Help Me With Sign Wisconsin Assignment of Partnership Interest

Help me with document type sign assignment of partnership interest wisconsin

[Music] I'd like to welcome you today to this week's ECE seminar a our speaker today is dr. Holly Hills he's a distinguished service professor of electrical Computer Engineering and public policy at Carnegie Mellon University twelve Onalaska for developing the state broadcast and telecommunication systems in the 70s and 80s and after that especially after building public radio stations across Alaska developed the state telecommunications networks so the Eaton's small villages television and telephone service des over 19,000 befell director please remember the alaska's inventors all fame and they've been named Alaska's engine of the year to travel over 60 countries in seven contests in connection with his guest lectures from Fotolia project today he's going to talk about building the first Wi-Fi network please help me welcome dr. [Applause] I'm going to tell you a story is xxx technology very about people tell you a little bit about the people that were involved in building this very first Wi-Fi network and by the way it's a reparation of a story the story that's in the book Wi-Fi and the bad boys of radio and I brought a copy as a gift to your department I think that we this will be available to you if you'd like to read it it's kind of a fun story it's not real technical it's mostly about the challenge of the adventure but the story happened at Carnegie Mellon University where I work and it happened way back in 1993 back then we had a big goal a very big goal and the goal was to find a way to provide access to people and information anytime anywhere now that probably doesn't sound like a big deal to you because that's what you have but we didn't have it back in fact our vision and this seemed like a really dramatic vision it's kind of like an Elon Musk vision so it sounds it sounded impossible that it sounded like something that we really might not be able to do but the vision was to have people carrying around small devices always connected to the Internet we did we had seen Star Wars where Captain Kirk has to have his communicator on his vest and that was definitely science fiction stars dogs you know so so this would this was really looked like a pretty big challenge back in 1993 here's what it looked like such a big challenge the state of the art of computers in 1993 now your professors will mostly remember this but this'll sound kind of like ancient history to you first of all we had mainframe computers big computers rats and equipment this these mainframe computers these went all the way back to the 1940s the company I mean I've actually seen a mainframe computer people in the front row I think that's the essential thing you might be able to see one in a museum there's a computer Museum in Seattle that was put together by Paul Allen and they actually have a mainframe computer in there so it is really quite a sight a huge room way bigger than this room and lots of special air-conditioning and things like that then long about the 70s many come so called many computers came along they took up a much smaller space but they were not as powerful in the 80s we started to see the public started to see what we called personal computers like they're going to bear my first IBM PC IBM personal computer think of a desktop computer nowadays maybe a tower under your desk something like that that was and in terms of portability somebody had managed to get one of these into a suitcase and I think we call it portable I'd call it loveable you know but you could actually move it around and then in the early in the early to mid 90s we started to see laptop computers they were pretty big you know but we started to see them and so that was the best anybody could do in terms of portability while I'm on this slide I just wanted to mention our Andrew system at Carnegie Mellon that one started in 1981 and it was kind of a radical idea - this is on a slightly different track than the wireless but it was kind of a radical idea it was a replacement for those big mainframe computers now in a mainframe computer if you were connected by a terminal or a smaller computer you were all the files all the resources were on the mainframe and everyone that was using the mainframe was sharing those files the idea of as a distributed file system where he had agree computers that were located in different parts of the campus the resources the files were on those machines that we would now call service and they were all interconnected by a high-speed network that was a brand-new idea in 1981 but that was the beginning of what we later started to call a client-server computing anyway that's the state of the art of computers hardware in 1993 now the other technology piece of the puzzle here would be Wireless so let me tell you about Wireless in 1993 and you'll see why this was really such a big challenge well we had cellular voice the whole idea of cellular telephones was developed by Bell Telephone labs in the early 60s but it did really wasn't really delivered to the public until the 80s I'd say the early 80s and by the 90s we were still working with some some of that original technology in terms of the original wireless waste technology which I here have labeled as one G we didn't call it one G but it was the first generation was called amps stem stands for advanced mobile phone service it wasn't really very advanced but it was all we had then in the early 90s we started to see the first digital voice service there were two versions of a time division and code division multiple access if we have some Q&A time at the end I can maybe you already know the difference but if we don't know I'll get into that but the most interesting part here is the cellular data so what could you do as far as moving data over wireless and there it started in just about 1993 a technology called CD PD was an overlay on the second generation digital systems so the idea was that they would separate out a few few voice channels and dedicate them to data service rather than voice service so it was just in its fledgling stages but take a look at the speed nineteen point two kilobits per second not megabits kilobits so we're talking about speeds that are roughly plus or minus a thousand times slower than what you have right now and so here's another question anybody ever use a dial-up modem now it'll be the people in the front row so the title of modem that you used was probably a 56 kilobits moment although there were nighttime 9.62 of authorities but it was probably a 56 kilobits so even so this 19.2 kilobits this is only one-third the speed of that dial-up modem which I'm sure you remember was pretty darn slow right so we had two challenges here in order to account in order to accomplish the vision I just mentioned briefly there were a couple of not cellular but separate commercial wireless data services called artists and Grahm but they weren't any help because they also ran a nineteen point two kilobits per second so those were the problems that we faced problems of computer portability I mentioned that and that was to some extent driven by battery size and weight to shrink down to a small device you'd have to get that battery size down but the battery size was driven by the power requirements of the wireless so with cellular you needed a cellular receiver transmitter that could communicate with the cellular network and the big power Bob there was the transmitter and then the display was also a tower guard so some work was needed on battery technology in order to get the size down for software we had some software issues first just with the computer networks computer networks in were designed around the idea that you had computers that were in fixed locations and they were on a fixed wired connection so the IP addresses were static for example with mobility you'd need to you need write dynamic IP addressing and you'd also need to accommodate handoff you that IP device would need to be able to be handed off from one base station to another so that's a little idea on the computer challenges and then as far as the wireless network itself I've mentioned this speed limitation and also the wireless system in those days was limited as the coverage it didn't it wasn't didn't provide coverage everyone okay so those were the big challenges that we faced in 1993 but we had some researchers who were hot on the problem and I said I'd like to talk a little bit about the people one of the people was my colleague professor Danza Warwick who is a well-known professor back there but he's also a dirty hand an engineer and this is a great combination maybe you know some professors that are also dirty kind of engineers in other words they could build something and in this case Dan liked to build small computers he called them wearable computers his vision which was related to our vision but his vision was a computer that you could wear and it goes back to Captain Kirk Star Trek communicator right something that you could wear on your person he headed down to something about the size of this book but maybe a little thicker so Alzheimer's will remember the Walkman radio they keep it hanging on your belt that's what and that was what he and his students were able to build they were able to build small computers not high not high-powered computers but they could do some computing and you could hang him on your belt the size of a Walkman radio each year Dan and his students would build a computer that was that size they called it a form factor the physical size but it was roughly that size but it was faster and more capable than that the students had built the year before so each year there was a class or students project course and their challenge was make it better than what last year's students do so that was Dan and he of course wanted to be able to connect his wearable computers to a wireless network because he wanted them to be connected to the Internet at all times so that part of the vision was the same as our vision here's another one of our guys can you pronounce his name well we couldn't either so so he's a pretty nice guy he's from India and he said well I'll tell you what he said just use the first five letters of my last name s80 why a just those five letters because all the other letters they're just for error correction the last one anyway Sonya's interest was in distributed file systems but had distributed file systems working in the context of that andrew project that I mentioned a little while but the entry project had wired links between the file between the server's and those wired links were there all the time and that were reliable now if you've tried to replace those wired links with wireless links you don't have such a reliable connection he use the word weak connections by diving meant sometimes it works depending on where you are right depending on whether you have a radio pack sometimes you have a connection sometimes you don't have a connection sometimes you may have a high-speed connection sometimes a low-speed convention so he was trying to figure out how to get distributed file systems to work in that kind of an environment being also just like a number of other software researchers wanted a real wireless muffler to try out his ideas so that's what Dan and Satya and the others needed was they wanted a wireless district now at the time I was running the central IT organization you probably have a similar person here was a vice president level position chief information officer so I was responsible for the network and by the way that's a pretty tough job at carnegie-mellon people expect to perfection all the time so that meant that my colleagues and I as well as some of our ECE professor colleagues were said about trying to build this wireless testament that the researchers needed we looked around we scanned the horizon in terms of what equipment was available and the closest things we could find was something called wireless local area network equipment wireless LAN so of course Ethernet existed in those days so a LAN and those thing a local area network was a wired setup and the predominant standard was he by typically 802 three there was also another one called token ring 802 five but 802 three dominated and so we have lighted lands for sure what was the idea of a wireless lamb the equipment that existed at that time the idea was that the building wiring isn't always up to the job at least it wasn't back then we have reliable buildings and when they were tired they were wired for white for a voice service but not people just didn't have data service in mind a lot of times when those buildings were wired so you had situations where the building wiring is inadequate the way this equipment fit in is if you could imagine a computer lab with a series of dentists or tables and desktop computers tower computers whatever but they were they couldn't be interconnected because the wiring was up to the job so the idea of the wireless local area network in those days was it was just wire replacement it was just a way to interconnect these computers by radio and then you'd also have to have a unit the real campus network through a good wired connection so that was what a wireless local area network was it didn't have anything to do with mobility it was just fixed computers and the actual the actual radio units were about this big it was a brick you know it wasn't something that you'd want to carry around with you so that was the closest we could come that was the closest the equipment the problems where the equipment was not portable and then another problem that we would soon discover was the problem of radio propagation anybody taken electromagnetic circuit pop out antennas do you have a person antennas yes okay so some of the things some of these challenges that I'm going to talk about will be pretty familiar to you but anyway that's what we were up against but you soon found a partner and this is a good thing I think you do it here at Boise State and have corporate partners you work with companies and maybe provide you some support and you collaborate in this case our partner was AT&T and you probably know AT&T as the cellular company as a as a telephone company but there was also a part of AT&T that sort of originated from the original metal labs and they were doing Wireless they were actually located in Holland in the Netherlands so I made quite a few trips back and forth across the Atlantic as part of this project anyway and then AT&T later spun off this work into a spin-off company that was called Lucent but anyway they had a wavelength there to wetterlund product called their product Waveland I was there their product Nazy it was one of the ones that I was just describing except they had a goal to move toward mobility so they saw just like we saw that this idea of mobility mobile wired and wireless connections would be important they were operating in the 900 megahertz unlicensed band called is called by SM industrial scientific Manufacturing nothing and they were using for their radio link they were using something called event sequence spread spectrum are you studying any split spectrum in any of your courses anybody know okay well we might come back to that too if we have some time at the end but spread spectrum was being used in these unlicensed low-power systems because well first of all the unlicensed spectrum requires that you use low-power and to establish a rule reasonably reliable communication link under those conditions we have to deal with issues of noise and interference and direct-sequence our spread spectrum was considered a way to overcome noise and interference there were two kinds of spread spectrum direct-sequence spread spectrum and the other one is called frequency off to stretch spread spectrum in my view at the time and my colleagues agreed direct-sequence looked like it was going to have work better it looked like it was going to work better and that actually played out because the Wi-Fi systems that we have today all these direct-sequence spread spectrum now this so we liked that they had we liked what their vision was and so we made a deal we made a partnership in any successful partnership each side gets something that they want right and so this is the way we broke things down 18 ties job was to first of all move their product toward mobility already talked about that also see if they could increase the speed because we didn't get speed problems and most importantly to us was to give us equipment that we could deploy then and on the carnegie-mellon side our challenge was to actually deploy the equipment to build a ultimately it wasn't the large-scale wireless LAN ultimate in the beginning it was a large deal within the in the end it was a large-scale world sign at the beginning it was just a modest goal that let's just cover one building and that was going to be the wireless disability it was the building where the computer science guides were located right so let's just see if we can get we can accomplish our vision in a limited way in just one building that was our goal and then what we were going to do for a TMT was to advise them on what we were running into in deployment so and advise them on improvements that they could make in their product and so it was kind of mutually helpful and it did actually work out very so here's a quick little drawing that shows what a the wireless land that we hit and run with look like schematically just simplistically it says large-scale we're in this land but this actually applies to the single building solution and then also when we extended it to other buildings so the idea is that you have a wired Ethernet the NATO two three local area network and then connected to that that you have these access points which have the capability to communicate with the wired network but then also by radio with laptop computers a little more detail so let's talk about these access points so following the weather the spectrum that ATT is using 900 megahertz we we moved they move toward an access point that we operate at 900 megahertz at a speed of 1 megabit per second it doesn't sound real fast but as I said earlier it was better than 19 the access point was a combination of a radio transceiver a transmitter and receiver and a data Bridge in other words it was a layer 2 device that then connected into the Ethernet and then ultimately that had to be connected to a router somewhere else but you know possibly in another site so that was what the access point was the other piece was the piece that connected to the laptop computer first it was large and then it got smaller but that was what we call the network the network adapter in this case was a two-piece unit I'll show you a picture in just a minute that will help with this but it was a two-piece unit that would work with the laptop computers that appeared around the mid 90s there were two pieces to it what was the electronics for the bridge and the other was the actual radio so the electronics for the bridge was built onto what we then call the PCMCIA card what was PCMCIA this is kind of another part of technology history but a PCMCIA card was a computer card that was about the size of a credit card and laptops those days started to come out with a PCMCIA slot so you could slide this credit card sized thing into your laptop computer and then an egg conductors in there just like you know heroes be connector except this was also a slot for the part - oh man and then that card in the act in the network adapter that card was connected by a cable to another dongle you know whatever you want to call it about this big that you could stick by velcro on the lid of the laptop computers and about a Nintendo Hana so I mean there was a real challenge to submit our miniature Ising all this stuff and this was as far as AT&T was able to go at that moment was a PCMCIA card plugged into the computer and then connected to this dongle and so I'll show you a couple pictures here that and I really searched last week for pictures of these units and I didn't I was not wildly successful but first of all the access point so with the access point you can see the bounding bracket up here on the wall at this technician has taken the electronic piece down off the bracket and he's working on it so you can get a sense there the dimensions are you know something like this of the access point once those access points were installed then they stated a fixed spot in a building and provided coverage of some limited area that was one of the things we needed to figure out was how much coverage area each access might have as far as the adapter the best I can do I pulled this from a cover of like Tripoli personal communications magazine the photographs that we had but I isn't it because this does have a network adapter on it so there's a PCI MCI a slot down here on the other side of the computer and that's got the electronics for the bridge and then the radio unit is is velcroed to the inside of the little laptop computer that's how it was back in those days what can I say so what happened oh I'll tell you what happened through the the next few years as I said in 1993 we started working on the the garrulous testbed in 94 the next year we signed the cooperation agreement with ATT in 1995 toward the end of 95 we finally after great difficulty which I'll describe in a few minutes we were able to get the thing working in one building that was the computer science building got talked about but then but then we kind of had the hang of it so it took us about another year and we and we were feeling good so we deployed into six more buildings so then at the end of 96 we had seven buildings working which seemed kind of miraculous compared to what we were thinking in 93 but there are a lot of buildings at carnegie-mellon just like you have here dozens so that was just seven buildings but what happened in 97 was had a big impact on at work because that was the year that the very first 802 11 standard was adopted now you have lots of different versions and flavors of 802 11 but in the beginning it was just one standard and so with the adoption of that standard that really changed the landscape for us and for a lot of people because up to that point we were just working with proprietary technology there were other companies starting to come into the space each company had their own some use direct sequence on his frequency hop they did this they did that and there was the I Triple E had a standards body and ultimately arrived at this standard that was adopted in 97 so that meant that we had to conform to the standard anybody that wanted to play in this game had to conform to the standard because the standard was the thing they guaranteed the equipment for different companies would interoperate right so that's the so the standard was a big deal but it also there was also a psychological effect too because at that point this wireless land they it migrated from kind of a vision concept a weird idea to something that seemed like it was one so we knew that we had to switch from what we were doing which was arguably the very first network of its size and we've villains ever challenged that to one that conforms a native to 11 so what we had to do was we had to switch all of our equipment and we worked with AT&T on this by this time they were Elizabeth and we switched everything over to 2.4 gigahertz and that was compatible with the with the new standard and in 97 we've done some of the buildings so we said let's do them all and so we got some more help from 80 listen we got some other funding and we put together a campus-wide deployment so now let's talk about the problems that we had in deployment ATT they were working their side of it but even in deployment it sounds simple just to cover a few access points wire into the network really was not that easy and over software issues the ones that are computer science guys were but there were also other deployment issues especially radio issues so I guess some of you are studying electromagnetics and antennas so you'll relate to some of this stuff it's the really these radio issues that are referred to in the title of the book Wi-Fi and the bad boys of radio the bad boy that everybody thinks that I'm a bad boy but it's not actually people it's the it's kind of a metaphor for the challenges these propagation issues so these are some of the things that we ran into first of all when you're studying electromagnetics you already know about free space propagation right and so for others even when you're in free space even when there are no obstructions the power of a received signal on an antenna will fall off as the square of the distance between the and the receive antenna so even in ideal conditions you've got that distance problem powered flawless office one over B squared then but then what about in a real building like this one you've got a walls you've got ceilings what about all that well that's the results for one thing whoops that results for one thing in the shadow and the shadowing problem so as as a radio signal penetrates a wall it's going to experience an additional attenuation above and beyond the free space propagation so that and then the same thing with a ceiling or floor so that obviously is going to have a big impact on the coverage area right it's not just free space it's not just a circle it's some funny shaped pattern that is influenced by the local environment in other words all the walls and ceilings so we went into the others here are also important reflection means that just like light a radio signals these UHF signals could bounce off of a smooth surface or they could be scattered by a rough textured surface those were dishes but not nearly as big as shadowing and then we also had interference of noise because we are in an unlicensed band there were other transmitters that were potentially going to interfere with ours we used to say that the great thing about unlicensed equipment is you don't have to go to an FCC license the bad thing is nobody else has to get one either so there's a lot of signals in the air and that was something that we had but we said about a design method you could call it a design for figuring out where you can place your access points and where the coverages are going to be in some buildings you can get a big coverage area and others you we've got a small coverage area and we really had the dual - step two step process one was where to put the access points and the other one was to what radio channels to assign to the access points which it turns out was a little more challenging than you might think so let's take take those one by one first of all placing the access points no if you just start thinking about this simplistically and to take a simple case of a narrow building where you can place three access points all in the line and get get a complete coverage of the building so that so that's well and good but you would need to space these access points in a way that there would be and would not be any coverage gaps right you want every place to have coverage from one access point or another and that's what we've done in this particular example notice that there are some overlap areas though right there are some places where you have courage from more than one access point is that a good thing or a bad thing well you might think it's not that because what the heck you you can talk to one you can talk to the other it turns out that it's not quite that simple that these coverage overlaps actually could cause a problem that I'll come back to in a minute but anyway there you have a setup for one for one single story arrow building then if you had a single story building that was why then you could just convert it to a two-dimensional same idea but then when you get into a multi-story building things get a little more interesting because signals penetrate through ceilings and floors so if you're on the first floor of the building with your access point you're going to get coverage of that first floor but you're also going to get some coverage on the second floor and maybe in the basement so going back to the simple-minded circles which are really oversimplified but just to communicate the idea now the coverage area could be described as three cylinders stacked one over the other the larger cylinder represents the coverage area on the floor where the access point is located and the smaller one above is the floor above and the smaller one below is the floor below generally speaking practically speaking we did not get coverage usually in fleurs beyond those and just usually the floor above and the floor below because there was nothing ten uation through the ceilings of course that the signal would really drop off say on the third floor so now as you'll see in a minute we really needed to minimize those overlap areas there were a couple of reasons why you might want to minimize them one would be cost if you minimize the number of access points then you would minimize cost that was what we thought in the beginning but it turned out these coverage overlaps were also going to cause other problems and by building suspense sir we're going to cause other problems and so we really wanted to minimize the overlap and we did not want the multi floor installations by kind of interlocking these coverage areas if you could imagine if you're on the second floor access point is on the first floor it's got coverage on the second floor but then on the second floor you need to put an access point that will supplement or complement that coverage area from that person so we ended up staggering them and you can this like first floor and second floor but an access point here on the first floor one up here in the second floor went down here on the first floor went up here in the second floor that's the idea of the the cylindrical shapes and now I've got a drawing here that might be a little hard to follow but this represents say the first floor of the building the solid line the solid lines represent the coverage area of this access point which is located on the first floor now there's also a nexus point which is located on the second floor and it has a solid line coverage area this is the coverage on the first floor right so you see how it kind of dovetails with the first one and then the next access point is back on the first floor that has a solid line coverage area on the first floor which dovetails with the one that was on the second floor so that was the idea that we use in placing these access points in multi-story buildings and then this is the same idea just extended in it's the same idea that I just described it's just okay now I'll resolve the suspense this is the this is where the this is what the channel assignment problem once we have all these access points physicians doing the best that we can on coverage areas but by minimizing the overlaps between the coverage areas now we need to assign channels each access point will operate on a channel and I'll just show you real quickly channel layout for 802 11 interestingly this is still the same for 2.4 gigahertz what is that this is now 20 more than 20 years later on 802 11 2.4 gigahertz it doesn't matter whether it's a B G and AC you name it it's still the same channel ok this channel layout these are the channels that are defined by the standard and in this country in the United States we're only allowed to use channels 1 through 11 that's by FCC because FCC only does a certain amount of spectrum that's allocated for this and the colors if you look real closely here the colors would tell you which channels are available in other nations outside but anyway and notice that these channels is not just a single frequency it's a lot of frequencies why because it's spread spectrum so these are these channels are 20 megahertz wide so are 25 minutes okay so you'd like to assign your access points to channels that do not overlap with respect to spectrum and also that did not physically overlap so you can see with 802 11 in the United States that channels 1 6 and 11 do not have any spectral overlap or at least not a minimal spectrum around these something like this is the impression that there's a sharp cutoff at the edges but it's actually more of a roll off but there's no significant overlap between 1 6 and 11 so at least especially in the early days you would find people using 1 6 and 11 because you knew that those access points would not interact with each other but let's go back now to what's the hat's the main problem what's the problem with channel assignment is you need to avoid Co channel coverage overlap so before I just just talking about coverage overlap between access points and just go back to this one so if these two access points have coverage overlap but if they're on different channels that's okay there's not going to be any interaction but if they're on the same channel that's a problem let me let me explain the problem the multiple access scheme in 802 11 is called csma/ca stands for carrier sense multiple access with collision avoidance now there's a lot of detail to this but the general idea is you don't want to have two stations transmitting at the same time because they're going to interfere with each other and the if you've studied Ethernet you're somewhat familiar with this idea from Ethernet but that's the whole idea is to avoid transmissions on the same channel at the same time in a really rough way it's kind of like truckers on CB radio anybody ever use CB radio again probably the front row we probably if you're on CB radio there's something called radio etiquette which is you listen before talk you just don't start transmitting you listen first and see if there's anybody transmitting on the channel because if they're transmitting and you're transmitting at the same time then probably nobody's going to hear anybody so in the 802 11 standard it's listened before talk and then wait and wait till the channels clear and then only transmit after the journalists clear okay so that's the way 802 11 works so let's assume now that these first two access points are on the same channel and therefore this overlap area is Co channel overlap code in other words they're not only overlap but they're also on the same channel and now let's assume that there's a local station that's in area that means it can be heard by both of these access points so and maybe he's associated with maybe he's connected to access point number one but that's his point number two could here so what does access point number two do nothing listen before talk right so access point one listens here's somebody transmitting okay I can transmit and then there's some detail to this there's something called the back-off algorithm where guess that wait so long before you try again and stuff like that but that's why co-channel overlapped is such a big problem is because if you have these both overlap areas and if they're on the same channel then it's going to really slow down the network and we saw this in our one network now this this problem I'm really the theme of this presentation is the challenges that we faced back in the day back from 20 years ago a lot of this has now been resolved because 802 11 and its extensions is also available at 5 gigahertz and there are really quite a number of channels and five gigahertz so there are a lot more channels possible now and those have cost some of these problems to go away but these are the issues that we face so we jump ahead here this week that is lights him but anyway we developed a detailed design procedure deployment procedure to figure out where to wear position of the access points out of the buildings and involve lots of signal strength measurements our objective though was to provide complete coverage and minimize the exit number of access points both for cost but also for the last point here which is minimized kuk-ja coverage overlap if you minimize the coverage overlap in the first place they have a better chance of minimizing the co-channel coverage overlap when you get to assigning those channels so we brainstormed up a procedure we had our little brain trust which consisted of professors and people like that and then we had the technicians that had to actually go out and do it this gets back to the very handed piece so every week we have a meeting in a conference room and we would try to come up with what's going to be the best procedure what's the step by step in terms of how to lay this out and then our technicians went out and there are two really great people mark I already showed you this picture and Lisa and they went out and they tried that procedure and then they come back to next week and they say well that didn't work very well here's what happened and then we regroup come up with another one so these two technicians I talked about them in the book a lot because they were kind of the heroes of this whole story they just made Gilman's work out there week after week ultimately and in that first year that I described it took us about a year to get this going and they were just really patient and they both still work there they're really great so I'm going to finish up with this one and then we can do the Q&A by 1996 we had the technology problems that were mostly solved we kind of we were kind of feeling good we had the testbed going and our computer science guys and the other researchers they were happy but we weren't sure about the capacity of this thing at this point we are still blazing along at one megabit per second but that was it and we weren't sure how much traffic it could handle so we put in password protection so we were only allowing the researchers to access this network only researchers you had to authenticate on and I always had students come to my office and they'd say we heard there's a wireless network how do we get out there it's like sorry I've been is for you it's guesswork protected but there was one thing that I missed students are smart students are really smart and so they figured out how to pack their way onto the system pretty soon I had hundreds of students somewhere and so that was when the light bulb went on remember this is 1996 maybe it's more to this and sure enough Wi-Fi turned out to be a pretty big deal so the rest is history the book is over here for any of you that are interested if you want your own copy it's on Amazon and ok I'm going to stop there [Applause] slide 34 videos are showing the different spectrums so there's only one I wonder what kind of topology way tell me again the question is what kind of topology because you have spectrums which we're not doing nothing you can assign different spectrums to your also they think the access points are all connected back into the Ethernet network right so that's the technology the is that widget so we were just concerned about these coverage areas right but each of each access point has a coverage area and then it's connected to the wired network like in that slide where I show the access points that were connected to the wired network using hopefully a different channel in those days we had three channels available each access point is one channel right if we could avoid having two of them nearby that we're on the same channel then that was what we tried to do if we only had one gentleman how about if I bring that slide back so everyone understands your question so you really looked at the colors right he's what she's saying is that the color here indicates that in Japan there's only one channel yeah so they couldn't they could not avoid Co general overlap like we did but then the but then the good news for them was then later on the 802 11 a and G a and then they see those went up to five gigahertz and so you know at five gigahertz there are more channels available but you're right back then they didn't they didn't have any way out like we did any others yes being a founder and how do you feel you have any that's a cellular network I live G well all I can say is we do it back then and it's still true people just want to go faster and faster it seemed that there's always an insatiable hunger for speed people just keep coming up with more applications or speed as a consumer I love it what do you think because of asbestos this way your thoughts on net neutrality net neutrality maybe we better say what it is this every should I explain that neutrality okay this is a regulatory problem not necessarily related to wireless but and the FCC has made a couple decisions about this first they made one decision and then the Republican administration they rehearse that decision but to concern about net neutrality and you can help me out see if I get it right is that a content provider on the internet will we'll be given preference by an ISP and so the concern is that the ISP will give priority to certain traffic from certain providers over others and that will mean that certain providers will be getting faster service more reliable service when the that a fair description of net neutrality okay so the idea of net neutrality was that internet service providers are not allowed to do that prohibited and the Commission vested Mississip under the jurisdiction of the Federal Communications Commission when Tom wheeler was the chair he was appointed President Obama he enacted regulations to to do what I just said to ensure net neutrality which meant preventing Internet service providers from giving that kind of preference then President Trump was elected he appointed his own dere CC Commissioner hygiene pi who was already a commissioner but he became chairman from and pi is kind of a free market guy and he said well this isn't really needed we're going to eliminate those regulations so that's where we stand now and that's where when the big controversy happened because there are a lot of people consumer-oriented people that feel like this kind of this kind of precaution is needed from preventing these so then there was a big outcry and I would suspect it's not over yet because Congress can overrule all that the FCC makes regulations but Congress makes laws as they can either agree on anything but so the FCC and I think someone will probably introduce legislation if they haven't already to come back and overrule the FCC and say yes we will okay any others Alba right did you find a significant difference between say modern buildings and the way they're constructed in the wall interior walls very old buildings that were all concrete or not concrete but stone the building's also you be pretty much the same in terms of shadowy reflection we had almost all the old buildings okay so I'm sure there's a difference I'm sure there's a difference as you know metal this real problem steel so it was it was all whole buildings and then getting so even I don't know I'm sure there's a difference though many others are we okay there's one over here okay should we wrap it up thank you thank you very much [Music]