platforms Online social networking sites constitute (PO code, street). to locate people and discover their social networks.
  • 77. Social Electricity University of Cyprus Social Electricity Facebook application Electricity data is real and accurate, provided by the Electricity Authority of Cyprus (EAC). Electrical information is aggregated in neighbourhood level (PO code, street).
  • 78. Social Electricity: Features University of Cyprus• Personal Comparisons: Compare your electricity footprint with the average amount of electricity consumed at your neighbourhood, village/town or the whole of Cyprus.
  • 79. Social Electricity: Features University of Cyprus• Social Comparisons: Compare the electrical consumption at your street with that consumed by the streets of your friends, who are tagged on the map of Cyprus where they live.
  • 80. Social Electricity: Features University of Cyprus• Location-based Statistics: Observe the most and least energy efficient streets in your neighbourhood as well as the most and least energy efficient areas and villages around Cyprus.
  • 81. Social Electricity: Features University of Cyprus• Historical Comparisons: Compare the energy behaviour of your street in previous months or at the same month in previous years. Make this comparison more social by including the energy behaviour of your friends’ streets.
  • 82. Social Electricity: Initial Facts University of Cyprus The application started officially at 1st August 2012. More than 1,000 users after 4 months, 1,280 likes on our Facebook page. Eponymous supporters like the Interior Minister Mrs Eleni Mavrou and the Commissioner for the Environment Mr Charalambos Theopemptou. The most popular group of users (39%) is between 25-34 years old. Younger people between 18-24 are also highly interested (32%). 65% of users live in an urban environment. 48% of users live in the capital of Cyprus, Nicosia.
  • 83. Social Electricity: Initial Facts University of Cyprus Extensive reportages and publicity in large media of Cyprus (TV channels, radio channels, newspapers, magazines, online blogs).
  • 84. Social Electricity: Initial Facts University of Cyprus First prize award at the 2nd Green ICT Application Challenge, organized by the International Telecommunication Union (ITU).
  • 85. Social Electricity: Initial Evaluation University of Cyprus • Via questionnaires. 178 subjects. 44% found the app very useful. 36% found it just useful. 55% were affected positively to become more energy-aware. 16% believed their energy consumption was high. 78% perceived their “energy profile” through the app. 62% claimed their consumption was reduced in regard to last year. 57% are aware of their “green” and “red” friends. 48% used the app from curiosity. 71% for environmental reasons. 71% for financial reasons. 14% for responsibility as a citizen. Most popular incentive for energy reduction is discount on the bill. 38% believe the app will be more useful in a few years. 64% believe the app will reduce their consumption more than 10%.
  • 86. Social Electricity: Current Work University of Cyprus A newsletter sent by email to the users of the application every two months, to inform them about their electricity footprint, comparing it with their local and social environment. (42% opened the email, 13.3% opened the app) Mobile apps for mobile Facebook users.
  • 87. Social Electricity: Current Work University of Cyprus More effective electrical comparisons between people that share common house preferences (e.g. home size, number of residents, heat type). Access to Social Electricity by people who do not have Facebook through a Web site that offers location-based statistics and general information.
  • 88. General Overview University of Cyprus1. Introduction - Problem Statement - Motivation2. Building an Application Framework for Smart Homes3. Web-Enabling Home Devices4. Using Request Queues for Enhanced Performance5. Technical Evaluation6. Blending Smart Homes with Online Social Networking7. Integrating Smart Homes to the Smart Grid8. Beyond the Smart Home Environment9. Conclusion10. Future Work
  • 89. Integrating Smart Homes to the Smart Grid University of Cyprus• The Smart Grid of Electricity• A smart grid describes the future electricity grid, enhanced with ICT and smart metering, applied to generation, delivery and consumption of electric power.
  • 90. Integrating Smart Homes to the Smart Grid University of Cyprus• The role of smart homes at the future Smart Grid• Great potential towards a coordinated, large-scale plan for energy efficiency.
  • 91. Integrating Smart Homes to the Smart Grid University of Cyprus• Web-based, Grid-ready Smart Homes Resource URL Method Parameters Return Value HouseName/electricity GET - JSONHouseName/reduceconsumption POST reduction (Integer) text/plainHouseName/increaseconsumption POST maxincrease (Integer) text/plain
  • 92. Integrating Smart Homes to the Smart Grid University of Cyprus• Device Categorization according to usage patterns• Permanent devices: Should never be turned off.• On-demand devices: Utilized by home residents spontaneously.• Schedulable Devices: Devices that are supposed to accomplish some specific task, but their operation is not momentarily urgent and can be postponed for a future time.
  • 93. Demand Response University of CyprusOffering dynamic tariffs, according to supply conditionsand current demand..• Schedule electricity-related tasks for future execution.
  • 94. Demand Response University of Cyprus• Example Scenario: Perform the washing when the tariff falls below 5%.• A RESTful Web server simulates DR functionality for EAC.• Event-based push notification of tariff changes in real-time.• Integration of a task scheduling mechanism to the framework.• Define the duration of each task, max. amount of waiting time etc.
  • 95. Demand Response University of Cyprus• Case Study: Identify schedulable devices and usage patterns• Monthly savings can be summed €6.00 in 10% tariff reduction, up to €19.00 in case of 30% reduction (using a tariff 20,07 cent/kWh).• Possible reduction in the bill of a typical home 3-10%.
  • 96. Load Shedding University of CyprusAn action taken to prevent frequency abnormal operation and is thelast resort to maintain frequency stability in case of contingencyscenarios or autonomous-islanded operation.conditions.
  • 97. Load Shedding: Controller Simulation University of Cyprus• Virtual Phasor Measurement Units have been assumed, capable of measuring the frequency and its first derivative online in real-time.• System parameters were identified by applying the Lyapunov Synthesis Method.• Based on a simple linear second-order system frequency response (SFR) model.• Identify the parameters of the plant by employing a suitable Lyapunov function, in terms of state variables and time, forcing this function to be at least negative semi-definite in order to obtain the desirable stability.• A (low-level) smart grid controller was simulated on Simulink, its main task was to maintain the stability of the system by determining the optimal (per unit) amount of electric load that should be shed to achieve frequency stability. Parameter Value Simulator Time Step 0.27424 msec Total Simulation time 150 sec Sampling Rate for Domestic 350 msec Consumption Total Number of HTTP requests 428
  • 98. Load Shedding University of Cyprus• Permitted limits +/- 3 Hz.• No load shedding causes under-frequency abnormal operation.• Conventional practices exceed the desired levels for 4 seconds and need 35 seconds to "absorb" the disturbance.
  • 99. Integrating Smart Homes to the Smart Grid University of Cyprus• Other potential applications:• Peak Leveling/Shaving• Fault Tolerance• Billing• A Market for Generation/Consumption of Electricity
  • 100. General Overview University of Cyprus1. Introduction - Problem Statement - Motivation2. Building an Application Framework for Smart Homes3. Web-Enabling Home Devices4. Using Request Queues for Enhanced Performance5. Technical Evaluation6. Blending Smart Homes with Online Social Networking7. Integrating Smart Homes to the Smart Grid8. Beyond the Smart Home Environment9. Conclusion10. Future Work
  • 101. Beyond the Smart Home University of Cyprus• From smart homes to smart spaces
  • 102. Beyond the Smart Home University of Cyprus• Online Social Networking of the Physical World
  • 103. Beyond the Smart Home University of CyprusEvaluation:• Nine workers at the farm - two weeks duration of experiment.Impressions:• “The application is easy to be used.”• “Excited with the perspective of controlling the greenhouse while amusing with my friends.”• “The notifications are quite difficult to understand.”• “The user must be online to be notified!”• “I increased my monitoring activity.”• “I became more aware about the farm.”• “How much does it cost to fully automate the farm?”• “This can be applied also in health monitoring!”
  • 104. Beyond the Smart Home University of Cyprus• Environmental Awareness in the Urban Environment • A “meta-smart home” perspective. • Web-based smart homes as the foundational elements for shaping the next-generation digital cities. • Community-based, real-time sensor sharing. • Location-based discovery of sensors through Cosm. • The UrbanRadar mobile application.
  • 105. Beyond the Smart Home University of Cyprus• Global Discovery of Environmental Services
  • 106. Beyond the Smart Home University of Cyprus• Global Discovery of Environmental Services
  • 107. Beyond the Smart Home University of Cyprus• Global Discovery of Environmental Services• Exploit the existing Internet infrastructure to achieve real-time discovery of embedded devices and environmental services.
  • 108. Beyond the Smart Home University of Cyprus• Global Discovery of Environmental Services 61.50 61.00 60.50 60.00 msec 59.50 59.00 58.50 58.00 57.50 57.00 100 500 1000 10000 100000 1M sensors
  • 109. Conclusion: General Overview University of Cyprus• A flexible application-level solution for home automation based on combining Web technologies.
  • 110. Conclusion: Meeting Requirements University of Cyprus • Support various interaction types (Ad hoc, pull, push). • Uniform access to heterogeneous home devices. • Multi-hop wireless communication, plug and play functionality. • Reliable operation, masking transmission failures. • Flexible design, light implementation. • Direct access for residents to their home environment. • Concurrent, multi-resident support. • Interoperable programming interfaces for end user development of smart home applications. • Graphical user interfaces. • Small waiting times for simultaneous requests. • Acceptable performance in terms of response times and battery lifetime of devices. • Scalability.
  • 111. Conclusion University of Cyprus• Web technologies for interoperability.• IPv6 is a feasible technology for home automation.• Web techniques for advanced performance.• Web-enabled smart homes promote the development of real-world pervasive applications.• Request queues for improved performance, reliability.• Energy awareness through social incentives.• Smooth integration with the smart grid.• Beyond the smart homes.
  • 112. Open Issues University of Cyprus• Security• Privacy• Energy conservation• M2M• Semantics• A killer application.• Fairness
  • 113. Thanks for your attention!Contact Details: Andreas Kamilaris Email: camel9@gmail.com
  • 114. Acknowledgments University of Cyprus