Sensor Cloud Computing

Summary

This article delves into the integration of sensor networks with cloud computing, a concept termed as 'Sensor Cloud.' The discussion emphasizes the proliferation of sensors in modern life, such as temperature, pressure, and biological sensors, and theorizes the potential of cloud infrastructure to enhance their utility. Sensors gather vast amounts of data, and through cloud computing, these data can be effectively managed, shared, and used in real-time applications. The concept of virtual sensors within the sensor cloud emulates physical sensors, allowing for scalable, flexible, and real-time processing of data. The article highlights the need to overcome challenges related to scalability, interoperability, and resource limitations inherent in traditional sensor networks. By integrating with cloud computing, sensor networks can better handle large-scale applications and enable cross-disciplinary utility, emphasizing robust data management, real-time processing, and broader access to information.

Highlights

• Sensor Cloud combines physical and cyber worlds using sensors as interfaces 🌐.
• Proliferation of sensors in modern life necessitates advanced data management solutions 📈.
• Cloud computing enables real-time processing and sharing of valuable sensor data 🕒.
• Virtual sensors provide scalable and customizable solutions by emulating physical sensors 🛠️.
• Integration overcomes traditional sensor networks' challenges like scalability and resource management 🚀.

Key Takeaways

• Sensor Cloud blends physical sensors with cloud computing, enhancing data processing and accessibility 🌥️.
• Integration helps manage vast sensor data, supporting real-time applications and decision-making 📊.
• Overcomes traditional sensor networks' limitations—scalability, interoperability, and fixed applications 📈.
• Virtual sensors emulate physical sensors to provide flexible, scalable solutions 🌐.
• Sensor Cloud supports multi-user, cross-disciplinary applications by sharing resources effectively 🤝.

Overview

Sensor Cloud represents the fusion of ubiquitous sensor networks and powerful cloud computing, creating a dynamic interplay between the physical and digital realms. In contemporary times, sensors ranging from temperature to biological are omnipresent, collecting numerous data points about our environment and behaviors. By capitalizing on cloud technologies, these data streams can be processed and analyzed effectively, enabling real-time decision-making and expanded application across various fields.

The concept of Sensor Cloud is particularly valuable in overcoming the limitations of traditional sensor networks, such as issues with scalability, fixed applications, and interoperability. By virtualizing sensors, the Sensor Cloud infrastructure allows for flexible management of sensor data and system operations. It supports creating virtual sensor groups that can dynamically meet application needs, thereby facilitating robust and versatile usage of the vast amounts of data generated by sensors.

Ultimately, Sensor Cloud serves as a mediator, connecting user applications with deployed sensing technologies to facilitate the seamless management of sensor data. This integration promotes efficient data lifecycle management—from collection to decision support—enhancing the relevance and utility of sensor data across different sectors, from environmental monitoring to healthcare and beyond. The Sensor Cloud paradigm marks a significant step forward in harnessing sensor technology's potential, offering tailored solutions that span organizational and disciplinary boundaries.

Chapters

• 00:00 - 01:30: Introduction to Sensor Cloud Computing The chapter introduces the concept of sensor cloud computing as a pertinent topic in the current context. The discussion follows on from previous sessions on cloud computing, indicating that the subject of sensor cloud will be explored further in the chapter.
• 01:30 - 03:30: Overview of Sensors and Their Increasing Usage The chapter discusses the proliferation and variety of sensors available in the modern world. It starts with basic sensors like temperature and pressure sensors and extends to more complex ones like atmospheric sensors and even cameras. The scope emphasizes the enormous number and variety of sensors that surround us today.
• 03:30 - 04:30: Decision Support Systems and Actuators Chapter: Decision Support Systems and Actuators This chapter discusses the various biological sensors available that monitor different aspects of human physiology. It highlights the increasing prevalence and importance of these sensors in gathering information. These sensors collect data about the environment, health conditions, and specific devices, thus playing a critical role in accumulating valuable information for decision support systems.
• 04:30 - 07:00: Integration of Sensors with Cloud Infrastructure This chapter discusses the integration of sensors with cloud infrastructure. It explains how sensors and actuators are connected via a decision support system, either physical or logical, that makes intelligent decisions and activates actuators accordingly.
• 07:00 - 10:00: Challenges in Sensor Networks The chapter titled 'Challenges in Sensor Networks' discusses how modern systems, such as cars, incorporate a multitude of sensors, sometimes surpassing the number of mechanical parts. It highlights the growing proliferation of sensors used for monitoring and management purposes, illustrating the evolution and complexity in sensor network applications.
• 10:00 - 13:00: Cloud Computing and Real World Integration The chapter discusses the integration of cloud computing with real-world applications, particularly focusing on the incorporation of sensors. It examines the necessity of connecting sensors to cloud infrastructures, creating a 'cloud of sensors' that mimics cloud properties.
• 13:00 - 18:00: Combining Sensor Networks and Cloud Computing The chapter delves into the increasing synergy between sensor networks and cloud computing, highlighting the ubiquitous nature and extensive capabilities of cloud environments to emulate a 'sensor cloud.' This concept is introduced as more effective than standalone sensors. The discussion revisits core properties of cloud computing that contribute to its popularity, setting the stage for a more in-depth exploration of its potential scope and applications, likely in subsequent sections.
• 18:00 - 23:00: Sensor Cloud Architecture This chapter discusses the increasing adaptation of sensing technologies, emphasizing RFID, cameras, and mobile phones as contemporary sensors. It highlights the role of the internet in facilitating real-time data acquisition for various parameters, including meteorological data.
• 23:00 - 31:00: Virtual Sensors and Their Configurations The chapter discusses the impact of the internet and cloud computing on real-time information dissemination. It highlights how blogs and social networks play a crucial role in keeping people informed about events in real-time, making these technologies an integral part of daily life. The chapter also touches upon the emergence of cloud computing as a key solution to manage the big data revolution. It emphasizes the pervasiveness of these technologies and their significance in modern society.
• 31:00 - 34:00: Sensor Cloud Infrastructure and Management This chapter discusses the management of data through sensor cloud infrastructure. It explores the potential of using data from sensors, combined with cloud capabilities, to address the demand for resources based on the volume of data. The chapter investigates whether it is possible to amalgamate these technologies effectively to enhance data management.

Sensor Cloud Computing Transcription

• 00:00 - 00:30 hello so we will continue our ah discussion on cloud computing ah today we will ah discuss on a ah topic ah which is ah very much relevant in todays ah context that is sensor cloud
• 00:30 - 01:00 right so what we have ah seen ah or what we are seeing ah or having these days its a world of sensors right its a its a you know enormous ah number enormous variety of sensors around us right it may start from ah temperature sensor or sensing ah pressure sensor to atmospheric different atmospheric parameters to if you can look at that ah camera as a sensors there
• 01:00 - 01:30 are other different ah biological sensors which senses different aspect of our ah human physiology and other things right so there are different aspects of sensors and its a ever increasing phenomena right so we what we have we have a set of sensors which ah senses or accumulates information about environment about our health condition or about ah a particular device or so and
• 01:30 - 02:00 so forth and ah based on the informations so there is at the back end there is some intelligence or decision making systems which takes some call and activate or actuators are ah activated so we have a sensors and set of actuators in a in ah and in between there should be a a ah either physically or logically a decision support system or a intelligent
• 02:00 - 02:30 systems which takes somethings right like if you look at our modern cars which are a bunch of sensors some ah somebody ah was saying other day that ah if the time is coming when where a a a particular car may have more sensors than mechanical ah parts right it may be a joke but it says that the amount of ah proliferations of sensors for a monitoring management and
• 02:30 - 03:00 activating different different ah mechanisms and processes around us right so what we try to see that there is a amalgamation or there is a ah there is whether there is a need to ah integrate this sensors with ah cloud infrastructure right or having a cloud of sensors which which has a property of ah emulating the properties of cloud right like
• 03:00 - 03:30 ah what we have seen that the different properties of cloud which is making popular whether this this type of ubiquitous sensors or a huge volume of ah huge ah number of sensors can emulate a sensor cloud which is much more effective than individual sensors right or that we will try to look at again as we see that it is more of a overview which we will open up ah the scope for ah
• 03:30 - 04:00 for ah you to read more somebody interested in research activity etcetera to look into the things so ah what we see increasing adaptation of sensing technologies right that it is rfid camera mobile phones are ah nowadays sensors so location as we have seen that a different ah meteorological parameter are being sensed so internet has become a source of real time
• 04:00 - 04:30 information on the hands internet has become a real time through blogs social network for events happening around us right so to to technologies which are ah extremely pervasive and it is it it has become a ah both of them ah eventually become a part of our day to day life right on the other hand cloud computing has emerged an attractive solution for dealing with big data revolution
• 04:30 - 05:00 which can able to manage and able to take a call on this type of huge or weak volume of data so by by combining data obtained from sensors we that from the internet we can potentially create a demand for sources that can be appropriately made by cloud right so what we are trying ah what we are trying to look at that ah whether we can have ah this amalgamation of this ah
• 05:00 - 05:30 the technologies or the ah or the processes which are collecting data there is a way of dispersing or dissipating the data that is a accessing the data ah so both sensors internet cloud coming together to make ah um make a our or our need or to full fill full fill the different aspects of our of our either day to day or commercial or industrial requirements
• 05:30 - 06:00 of the thing right so so rather ah we have referred a one of the pioneer paper that what next it is a whether sensor plus cloud is the ah things which are coming in this particular decade so already we know about wireless sensor network ah i believe that you have ah more or less it is known just to put some points so its a seamless ah ly couples physical environment
• 06:00 - 06:30 with the digital world so its a what it is there there is a wireless ah sensor networks which which has a which has ah different sensor nodes or modes sometimes we say which collects information and ah those are digitized and pushed through the scene to the central ah central infrastructure so it is it is ah so what we do that whatever the physical environment is there ah is being captured and coupled with the our digital world so sensor nodes
• 06:30 - 07:00 are small low power low cost and provide multiple functionalities right so typically there are small they are power hungry cost wise ah also not so costly and as different type of functionalities like sensing capability sometimes it has a processing ah capability memory communication bandwidth battery memory power and these are the different type of components of the things so sensing ah sensing and processing at times or transmitting
• 07:00 - 07:30 or store and hold like memory and then push it so communication bandwidth and and it requires a management of the battery power in ah ah aggregate sensor nodes are substantially data acquisition processing capability usually ah when it use in a aggregation mode there are primarily ah using acquisition of the data and processing ah some sort of a processing
• 07:30 - 08:00 and push it to the ah next node or to other h node so useful in many applications and we have we know already that environmental health care education defense manufacturing smart phones and anything where where you require some sort of a sensing and monitoring of the things so what we see here there is a battery of sensor node these are been organized in a particular fashion but that can be ubiquitously thrown and can ah reorganize among themselves there are ah some ah gateway with which it
• 08:00 - 08:30 connects to the rest of the world right and this can be different type of things like it can be management of forest planes city transport and so and so forth so it finds application in different aspects like so it is a either a the equipment itself as a having a sensor thing like if i if i say a car or a airplane so those are different sensing units doing it or they are sensing and sending data like a temperature center
• 08:30 - 09:00 moisture sensor or hey sorry humidity sensor ah say velocity sensor they basically collect information send it out and with all different sensors collected information a call is taken by the management ah or the management or the control unit right so there are ah typical limitation in sensor networks that we already we understand that is a very challenging to scale sensory network to large sizes suddenly scaling up or scaling
• 09:00 - 09:30 down this sensor networks is a major challenge right so you whatever you have deployed in the physical sensor is there suddenly instead of ah collecting a say i want to increase i am collecting temperature or ah acquiring temperature and suddenly increase it to to a all larger scale and to on to collect a large thing its a its scaling up is different even changing the granularity is different right so proprietary vendor specific design in most of the cases difficult for different sensor
• 09:30 - 10:00 networks to be interconnected so there is another problem interoperability sensor they not data cannot be easily shared between different groups of users because the everything if this isolated then it is extremely difficult to think ah share first of all there is a problem of the proprietary formats secondly there is no there may not be there may not be mechanisms to how to share data so i have a bunch of sensors collecting and putting somewhere and another bunch of sensors and putting somewhere and those need to be talking
• 10:00 - 10:30 to each other right so that that need to be enforced right in sufficient computational storage resources in sensors to handle large scale applications right if i have a more complex application it is difficult to do on this use for fixed and specific application that cannot be easily changed once deployed so usually the sensor requirement are for fixed and specific applications that cannot be easily changed or deployed slow adaptation of large scale sensor network applications so if we have large applications
• 10:30 - 11:00 so the immediate deployment and adaptation becomes a challenge so these are some of the things which are we see ah so to say with this type of sensor network including wireless sensors networks right and similarly if we look at the there are a a in ah terms of the cloud other than all the limitations we have looked into there are other typical limitation like immense power of cloud can only be fully exploited
• 11:00 - 11:30 or fully appreciated so to say if it is seamlessly integrated with our physical lines right so it is it is not that when i demand i take things that it it takes a some sort of a real world situation and process it on a real world things then only that actual power of the cloud is manifested or appreciated by the ah community at large so what it means that providing a real worlds information to the cloud in real time and getting cloud to act and serve us instantly right that is one thing or another thing adding
• 11:30 - 12:00 the sensing capability to the cloud so cloud as such as the infrastructure wise ah just waiting and ah waiting for something to use it rather than it has a it has coupled with this sensing capability which sense do some processing and react to the things right so having this sort of sensing capability will give a definitely a extra age to these ah sort of a cloud computing so so to say um it is ah if not ah if we if we dont want to tell you the limitation of
• 12:00 - 12:30 cloud but to make a cloud really useful and ah effective um we need to do this right and so one side we have sensor networks ah typically ah we know that there is a ah bunch of sensors there is a way there is a sink node or gateway through which it is connected to the rest of the world and this there are different way of um looking at the ah this ah sensing
• 12:30 - 13:00 things like ah ah like they can follow a particular path to the ah to the say to the sink or they can ah form different clusters make the cluster hate and so and so forth there are technologies what you can see in the sensors network or on the other side if you look at the cloud infrastructure we have different sort of things like there are cloud server cloud computing
• 13:00 - 13:30 platform several cloud level applications mobile computing cloud security aspects economics and so and so forth right and different models and different level of manifestation are there so ah there is there is a missing link which we want to connect like making these sensor overall as a sensor cloud or connecting with the rest of the things which the cloud support which which exactly the sensor cloud ah infrastructure or sensor cloud network try to address like
• 13:30 - 14:00 so like ah if we look at a say a typical scenario like ah this sort of things where ah there are multiple parties multiple information are being exchanged these different type of resources are being ah different type of ah equipments specially mostly most cases our mobile phones are used to capture this informations right and if we look at the cell phone is
• 14:00 - 14:30 somewhat omnipresence and records the tourists gesture activities applications such as in camera microphone etcetera which could have been captured cell phone produces very swift responses to real time ah after the processing geographical data to show that if it is interfaced with the geo cloud type of things acquiring tourist physiological data and wearable physiological ah sensors ah so physiological sensors like blood sugar etcetera and cross comparing with
• 14:30 - 15:00 the medical records like so if if somebody is need some medical help this automatically instantiated or before the some something serious happens there are speech recognition system to find out things image processing of the some restau[rant]- logo or accessing internet based things accessing tourist social network profiles to find out his friends etcetera so what we see that even though is a simple scenario of visiting a particular place but
• 15:00 - 15:30 there are several type of sensors at different levels which are there some are the physically doing that some are may be doing some derived informations right ah so its a its according like ah whatever been posted in the social network finding the friend etcetera it is also trying to do some data information and try to couple with this physical information its location ah if if as at all we are trying to do monitor its a other type of infrastructure
• 15:30 - 16:00 those type of things are ah being ah used to that type of things so the fact the cell phone cannot perform all the task which is needed right that we see that the cell phone cannot perform so we need require some of the something which is ah which is required for ah which is something more then this or coupling sensors with cloud ah philosophy so there is a need to integrate ah sensors with ah cloud right ah the acquisition
• 16:00 - 16:30 of data feeds from numerous body area networks right like a different type of body parameters and wide area ah like water quality monitoring that is a small or large scale things what we need to do ah that sensor networks in a real time type of things so what we not need to do that it is not only real time informations but also real time
• 16:30 - 17:00 processing of the informations is the is a requirement which is coming up where sensors integrating as sensors with cloud may may give some results real time processing of heterogeneous data sources in order to make critical decision right so we need to do some real time processing of heterogeneous data sources coming from different type of sensors like ah something coming from the body area network or sensors ah which are ah taking
• 17:00 - 17:30 a more ah health related data accompanied by the other type of things like may be the overall environmental things along with ah some of the sensors which gives that ah like related to the heights and so and so forth and try to take a call integrating those reason these are the different heterogeneous ah sensor informations which need to integrated so automatic formation of work flows and invocation of services in the cloud one after another to
• 17:30 - 18:00 carry out complex task tasks highly swift data processing using immense processing power of the cloud to provide quick response so one side we have a capability of wide variety of sensors collecting information another side what we have that computing power and the ability to interoperate between different type of heterogeneous data ah from the cloud we want to integrate together so it if comes to a scenario of what we say sensor cloud so what if you try to look at
• 18:00 - 18:30 a the sensor cloud does an infrastructure that allows truly pervasive computing using sensors as interface between physical and cyber worlds the data computing clusters as a cyber backbone and internet are the communication media right so what a let me repeat its a infrastructure that allowed pervasive computing using sensors as an interface to the physical and cyber world right and data computing clusters at the cyber backbone at the internet as the
• 18:30 - 19:00 communication media so it is a amalgamation of this different technologies to realize a or to ah basically support a real time application or real time processing of informations right so it integrates largescale sensor network with sensing applications and cloud computing infrastructure it collects and process data from various sensor networks enables large
• 19:00 - 19:30 scale data sharing and collaboration among users and applications of cloud delivers cloud services by sensor rich devices allows cross disciplinary application thats span organizational boundaries like it says lot of things right so its a its says that what we are discussing that to make it the whole thing ubiquitous and omnipresent and to serve a variety of applications so just to reinstate that enables user to easily collect access process visualize archive
• 19:30 - 20:00 share and search large amount of sensor data from applications supports complete sensor data lifecycle from data collection to the decision support system as we are discussing vast amount of sensor data can be processed analyzed and stored using computational and the storage resource of the cloud right so its as a cloud as by virtue of it that having
• 20:00 - 20:30 a huge infinite storage theoretically i can store this huge volume of data for further processing ah so many sensor are sending its a really a data challenge or a big data problem allows sharing of resources by different users and applications ah under flexible usage scenario enable sensor devices to handle specialized processing task so this these are sensing processing actuating type of services so sensor cloud enables different network spread in
• 20:30 - 21:00 huge geographical area to connect to together be employed simultaneously by multiple users on demand right so so if you can see that from the sensing technology we are in the we are trying to incorporate or what we trying to do in the sensor cloud is putting that that good properties or good use of this cloud in to this sensing overall sensor network infrastructure
• 21:00 - 21:30 so you can have a something overall overview of the things one side is sensor network ah where it ah there is a sensor network a proxy which proxies it for the rest of the network this handle devices or the handle ah sensors like mobile phones another type of things there is a traditional cloud we have a sensor cloud proxy which handles it and there are satellite connection standalone gps sensors there are network of sensors which capabilities different sort of things are being connected over the internetwork so sensor cloud proxy
• 21:30 - 22:00 it primarily interface between the sensor resources and the cloud fabric manages resource network connectivity between the sensor and resources which proxies on behalf of this this say your sensors resources which can be heterogeneous which can be which homogenous type of things so it handles data format conversion like using technologies like xml interoperability data cleaning and aggregating data transfer to the cloud storage and so and so forth
• 22:00 - 22:30 sensor network proxy for sensor resources that do not have different connection to the cloud this acts as a proxy so it is is unlike it is not ah it cannot be expected that all sensor network will have that connection to the could and able to interface to the cloud so thats why we have a proxy where by which it is connected to the cloud right there are different other use cases like one is the traffic ah we can say that traffic flow sensors
• 22:30 - 23:00 that deploy large number in ah place city places like large cities right these sensors are mounted on traffic lights and provide real time traffic flow data right so these are ah on on traffic lights systems on signaling system they can be mounted and real time things driver can use these data for his route planning right you see that ah there are this this amalgamation of information and taking a call for somebody from the sensor
• 23:00 - 23:30 things is there in addition if the traffic flow sensors are augmented with low cost humidity and temperature sensor that can also provide customized local view of the temperature and heat index on demand right so it is also support the metrological department which other ways have fix number of sensors right so i can have a rough ideas i i have that precision sensor of the med department which keeps more precision and i have this local ah sensors which are on the lamp post and traffic ah signal and lights which keeps me a much more ah real time ah scenario of this
• 23:30 - 24:00 temperature so the same type of infrastructure with little augmentation we can have different layer of informations or in other sense you can see that you have virtualized these things into different type of things right i have a one side that metrological departments looking for the med type of things traffic management department looking for that what is the overall traffic managing things i can have ah ah drivers on the road which uses for road planning and
• 24:00 - 24:30 so and so forth right so we can have ah one over view of the sensor cloud infrastructure ah like ah there are the standard ah way of looking at it data processing routing power management clock synchronization localized fault detection this is which are their basic physical sensors which has to be there and i basically what we try to do is a having a virtualized scenario
• 24:30 - 25:00 or the physical layer over that as we have seen in the cloud a virtualized scenario which helps me in looking at different type of user related issues right so i have now a set ah virtual sensors or virtual sensor groups which which can answer to the different category of applications like as we are discussing one may be med type of one may be traffic type of applications one may be action management type of things
• 25:00 - 25:30 and ah something may be ah overall ah looking at that ah other phenomenon like ah how is that ah lightning ah or how is that overall ah environmental situations of that particular place so we talk about virtual sensors what here is as we have said it is a emulation of the physical sensor that obtains it data from the underlying physical sensor as we have
• 25:30 - 26:00 seen virtual machine etcetera so it is emulating the machines from the ah underlining the physical machines right so virtual sensors provide customized view to the users using distribution and location transparencies right in a sensor ah wireless senor these hardware is barely able to run multiple task at time and difficult to multiple vms and as in a traditional cloud to overcome this problem virtual sensors acts as an image in the software of the (Refer
• 26:00 - 26:30 Time: 26 :00) corresponding physical sensors right so thats as we have seen that ah in case of our cloudy cloud things also so virtual sensors contain metadata about the physical sensors and users currently holding the ah physical sensors so we have a virtual sensors layer so there are physical sensors over that we emulate this virtual sensors and there are different virtual sensor group which uses that it may so on happen one physical sensors may be contributing to the physical
• 26:30 - 27:00 sensors so one to many many to one many to many and this type of things ah scenario may come up so as we are ah discussing so there are sensor cloud admin so this is a sensor cloud infrastructure which has different physical sensors have virtual sensor groups which have underlining physical a virtual sensors which are coming from the physical sensors and then we have end users which are more more ah looking for
• 27:00 - 27:30 this ah more inclined to the applications or more bothered about the applications which talk to this uses this ah sensing network to look at it so there is a management of this ah sensor cloud ah like ah registering physical sensors management ah mange infrastructure and that it may happen the sensors because this sensors are low powered manage by different ah type of ah in different situations so sensors may come up or can go down and type of things so this sensor cloud need to be managed appropriately
• 27:30 - 28:00 and ah the something if we look at the client the portal it goes for a provisioning resource management monitoring virtual sensor group physical sensor groups and so and so forth this ah overall architecture if we look at ah in other way its a its ah so client request ah to they to a particular portal servers which intern ah go for that ah process the
• 28:00 - 28:30 query ah it needs to look at the different virtual sensors which can answer to the queries and ah which intern look for the physical sensors and so and so forth so as we have discuss ah discussed there are virtual sensor configuration it can be one to many many to one many to many and can be derived right so ah there are several scenario many to many ah one to many then we have this derive type of ah scenarios so in case of
• 28:30 - 29:00 a many to one to many ah in this configuration one physical sensor corresponding to many virtual sensors although individual sensors on their virtual images the underlining physical sensor is shared among all the virtual sensor accessing it right so its the middle ware which computes the physical sensor sampling duration and frequency by taking into account all the users it reevaluates and duration so it is basically the synchronization
• 29:00 - 29:30 is ensure by this ah underlining middleware and type of things so it is a so where we have one physical sensors corresponding to ah more than one virtual sensors many to one in this configuration geographical area is divided into regions each region have one or more physical sensor and the sensory network when a user requires a aggregate data of specific phenomena from a region all underlining wsns or value sensory networks which switch on
• 29:30 - 30:00 with the respective phenomena enabled and the user can access the aggregated data so here what we have have a large geographical space why which is different into regions and which has say for wireless sensor network and if there is a aggregated ah some processing is required then these are switched on and give a aggregated view on the things right so its a scenario of many to one we can have as a many to combination of one to and many to one type of scenarios there is a special scenario which is a ah derived so its a derived
• 30:00 - 30:30 configuration refers to a versatile configuration of virtual sensors derived from a combination of multiple physical sensors right so there are virtual sensors which is a combination of the ah derive sensors like here you say there is a proximity sensor light sensors there are say habited monitor temperatures humidity sensor and these are integrated into to take a call right
• 30:30 - 31:00 so this configuration can be seen as a generation of other three configuration generalization basically generalization of the other three configuration lies in the type of physical sensor with a physical sensor communicate so it is it it it emulates the sensors looking at different other physical sensors so it derived a sensing capability from the other physical ah sensors right so this type of scenarios are being emulated so there are there can be many different kind of physical sensor that can helps a complex question for
• 31:00 - 31:30 a for example are overall environmental condition safe in a wild life habitat so this is a very complex question right there there is a there should be a definition of what is meant by the overall condition what is the meant by the safety of a wild habitat and this need to be known so virtual sensor can use reading of number of environmental condition from physical sensor to compute safety level values and answer the query type of things right so there is ah quite ah um ah pervasive applications which can use this type of scenarios and if we look
• 31:30 - 32:00 at a layered sensor cloud architecture so we have different wireless sensor providers right they provide data information and type of things we have a sensor centric ah middleware which is a wsn maintenance wsn registration type of things and then we have a other middleware where we say as a virtual sensors emulating virtual sensor provision management image
• 32:00 - 32:30 life cycle and so an so forth and then at the top interfaces so these are multiple layer one is that physical ah underlining layer over there ah one to have this ah wsn registration wsn maintenance data collection part over there that virtualization layer and then we have that physical ah user interface to connect to the things and they are we can have other type of things like ah user repository data repository session
• 32:30 - 33:00 management membership management and so and so forth so there are variety of things which are can be there so if we can if we summarize sensor cloud infrastructure virtualizes sensors and provides management mechanism for virtualized sensors sensor cloud infrastructure enables end user to create virtual sensor group to dynamically by selecting the template of the virtual sensors ah or the sensor groups which are there so the based on the user application i can select a particular template to work on that sensor
• 33:00 - 33:30 cloud infrastructure focuses on sensor system management and sensor data management so this is more inclined or the vertical towards sensor data management and sensor ah system management and aims at ah to take the burden deploying managing network away from the user acting as a mediator between the user and the sensor network so that is a its its a sort of a connecting user applications with the deployed ah data or the sensing environment
• 33:30 - 34:00 so as as as we have started todays ah discussion that that is that is a there is a omnipresent ah presence of different kind kind of sensors starting from our mobile phones to different category of sensors which is collecting huge volume of data which could have been used for processing and supporting different user applications so that a a virtualization mechanism is supported and evolving a sensor cloud which are ah which are which may help in ah addressing
• 34:00 - 34:30 different real time ah applications or serving different real time applications with a variety of heterogeneous ah sensor collected or sensor acquired data system thank you