Week 1: Internet of Things

Week 1: Internet of Things

“Introduction … Wireless Communications … Cellular Mobile Systems … Wireless Communications … Comparison of Wireless Systems … Wireless Standard: WiFi and Bluetooth … Wireless Standard: BLE and Zigbee
(Source URL)

Summaries

  • Week 1: Internet of Things > 1a Internet of Things: Introduction 1 > 1a Video
  • Week 1: Internet of Things > 1b Internet of Things: Introduction 2 > 1b Video
  • Week 1: Internet of Things > 1c Wireless Communications 1 > 1c Video
  • Week 1: Internet of Things > 1d Wireless Communications 2: Cellular Mobile Systems > 1d Video
  • Week 1: Internet of Things > 1e Wireless Communications 3: Comparison of Wireless Systems > 1e Video
  • Week 1: Internet of Things > 1f Wireless Standard: WiFi and Bluetooth 1 > 1f Video
  • Week 1: Internet of Things > 1g Wireless Standard: WiFi and Bluetooth 2 > 1g Video
  • Week 1: Internet of Things > 1h Wireless Standard: BLE and Zigbee 1 > 1h Video
  • Week 1: Internet of Things > 1i Wireless Standard: BLE and Zigbee 2 > 1i Video

Week 1: Internet of Things > 1a Internet of Things: Introduction 1 > 1a Video

  • Welcome to the sequence of lectures on the topic of Internet of Things.
  • The lectures om Internet of Things are going to be given by a number of faculty from Columbia University.
  • This lecture is a introductory one giving an overview of what Internet of Things is, as well as some insight into what are the topics that we’re going to cover in our lectures.
  • So a key piece of the concept of Internet of Things is a Thing that does something useful to make Internet of Things happen.
  • So what happens with that chip, that chip gets attached to the thing enabling this thing to become a piece of a bigger thing.
  • Typically the data that is collected off of the thing is sent initially to a gateway.
  • In gateway we collect the data that comes from the thing, really that comes from a number of things.
  • A Cloud in a computing Cloud, we can process massive amounts of data, we can clean the data, we can analyze the data, we can deduce some things from the data, and finally in sophisticated applications, we can provide feedback, in our case to this particular thing.
  • So a feedback will help the thing either decide something or do a particular function.
  • So this is a reasonable, simplified explanation of what Internet of Things consists of.
  • If we nail down two most critical pieces of the story, well, we have a thing that does something useful.
  • More than a thing, what we care about is the data.
  • So the world is going in the direction of having many things.
  • By the way, many of these things being enabled by the capabilities of Internet of Things features.
  • So one aspect of Internet of Things that really deals with massive networking connection of all of these devices in a fashion where the connectivity can be maintained, can be reliable, can be secure, and that it doesn’t overwhelm computing resources, or bandwidth in the radio frequency domain, or any of the equipment that we need to support such a network.
  • Indeed if we didn’t have networking or if the devices were not interconnected, the value of the Internet of Things would not be at all obvious.
  • Now this makes a lot of sense but at a time when Internet of Things was not an overwhelmingly dominating technology there were many other alternative technologies and typically they would use much more of a centralized control.
  • The next thing that internet follows is a design principle, is the so-called best effort service model.
  • One should realize however, that in reality there are many different tools, processes, procedures mechanisms which are used today in concert and on top of internet to facilitate the reliable data transmission, high quality of service data transmission, and so on.
  • It is desirable to preserve the same principles for purposes of the world of Internet of Things.
  • So what is Internet of Things? Another way of looking at it, well we have things and we want these things except for being physical devices, we want them to be able to sense- provide some useful information about some physical phenomena.
  • Some of the things since their feature is anyway to do something useful, they can act.
  • Acting may actually be influenced by all other processes that occur within the Internet of Things concept.
  • There are many different ways of calling the Internet of Things network.
  • Some industry players or players in this field call the network a web of things, some of them call it an internet of everything, some of them call it a Cloud or Fog network.
  • It is worth thinking through what is it that made Internet of Things possible.
  • How is it different from disciplines that we have been aware of and deeply pursuing in the last 30 or 40 years? So many of the concepts that are being covered under the name of Internet of Things existed before- embedded computing, scatter control.
  • IC technology was a crucial piece of the technology contributing to the emergence of what we think of today as Internet of Things.
  • Wireless allows us, and most often the wireless means radio although it’s not the only wireless technology- wireless enables us to reach things that are not reachable by any other means.

Week 1: Internet of Things > 1b Internet of Things: Introduction 2 > 1b Video

  • So it’s worth saying since internet has sort of two words, Internet of Things, let’s say a couple of more words about things.
  • So the thing is to make them useful, more technical stuff that needs to happen is that you must identify, you must be able to identify every single thing.
  • We have great motivation to make sure that communication that happens between things or between things and networks is secure, fully authenticated, and fully authorized.
  • Since we are anticipating billions of devices who play in the Internet of Things, it is crucial that we come up with a limited set of standards, APIs, such that the devices that we deploy can actually work together.
  • We listen to it and we say yeah, that’s true, but if somebody were able to invent a battery or some other energy source that could be five times better in whatever fashion, certainly in capacity than what we have today, the contribution to many disciplines but certainly to Internet of Things would be much greater than incremental improvements in protocols and methodologies that we keep pursuing, with good reason, though.
  • In the wording of ITU, internet of things is dynamic, it’s global network infrastructure that can self-configure using standards and using interoperable protocols where physical and virtual things have identities, attributes, and personalities, use intelligent interfaces, and can seamlessly integrate into the network.
  • So one word that catches our eyes may be “virtual” because the things that we were mentioning so far certainly look like always a physical thing.
  • Anything can be defined as a thing as long as there is some data to worry about and there are ways of communicating with it and exchanging information.
  • Internet of Things has a very large number of different classifications, so we’ve chosen to show one that is very often shared among people who participate in this area.
  • As a matter of fact, the concept of Internet of Things in itself may embody all other internet [INAUDIBLE].
  • Taking a look at what are the service sectors that Internet of Things can service, where we start with information technology and network security, retail, transport, industrial, health care, consumer and home energy, and buildings.
  • Many of us have been exposed to some notion of Internet of Things mostly through retail devices, such as activity monitoring bracelets or similar.
  • It is certainly true that these retail devices are a piece of Internet of Things and that they bring important data, but it is more and more obvious, and there is a large focus on the so-called industrial internet of things because it is believed that the monetary value of deploying technologies associated to the Internet of Things in the industrial space brings large monetary benefit.
  • We can say in the example of energy, if you deploy Internet of Things, the first thing you can think of, some sort of measurement devices that are attached either to turbine blades or to batteries or to generators, devices that can assess vibrations, the amount of tension, the speed of rotation.
  • Well, if we constantly monitor all of those things in a simple fashion, we can regulate things such as energy supply and demand, we can think about incorporating in an easier fashion many different alternative energy sources, and we can much easier and more effectively manage oil and gas fields by planes and distribution.
  • Internet of Things can play a big role in that area.
  • Many do ask, is Internet of Things here? Or when is it coming? Or was it always around? Or is it going to come ever? So these are interesting questions and valid questions.
  • This shape of a diagram is used across all sorts of different problems, and in this case it was used to represent, as of July of 2014, what may be the status of any individual technology, service, concept, that somehow is associated with Internet of Things.
  • Or it might be that we at some moment think that it is the greatest thing in the world and very useful, but still not embraced by everybody or deployed in the field.
  • What are these dimensions that we ask the question against? Does it have to be very reliable? Is the latency or time delay required to be super low? Is the thing that we have to deal in that application mobile? Does it move? How secure does it have to be? Does it have to be a really, really low-power or low-power but not really, really low-power? And so on.
  • Some of them have very high needs, so unfortunately, that means that the way we have to design the overall IOT, Internet of Things, has to be able to service and facilitate services for very variable set of requirements.
  • If we think about gaming, video gaming, then things change dramatically.
  • Now, Internet of Things is also dabbling in the area of machine control.
  • So Internet of Things has to be able to cover all of these cases.
  • So there are many words we use today for describing what is the world going to look like a number of years from now, and we hear about ubiquitous computing or pervasive computing, ambient intelligence, things that will come in the future.
  • We talk about the Internet of Things, we have been talking for a long time about SCADA, which is an industrial standard for supervisory control and data acquisition.
  • One must say that in many ways, the concept of Internet of Things is just an embellishment of a SCADA concept at the level of machine-to-machine communication and control.
  • So this set of lectures on Internet of Things is covering the following concepts- first we talk about basic design principles, we spend time talking about wireless communications per se as well as about wireless standards, which are key to Internet of Things.
  • On one hand, we wanted to cover the basic issues, the primary technical issues, that surround all aspects of Internet of Things.
  • On the other hand, we wanted to showcase some of the advanced work that happens in particular areas in the domain of Internet of Things where Columbia University has advanced work.

Week 1: Internet of Things > 1c Wireless Communications 1 > 1c Video

  • Today’s lesson is going to talk about wireless communication systems.
  • So wireless systems can be classified in various ways.
  • One reasonable way to distinguished them is to classify them into licensed and unlicensed systems, which means that they either use licensed frequencies or unlicensed frequencies.
  • Cellular communications, paging, fixed wireless, and satellite belong to systems which are licensed.
  • So the service operators, like AT&T and Verizon, actually have to license the band for use and pass that right to their users.
  • Unlicensed bands- systems that work in them are cordless, wireless LAN, Bluetooth, machine-to-machine communication, and Personal Area Networks.
  • Typically, systems that work in licensed bands require building out an infrastructure.
  • In unlicensed bands, we usually deploy systems which do not require infrastructure.
  • They’re using the so-called peer-to-peer and ad-hoc types of communication, which implies that they’re self-organizing.
  • Wireless communications are peculiar in the fact that signals which are transmitted over wireless channels have extremely high variability.
  • So just for an example, we’ll pick up voice and make a statement how the data rates required to transmit voice are relatively low, that bit error rate requirements are medium, and the latency requirements are rather aggressive.
  • How does one build a system that operates over a wireless channel such that we can meet particular quality of service targets? So these requirements map into capacity, which means how many bits per second can be transmitted with some notable reliability over the system.
  • What percentage of geographical locations will get service with some acceptable level of service? And finally, the cost of service is an important parameter, as well.
  • On top of these requirements we have a peculiarity that many of the wireless systems rely on portable, mobile equipment that works on batteries.
  • Wireless communications imply that you are using frequency which may or may not be regulated by the government.
  • So in the United States, the Federal Communication Commission dictates how is spectrum used for different applications.
  • In the case of server communications, the auctions yielded large income for the United States government, because some service operators are given the exclusive use of that band.
  • The intent of the government is to make sure that some bands are used in a very spectral-efficient fashion.
  • When we have an unlicensed band, it means that any piece of device, any equipment can use that band as long as they don’t violate the spectral densities.
  • It is reasonable to expect that everybody will try to use those bands.
  • So you might have an example where TV broadcasters are using some bands and the government might allow that there is a secondary user can actually transmit in the same band, but then it has some limitations- again, most often in terms of power.
  • So unlicensed users can actually have restricted access to operating even in licensed bands.
  • In the United States, the Federal Communications Commission regulated the spectrum.
  • Then we continue with the next band on this side.
  • Just for a brief orientation, I will say that for cellular communications, mobile communications, a seemingly tiny amount of spectrum is used somewhere along the area in this domain.
  • We have zoomed in in an area of a frequency band which is used for cellular communications, at least some part of cellular communications.
  • There is also a piece of unlicensed spectrum, ISM- Industrial, Scientific, and Medical applications- exactly in this area here.
  • The big picture is such that a fairly small amount of spectrum is dedicated to a massive amount of communication that we today utilize for cellular communications.
  • For applications in the Internet of Things, some bands are of particular interest.
  • So very similar to what we have in case of WiFi and Bluetooth, IoT can take advantage of the band between 900 and 928 megahertz, which is an unlicensed band- the ISM band we just mentioned a few moments ago- or in the areas of 2.4 gigahertz and 5.7 gigahertz.
  • There are some bands at lower frequencies that are also appealing.
  • In these bands, we are essentially counting on the license-free operation so that a very large number of Internet of Things devices can be deployed without having to pay for the usage of the band.
  • Radio channel propagation determines virtually all capabilities of all wireless systems that can be built on top of them.
  • This figure illustrates the relationship between the received power and transmit power as a function of distance between a transmitter and a receiver.
  • So what happens is that because of the obstructions that exist between a transmitter and a receiver, there is variability around this blue curve.

Week 1: Internet of Things > 1d Wireless Communications 2: Cellular Mobile Systems > 1d Video

  • A crucial invention in the area of radio communications was the invention of cellular concept, which essentially teaches us how can we reuse frequencies? So before we say how that works, we should realize that from the previous graph we saw that the received signal strength reduces with the distance.
  • We are transmitting with some power in the center- imagine this is some geography and we are right in the center of some hexagonal area.
  • Well, if we are far enough from the center of this area, the signal strength is going to be reduced, and it can be notably reduced.
  • If you look at this blue area and this blue area, we’re making a statement where we can build a cellular system in which we use frequency one here in the geographical area, and then we reuse it once again here in another geographical area.
  • So neighboring base stations, which would mean if this is one base station in the center of this area and they have another base station in the center of this area, then these two neighboring base stations have to use different frequencies.
  • We drew here hexagonal areas, and that’s just an ideal form.
  • Again, a little illustration- we can reuse frequency one, we can also reuse frequency two, f2, we can reuse frequency f4.
  • So cellular concept implies that we have geographical regions which are divided into cells, and each cell is served by one base station.
  • So if you looked at the diagram before and if you looked at the cells that are colored with the same color, let’s take a look at that.
  • We say that there still may be some interference and that type of interference we call co-channel interference, and it has to be very low.
  • Base stations which sit in the center of those cells are connected by some wire line network rather than wireless network.
  • If we reduce the power of transmission, we can more frequently repeat the usage of frequencies across different cells.
  • So in cellular communications, typically a cellular service operator would buy or license the spectrum, this whole spectrum, and then it would divide it into smaller chunks, into smaller pieces.
  • So there could be as many as n users using each of the little chunks of frequency and never creating interference to each other.
  • So if it only needs a piece of time to transmit all of its data, it means that another user can use channel two and transmit its own data.
  • So we have cells in center of which we have base stations.
  • Another key concept in case of cellular communication systems relates to the fact that every user and its phone number is associated with a home mobile switching center.
  • If our user is in the area of a home mobile switching center, it is obvious how it will be able to receive the call.
  • If our user moves to some remote location, and if there is an outside phone call coming to that user, well, that outside phone call is still going to go to a home mobile switching center.
  • There is yet an evolution of those, LTE advanced, but prior to LTE, we had the third generation systems based on some different technologies using the so-called wideband CDMA modulation and particular type of packet switching.

Week 1: Internet of Things > 1e Wireless Communications 3: Comparison of Wireless Systems > 1e Video

  • We have second generation systems, and we can see the theoretical data rates are at the order of 200 to 300 kilobits per second for the download, which means from the base station to the user terminal.
  • Third generation cellular systems, which started being deployed in the middle of 1990s, enable communication on the download side in the order of several megabits per second for download, and close to 1 megabits per second, or around 1 megabit per second, for the upload from the user to the base station.
  • Today in North America, we have deployed fourth generation cellular systems, the so-called LTE, and LTE, LTE, and LTE-advanced in the deployment.
  • For this system, recounting at about 100 megabits per second transmission on the download and about half of that much on the upload. So from each generation we had- from generation to generation, we had almost an order of magnitude improvement in the data transmission rates.
  • Some more exotic technologies, UWB and 60 gigahertz millimeter wave, claim to be able to transmit at much higher data rates.
  • This is- from here we have second generation, third generation, and fourth generation cellular systems compared against wireless LAN and Bluetooth.
  • So for all of the cellular systems, the range is notably larger, in the order of some number of- several kilometers, whereas wireless LAN may be reaching at about the maximum of 100 meters, and Bluetooth and Bluetooth Low Energy would be reaching the distances of several meters.
  • Technologies such as ultra-wideband and 60 gigahertz millimeter wave would have the ranges which are similar to Bluetooth and Bluetooth low energy.
  • Thereby, wireless communications are error-prone, they’re unreliable, and they have latency issues.
  • What I can safely claim, that as of 2015, the basic fundamental issues around sending wireless- sending data over wireless channels have been solved, and that these issues of error-prone, unreliable, and latency issues, the solutions have been found for them.
  • The technology at this point is all about making incremental changes and dealing with distances trying to improve the data rates over shorter and shorter distances.

Week 1: Internet of Things > 1f Wireless Standard: WiFi and Bluetooth 1 > 1f Video

  • It evolved within the auspices of a standardization body, IEEE, and 802.11 is a set of medium access protocol and physical layer specifications for implementing the wireless local area network communication.
  • In the 2.4 gigahertz band there are altogether 14 channels, and each of these channels is spaced 5 megahertz apart.
  • So typically you would deploy maybe channel one, channel six, and channel 11 in the same room or location.
  • When the evolution of the standard went beyond the initial version- so this is the initial version- the overlapping of the channel was defined and proposed in the following fashion.
  • As opposed to cellular communication systems which have a very rigid type of allocation of channels, wireless LAN, WiFi, is based on so-called random media access control protocols.
  • The channel is essentially- the channel access is regulated through the so-called carrier sense mechanism, which is rather robust, and it allows for scalability of the network without having to have a centralized controller that negotiates between different users.
  • This figure illustrates the basics of CSMA-CA, which stands for carrier sense multiple access with collision avoidance.
  • So although we said that this methodology essentially uses random access, this collision avoidance portion gives us the ability to minimize the amount of collisions that occur over the channel.
  • Now, if it turns out that the packet didn’t get successfully transmitted, then the same packet will have the opportunity to go over the channel again, and the time at which the second transmission would start or the following transmissions may start is determined by the notion of contention windows and [? back ?] [? offs.
  • So as we said for carrier sense multiple access, the more contenders- the more users attempt to transmit, the worse off the performance is.
  • So access point is a somewhat special device, in that the stations that communicate with it- want to communicate between the cells- are essentially transmitted through an access point.
  • Access point is the component of a wireless LAN or WiFi that is connected to the network.
  • So if you have a room, or a building, then one or more access points can extend the coverage from several rooms to potentially several miles, square miles, depending on how close they are to each other.
  • Just like in cellular communications, to have reliable coverage one often requires, or one does require, that access points be located in a fashion where there is an overlap between them.
  • On the other hand, the other configuration is the so-called- or type- is the infrastructure BSS. Infrastructure BSS is the one where we have access points, and thereby all the communication goes from a laptop or from a computer to an access point, even if two laptops might be in the same physical location.
  • How does one establish the WiFi contact? So when a device with the WiFi connectivity is powered up, the media access control layer of that device starts establishing the contact.
  • In active scanning mode a device, let’s say a laptop, tries to locate an access point by explicitly transmitting a probe request frame.
  • It waits for a probe response from the access point.
  • So the access point periodically broadcasts a special frame, a so-called beacon frame, and that happens at regular intervals.
  • Through this beacon frame the access point advertises its capabilities, and this information is used by clients who are in the passive scanning mode to make decisions to connect to the access point.
  • Synchronization is a necessary step to make sure that all the clients are synchronized with the access point.
  • So every single station needs to be authenticate by the access point in order to be able to join its network.
  • If there is an open network, then a device can send and authentication request and the access point can send the result back.
  • The authentication protocol involves three parties- the be access point, the device, lets say a laptop, and the authentication server.
  • It sends a request- it actually sends a request frame to the access point, which replies to the client with an association response frame, which is either going to allow or disallow the association.
  • If the association is successful, the access point issues an ID to the client.
  • At this point, we’re now ready to exchange data between the client and the access point.
  • So if your device sends a data frame to the access point, the access point has to send an acknowledgement.
  • If an access point sends a data frame to the device, the device must send an acknowledgement.
  • Access point takes frames received from the client and it forwards them to the required destination, wherever it is, on the wired or some other wireless network.
  • In the opposite direction, if there is data or packets coming from the outside towards the client, the access point will be directing those packets to the client.
  • Access point can also between two clients which are local to itself.

Week 1: Internet of Things > 1g Wireless Standard: WiFi and Bluetooth 2 > 1g Video

  • Another standard that we’ve all heard about is the so-called Bluetooth standard.
  • So as opposed to wireless LAN standard- which is a local area network- a Bluetooth standard is built for purposes of personal area networks, WPANs, wireless personally area networks.
  • A traditional Bluetooth wireless network can contain up to seven active devices.
  • So the finding available to devices happens on an ad hoc basis within small networks.
  • Types of services that are supported by Bluetooth is very wide.
  • So examples are things like device control, mice and keyboards, voice transmission for headsets, video for webcams.
  • All of these different services can actually share the same ad hoc connection established by the Bluetooth.
  • Bluetooth has defined several different radio frequency power classes, 1, 2 and 3, limited with different amounts of power- 100 milliwatt, 2.5 milliwatt and 1 milliwatt.
  • The range of Bluetooth depends on which class the device belongs to.
  • Essentially, if you’re sending very large amounts of power, you want to be able to back off when you have a large number of Bluetooth devices next to each other.
  • As we said, seven Bluetooth devices can become a part of a temporary small network.
  • Then a Bluetooth operates at any one instant only over one channel, but it keeps changing which channel it operates in a rather rapid fashion.
  • Frequency hopping is also making it possible for several Bluetooth piconera to operate in the same area at the same time with minimal interference.
  • So the master unit in a Bluetooth piconet is the one that dictates how the communication is going to occur.
  • All of the other devices have to follow what the master and its [INAUDIBLE].
  • More recent versions of Bluetooth have data rates which go to about 2 or 3 megabits per second.
  • There are two types of data transfers in Bluetooth, the so-called symmetric data transfer and an asymmetric data transfer.
  • So the total frequency band used for Bluetooth in the area of 2.9 gigahertz is broken into 79 smaller frequency chunks.
  • So in a Bluetooth system, every device has the opportunity to discover other nearby devices using Bluetooth and create a connection between them where the information is to be exchanged, which validates the identity of each device and another piece of control which adds channel paths between these services or applications.
  • The terminology used is device discovery for purposes of establishing this initial communication.
  • Again, it’s used for finding other devices and figuring out what are their capabilities.
  • So if a particular device wants to learn about services that another device can offer, it must use the address and establish a temporary connection.
  • Each of these devices has an address, number three, number one.
  • If there is a device that is not in the address space, there will be no response.
  • If there is a device which is in the legitimate address space, it can say well I can cook, send the reply.
  • The next operation is the so-called Bluetooth pairing which validates the identity of other devices.
  • Bluetooth has a bunch of layers just as other protocols.
  • Distinctive feature of Bluetooth protocol is it has a very large number of profiles.
  • Essentially, these are particular versions of Bluetooth, different implementations, processes, definitions which in some detail describe what are the required operations and protocols, such that specific service or application can be serviced.
  • So why so many profiles? In the implementation of any particular device, the vendor manufacturer wants to minimize the amount of hardware and software that is necessary to deploy on the device.
  • The less they deploy, the smaller the power consumption is and the cost of the devices.
  • So Wi-Fi and Bluetooth are two very widely spread communication standards that are actually crucial for the operation of the internet of things.
  • Their native versions are not necessarily the lowest power or they’re not the lowest power communications standards, but in many instances they’re good enough and can be used for either end devices in internet of things systems, or for hubs that required in internet of things systems.
  • The main focus beyond Wi-Fi and Bluetooth is reduction of power consumption which will be addressed in another lecture.

Week 1: Internet of Things > 1h Wireless Standard: BLE and Zigbee 1 > 1h Video

  • This lecture is devoted to low power wireless communication standards, Bluetooth Low Energy, and Zigbee.
  • Key features of wireless communications systems are the range, the data rates, and power consumption.
  • Then followed up with all other systems whose main reason for existence is to provide as low power communication as possible, providing that a reasonable amount of data method.
  • Differentiation between Bluetooth, Bluetooth Low Energy, and Zigbee is illustrated in this table, where Bluetooth is the original standard built for short-distance communication with relatively small amount of power.
  • All of these systems provide security, but Bluetooth Low Energy and Zigbee go to a greater length to ensure that power consumed is significantly lower as compared to what is currently required in the Bluetooth system.
  • Another feature that distinguishes both BLE and Zigbee is the latency required to establish the communication, which is notably lower than what is required for Bluetooth.
  • Whereas a Bluetooth device can last for several weeks, both Bluetooth Low Energy and Zigbee devices are meant to last for months or years.
  • The fact is that traditional Bluetooth devices cannot operate on a coin cell battery in a meaningful fashion.
  • Traditional Bluetooth is connection oriented, which means that a device is connected to another device.
  • A link is maintained even while the data is not It is true that there are sniff modes, which allow the device to sleep and reduce some power to give a little bit more of battery life.
  • Although it is lower than many other radio standards, it is still not low enough to enable to run these devices using coin cells or some energy harvesting devices.
  • Finally, the state machine that’s running the BLE has been simplified, compared to BL. Whereas traditional Bluetooth is designed for purposes of transmitting significant amount of data, which means we care about the data throughput, BLE’s really designed for purposes of exposing state, saying some critical, meaningful, and limited information about a device.
  • Although BLE can support the data rate of up to one megabit per second, it is not optimized for file data transfer at all.
  • When is data sent can be triggered by some local events.
  • These protocols are implemented typically in system and chip devices, together with radios, such that minimal power consumption is achieved.
  • Layers above are typically implemented in some general purpose computing device.
  • Bluetooth Low Energy, just like Bluetooth, supports a number of profiles, some of them which are not core.
  • Or we might implement controller on some specialized device, and then have a main control processing unit for host and application connected to the HCI.
  • Bluetooth Low Energy advertises channels and has data channels.
  • Advertising channels are used for device discovery, for connection establishment, and for broadcast transmission.
  • Whereas data channels are used for bi-directional communication between connected devices.
  • This illustration shows that we have three advertising channels and 37 data channels in the span of frequencies available in the ISM band.
  • Since this band can be used, and is used, not only by Bluetooth but also by Wi-Fi devices, the goal of Bluetooth Low Energy is to minimize the interference.
  • In the broadcast topology, the Bluetooth Low Energy broadcaster sends the data.
  • All other devices are in a mode where they observe what data is being sent.
  • So when broadcasting the data, the implication is that other devices cannot connect.
  • One advertising packet, a typical one contains 31 bytes of data, which describes the broadcaster and the capabilities of a broadcaster.
  • If the standard packet isn’t large enough, then there is an optional secondary advertising payload, which is called the scan response, which can be used if a device is to detect a broadcast that requests a second advertising frame.
  • So the major limitation of broadcasting compared to regular connection, there is no security or privacy provisioning for it, which means that essentially any observer is able to receive the data which is broadcast.
  • So when do you use a connection? If one needs to transmit data in both directions, or if the data required for transmission is larger than supported by advertising payloads.
  • So connections are a permanent periodical data exchange of packets between two devices.
  • In support of connections, we have the definition of a master and the slave device, a central and peripheral device.
  • So a central device repeatedly scans the preset frequencies for the connectable advertising packets, and then initiates a connection as needed.
  • Once the connection is established, the master manages the timing and initiates the periodical data exchange.
  • On the other hand, the peripheral slave device sends connectable advertising packets periodically, and then excepts incoming connection.
  • In active connection, it has to follow the timing of the master and it exchanges the data regularly with it.
  • So a short analysis as to how long can a device last if we use a coin cell battery.
  • If we only use this device for Bluetooth Low Energy transactions, in theory this could support as many as 15,000 days, or a number of years.

Week 1: Internet of Things > 1i Wireless Standard: BLE and Zigbee 2 > 1i Video

  • So ZigBee has been designed with the thought of having mesh connectivity and be able to consume very small amounts of power and support a large number of devices, about 65,000 devices.
  • The main objectives in designing the ZigBee was to ensure the efficient use of the bandwidth and to provide the platform and implementation for wireless network devices.
  • Essentially, if a device is manufactured and software has been loaded to support a particular application profile, it would be possible to use it straight out of the box with all of the devices that support the same application profile.
  • The standard has been designed to separate the concerns of the specification design and implementation, such that an arbitrary manufacturer can create and coordinate any devices that are using the standard.
  • Networking basics of ZigBee, the networking components are scanning, device discovery, creating or joining a personal area network, service discovery and binding.
  • The devices need to be identifiable by neighboring devices and the services need to be identifiable as well, for purposes of using particular profiles.
  • So any device can initiate the service discovery and the discovery can be performed via gateways from devices outside the network, at least in the future ZigBee releases.
  • Any particular device can be bound to other devices, which right complimentary services and that is supported by command and control facilities for those devices.

Return to Summaries

(image source)

Leave a Reply

Your email address will not be published. Required fields are marked *