0:00:13Q are like
0:00:14so with that and everyone
0:00:16so i would be in about uh that this a power control
0:00:19um i know that uh i i is in and doing thing
0:00:23and resource allocation for wireless multihop networks
0:00:26is don't work with my
0:00:28but a student
0:00:29a a a a like my gone
0:00:30my friend and probably members post you
0:00:34oh i have a is i know that maybe be in the audience no about um i think that being
0:00:39but for those of that then uh we can that action
0:00:43uh if you have they can that of course in the network you know that uh
0:00:47the you you land in our thing is that a big
0:00:51uh where in
0:00:52oh you have a graph and it would the to present as a graph
0:00:55uh you could uh a it is on the go
0:00:57every L
0:00:58as a set and way
0:01:00typically this way to lex
0:01:02the back
0:01:03or have a that's delay or or uh some other
0:01:06uh back to a let's say your delays that this big on that particular thing
0:01:11and we do so this that thing meeting that we compute the do this but when the source to a
0:01:16desired destination
0:01:17and we like it's along long what was but
0:01:20a this of this path thing is well motivated for a number of reasons
0:01:24uh by five to his motivation or the claim of so was but out thing is that is a service
0:01:30and if you
0:01:30you weights that
0:01:32basically got to the reflect
0:01:34uh you have is a delay let's say about link
0:01:37uh what you get
0:01:40but this is a
0:01:42at low system loads because of kind system loads the so this but may not have enough capacity to can
0:01:47bit your traffic
0:01:49and you may need to use
0:01:50secondary path
0:01:52and that a and number of other issues
0:01:54um the going for stuff set this but i think in but the fact that you can set it
0:01:59uh with a relatively low complexity difference
0:02:02i including a distributed algorithms
0:02:04or these a dynamic programming based the average
0:02:09difficult of time and is associated with a this bad that i thing in of the fact that you must
0:02:13know the i've and rate so we can prove a long
0:02:16the service that
0:02:17um i i so that is are going to that they should
0:02:21and fighting
0:02:22all the service that
0:02:23uh that so
0:02:25in fact is slow
0:02:26uh by wireless fading on that
0:02:28that bad to changing i as
0:02:32in the lower than patterns of a lot but or or of that the nodes
0:02:35and most importantly in wireless network in
0:02:38uh i mean
0:02:39uh change is due to fading
0:02:42i i ninety to my uh my colleague as that's you less of that was part of his P the
0:02:48you you use a new but dying for a uh i think
0:02:52no as back last a minute working common
0:02:55uh is that on a concept
0:02:57a that is a concert there a difference of but
0:03:00which measures the sense only
0:03:02a a a a a a a back to is with respect to an ending no
0:03:08and that the decisions are but and this one difference and back
0:03:12a and about the surrounding "'cause" one lots of plane
0:03:15in the network i
0:03:17for this is the most and the reason
0:03:20is is that the K is not maximal stable throughput place so basically
0:03:24but this work as the policy
0:03:25so is probably
0:03:27that if the network can sustain a set and
0:03:31seven and
0:03:34a a and that big you is going to instead but
0:03:38a a a a so we'll will start has a network and push all the packets through the network
0:03:42do you have a site to this nation so it was a maximum flow
0:03:45my can use they can private in that but
0:03:48um the biggest plane
0:03:50we should have as it must recognise the past
0:03:52so it doesn't that a long run only
0:03:55uh let and and the sense
0:03:57because that of to a C is a very way
0:04:01a is this predicted by to a teenager read quickly that's to changing
0:04:06and like sort of stuff helping
0:04:09that it is a
0:04:10so of that a problem as well do not think that can be that
0:04:15a because it uses but that's with to guy packets to the network it there is no
0:04:19in the network that work is empty a back and kid can simply go through a time
0:04:24uh would to is that is basically a you a to go
0:04:27with respect to learn
0:04:28okay there's an active node
0:04:30but nice is delay and if you can be like i think you could in late
0:04:34okay very
0:04:35uh a uh uh a have a Q or something
0:04:37i like but this can we get the sense that it can you can combine the said this
0:04:43that are sets of uh
0:04:45that you can play
0:04:46so can as a a uh a sort of a picture of set that of four by this of thing
0:04:50so that
0:04:51or that that in the network and sitting here
0:04:54i i think about the
0:04:56and let's my by contains the back it's in this particular example
0:04:59uh this guy was that it contains
0:05:01that the packets
0:05:02a the difference in
0:05:04is the maximum of zero and the difference of this look like this
0:05:08so this particular uh example of so uh but we didn't know to and not five is
0:05:14because this does is not known
0:05:19but as we can from these because the you of that guy was is you back risk
0:05:23that's is but
0:05:25a a a a a a a is the that to note is for thing to do this to set
0:05:30of an in between this can know this two
0:05:32but that's of adding for a
0:05:36in that decisions
0:05:39a a a bad thing
0:05:40and you that as a down here it's of any possible to is that the only possible for for the
0:05:44so to get back
0:05:46a packet that has been
0:05:48if you look at a lot what he's this but the you when that in how it can be
0:05:52the fact that the market is
0:05:53that is the low builds that
0:05:55okay one big are given below that this one cup
0:05:58a "'cause" that the idea is proof of that
0:06:02you we have not of this nation's
0:06:06um a multiple source destination pair or even different types of that uh i think this these are a come
0:06:10and it is in networking
0:06:12i then
0:06:13uh actually live that's to do this
0:06:16what happens to you you have to be not use but no
0:06:19well that commodity died
0:06:22a a and then like to do is on it to you take the maximum difference so but so the
0:06:28of of um and that is that a this link
0:06:30or the different so that no
0:06:32and basically you you the entirely the capacity about lee
0:06:35to to do the
0:06:37to the commodity that is a uh i i i and it's about a lot
0:06:41so i if it was ever been i me used to a maximum a can which a maximum
0:06:47task got lost
0:06:49if you use that this
0:06:50not if you
0:06:53uh that but also i have been and some of the mitigation lead to computation you compute D for the
0:06:58difference is basically on
0:06:59let an occasion because i only get them as or
0:07:02but if it so that was use but we and and a handle but not to the rest of you
0:07:06guys for the don't at all
0:07:08so it's a very nice but to the problem
0:07:11the i think a station is a different because of was is you know is a broadcast medium
0:07:16and a and a link that is dependent transmission scheduling and power of are added that the links
0:07:23so this is what i call love a so
0:07:26yeah i where
0:07:27my to do this but increase
0:07:29but the this is this is an example of the values so
0:07:32but also how to me
0:07:33uh a so that playing field
0:07:36you know it's to to
0:07:38so is the physical you um
0:07:41interference patterns it
0:07:43a a to control K this is what the name of the game here is all about
0:07:48so two
0:07:49i like the problem mathematically
0:07:50uh if you got a set and you let's say between a a time
0:07:54uh with of a a not the L L for link L
0:07:59but i'm duration power my recent and parties be L
0:08:03is that of a cost i mean you games between i have a
0:08:06the the time to meet those have a a in this is a like but with a link a and
0:08:11and uh the receiver of link L
0:08:13but time use of power of link eight and then the background noise
0:08:18so this is a lot of in of last noise
0:08:20uh a as you know a the C value of
0:08:22but that
0:08:23i if that sounds st very low
0:08:25a i think that link will be of random is bad
0:08:29the defence
0:08:31i and
0:08:33i would have that i'm to see but would be this one to you the maximum the Q
0:08:37zero and the difference of the queue line
0:08:40and the and
0:08:44that is king
0:08:45oh was a some problem
0:08:47you was so that is
0:08:50that is that's good isn't because it because he he's
0:08:52so so so is E to zero you decide of something
0:08:58so that is
0:08:59and this this
0:09:01oh maximise the reference back but sound
0:09:07oh the it's not bad
0:09:11well as bad for the with an upper bound on a some power of that and the have at time
0:09:17because the right
0:09:19and that's
0:09:20the fans
0:09:22okay you may
0:09:25by a constant
0:09:26so that
0:09:28oh so this problem
0:09:30there was networks
0:09:32we do that he didn't have scheduling and power allocation
0:09:35from the viewpoint of maximizing that works that but that the basic
0:09:47at this
0:09:47now ah
0:09:49so that was like that was my buttons the since then and that in many ways
0:09:56this is a a survey but
0:09:59a a problem time my but to the sun
0:10:04so was not for many he has to be a problem i see
0:10:11but but not count can perhaps the most
0:10:14for example if you get you can sort
0:10:17times with that but not have so either by
0:10:21yeah that was a that this is a
0:10:23this a problem you can smell the
0:10:27if you are so that you the core
0:10:30and you can
0:10:32so for hours by the powers
0:10:35you set of it is but that for you are present
0:10:38a about the time but in the is where Q is it is but that you had but it if
0:10:42not you can be to use one you don't like this
0:10:45this can you see surprisingly
0:10:48well also so to use maximum stable
0:10:51and and there is very nice intuition of that can exploit
0:10:53explain of states
0:10:55if you it
0:10:55well as far
0:10:56but this
0:10:57of course it is a delay but again we could maximise
0:11:00a a maximum throughput
0:11:03the question that the so of this paper is and the need to compete or something a some bit about
0:11:08this bad is leading powers that
0:11:10you know
0:11:11you can choose you can
0:11:12to to between zero and an upper bound
0:11:16but that stay on pattern
0:11:18not not very well
0:11:19a know if you think that this problem for the that have that this is uh it it is a
0:11:25end of the sound of in is a some problem in not networks well as an words in S cell
0:11:32since sensing at this meeting that not i think that was that could be but assume that that that's and
0:11:36once in the difference
0:11:38we have is that as you can have just a in a as you can
0:11:42and that's that like to answer that has that the receiver
0:11:46and this but quite X
0:11:50so you that is part of the optimization invest
0:11:55as is that
0:11:56that is
0:11:57in this and i i have think that now that's a lot for change or games
0:12:03a i think that it is bad and that was that was that that is the for time that that
0:12:09of the channels are best
0:12:10in a direction
0:12:11so you and to for me to do that um
0:12:14so uh has and so by
0:12:17from and is that
0:12:19a a uh hmmm so that this key difference is the best of them require
0:12:24if you in the back
0:12:25so that
0:12:27uh_huh that is that
0:12:29you can
0:12:30basically when you prevent be is used to show that that is
0:12:34and then P that problem
0:12:36and by
0:12:37but are chosen the back
0:12:39you that night
0:12:41so that that them for now that a configuration of back
0:12:44if fact can set so that it can "'cause" not consent to a special configuration if that a configuration
0:12:50so that think that that the that does not sales
0:12:53but has a given
0:12:55i i see that as a
0:12:59is that is or so
0:13:01and that is not to the for this that's me that on this that is not for this me that
0:13:06i in fact is that i have a set of chords and links
0:13:10and the for back into the the that's another
0:13:12okay and that this so that uh
0:13:15now if you have found this that's will discuss look not because of miss it works you also to find
0:13:21that estimation algorithms
0:13:23and we can basically uh
0:13:26and i do an on a scale from a and you layers
0:13:29a a and and and button thousand six
0:13:32i'm going to all the level was the is but the basic idea of scale
0:13:36is to a
0:13:37uh no and
0:13:39uh approximation a in a a and of the sound a uh with the an objective function that is
0:13:46have to that is exact
0:13:48and then to a solution
0:13:49right instead of doing it and that approximation to has now to the specification of those
0:13:54in this you did you that is
0:13:57so this is to your this realisation
0:14:00you you of the sequence of convex problems
0:14:03and use the second step of to move this omission of it
0:14:05or an np-hard problem
0:14:08it can do is with us that is that in that that the of the same that well at the
0:14:13same problem that
0:14:14that problems
0:14:15with different objective function
0:14:17and then
0:14:18a if you have but as it is working but complex this you need something not fast than and that
0:14:25a so
0:14:27that's and look at the algorithms of that to do
0:14:29shall this problem much faster than than it does the service
0:14:33where a is a is it can use here and that was that was the of something that was left
0:14:39and then they fans the
0:14:41this person is ugly
0:14:44it is a it because i was network
0:14:47but uh you have not listen what we talk states
0:14:50the that
0:14:51is like you are you're you use it is post by that that it is by the car so it
0:14:56good like that
0:14:58so there is so that that i is and the network
0:15:00and that let you the solution that you're sitting gets that E is in fact a very different from the
0:15:05previous one because in the previous money make for them are used you
0:15:09a way for
0:15:11so in fact this is why is doesn't work
0:15:13which also said to
0:15:14and we could be able to show the cases
0:15:19well can expect you are they can be here in the network
0:15:22if you choose a window of past solutions in single a which one of the past babble use an issue
0:15:27is better
0:15:30do an excellent rents that
0:15:33a a i a that D V you a and that
0:15:39extensive uh experiments
0:15:41uh it is interesting that that is a thing that and the simulation
0:15:46but we bring signal processing folks we we uh
0:15:49we the level of simulations
0:15:50and i would have to give it just the
0:15:52um so that that you'll and basically i'm just
0:15:56i we just like some of them a result
0:15:59is that is a that is it difficult scenarios uh no this is in that not if and this one
0:16:04that has an is this is kind the file as the interference
0:16:08a a this is a i think that we use that as an approximation of than and become problem
0:16:14in terms of a uh not the last
0:16:16so you should my and seven
0:16:18if you do has access to that two nations that it's is the it's and that for
0:16:23this is that this study of for all our out that is actually do do level and sound of his
0:16:29i have a so that can do a lot of that is the look up as
0:16:33all that could that the a or using all algorithms
0:16:37at at at next
0:16:39at the
0:16:41well as the estimate that sense of those of was that a as a sources
0:16:47i would actually used
0:16:49uh or or as as is what happens with a it is that
0:16:54but have a thing so that was that that it might don't know
0:16:59yeah it a suppose that like sing
0:17:01not i concerned that
0:17:04the problem is that car
0:17:05and the for a approximation
0:17:07so for and
0:17:09so that that the minute i think course
0:17:29at some point you are
0:17:31and and
0:17:32taking not as a constraint but but as the way in the capacity of ring
0:17:36at the beginning of your two
0:17:38yeah and you are
0:17:39acting acting as if
0:17:40the cat the C D's where in that and then
0:17:43and at some point
0:17:45some no
0:17:46we E a
0:17:49contacted by to transmit
0:17:51in this case these is will be we act
0:17:54as a multi a model in the multiple access situation
0:17:59and to get a C D's will not be in that then
0:18:02okay what uh is that that uh
0:18:05you are assume that every E
0:18:07we we
0:18:11a single a of mission yeah will do that to do the fist that's is something like that is you
0:18:15see then
0:18:17this is a nice then the formulation you that
0:18:19that nodes i and everything because
0:18:22you know see still
0:18:24but a these case
0:18:25margin for
0:18:26you could use disk
0:18:28which are set and that mean so that it is not
0:18:32what that that that i when i say that is a a a a a a is a much not
0:18:36stable through
0:18:38assumption sounds