0:00:13okay
0:00:14a could have not everyone
0:00:16um my talking case about a transceiver optimization for multiuser remote a on a each channel
0:00:21and in this work
0:00:22we can see such scenarios
0:00:24well base station at the multiple users exchange a pink and downlink channel uh uh uh i pink and only
0:00:30information
0:00:31where a to will eh
0:00:33a we're a station
0:00:34in such scenarios
0:00:35the
0:00:37given the user interfere with each other
0:00:39and that we exploit the multiple antennas at the base station the station to in to use their interfering
0:00:45and you our work we can see that the i'm don't for the protocol
0:00:50uh here we were talking about using much what time that's to come have to interference
0:00:55that zero focusing is a straight for the solution
0:00:58we have uh
0:00:59system with and user
0:01:01the zero forcing to game need at least a two times and and ten us at the relay station to
0:01:06separate a with and downlink signals
0:01:08oh in pink still
0:01:10and the here and now the interference free game called the signal alignment is propose is proposed to by this
0:01:16paper papers
0:01:16uh the basic idea is that
0:01:18through proper a base station precoding the downlink signals are projected onto the same directions
0:01:24a pink signals
0:01:25then the really station only needs and and a two step rate
0:01:29given the user to imposed the signal
0:01:32and
0:01:32in this paper we have also proposed a of balance this P
0:01:36and they is also a interface for scheme and it can achieve a higher by jewish bidirectional sum rate than
0:01:42both of the zero forcing on the signal i'm i
0:01:45but so we want to ask
0:01:46is this game is good enough and the can we achieve a higher by just close summary
0:01:51all that these questions we want to first all four
0:01:54a performance benchmark uh problems but map for that concerns sister
0:01:59a a a our system or do
0:02:01uh
0:02:02base station and the relay stations uh is delay you keep the with and B and and are and that's
0:02:07each of the and you users
0:02:08is equipped be
0:02:09thing going on a
0:02:11uh here H B R is the channel matrix from base station to the relay station
0:02:15H I R is that
0:02:17channel what or from the i-th user or to the relay station
0:02:20and uh the that will be G and that that will be a a are respectively the
0:02:24base station transmitter and the receiver or which a matrix
0:02:27that we is the really station to zero with me check
0:02:30in the first phase pose the base station and all users transmit to the really station
0:02:35and in the second phase
0:02:36the really stations a cost uh or its received signal
0:02:39to the base station use user
0:02:41we assume that the channels in the two faces
0:02:44uh we C program
0:02:46uh you know we introduce the interference free on screen
0:02:50after the two face transmission
0:02:52uh a the base station and all the users will receive that
0:02:55or what a pink signals and all the downlink signals
0:02:58for the base station
0:02:59it's a knows the downing signals
0:03:01it can remove this signal by
0:03:03self interference cancellation
0:03:05and uh what it has to do is to separate of that
0:03:08a pick signals of from a different user
0:03:10therefore we have our first the into free screen
0:03:13which means that a
0:03:15different
0:03:16the a pink signals from a the users we not interfere with each other at the base station receiver
0:03:22and for each you there
0:03:24or or a pink and downlink signals from other users at in to user interference
0:03:28we should be there are therefore we have a corresponding to intervenes free screen
0:03:33which means that a pink signal and
0:03:35oh you know
0:03:36oh each user we will not interfere don't
0:03:38we were not need for other user
0:03:41well all these interference free "'cause" that come be satisfied
0:03:45we can calculate this is system
0:03:46i pink read and the only three that
0:03:48by this equation
0:03:50here the P V P R and the P U are respectively the transmit power of the base station really
0:03:55station and each user
0:03:57uh the one or more to pretty factor is due to the two phase transmission
0:04:02on of the people meeting on that in a minute it as the first the term is the really station
0:04:05amplified noise
0:04:07and the second term is the noise at the receiver a base station and you use user
0:04:12and they say the bidirectional sum rate
0:04:15uh
0:04:16to at the maximum bidirectional sum rate can be achieved a
0:04:19by talking to lay optimising the
0:04:21street transceiver zero matrix on there these
0:04:24can these days that base station power constraint screen and it is a is the really station power constraint
0:04:29how one
0:04:30this hmmm joint optimization problem is too difficult for us to
0:04:34so they
0:04:35therefore in this work
0:04:36we we start to the alternating optimization
0:04:40two all for or from a span to mark
0:04:43here is our procedure
0:04:44first that we was then initial is for each of the really station to you were and the base station
0:04:49transmitter and uh
0:04:51based there is C we C one
0:04:52then you each that we optimize one of them by fixing the other two
0:04:56after each side
0:04:58we will check is the sum rate higher since that
0:05:01a lot of the last as that
0:05:02if yes to be go to know i mean it's the loop
0:05:05and in the pouring
0:05:06we will respect so with this three sample problem
0:05:10hmmm
0:05:11first we so the
0:05:12we optimize might the relay station ten you work by fixing the base station to work
0:05:17and the
0:05:17we
0:05:18well P mike that that we are to maximise the bidirectional sum rate
0:05:21hi were these days that not convex summary the maximization problem
0:05:25and we use the concept that would be to in this paper a to so to so that kind of
0:05:30problem
0:05:31the basic idea of the with to pull
0:05:33it's clean here
0:05:34the that we are maximising the summary i one plus had two
0:05:38and the say the a two paul a region
0:05:40and each are here corresponds to a with to pull
0:05:43if we maximise the summary
0:05:46uh uh with a P will be to pull we will achieve a boundary point
0:05:49on a two on the triple be treated
0:05:52if we can find the optimal more with me to paul we achieve
0:05:56the optimal boundary point which will result maximum sum rate
0:06:00and that
0:06:01you are what system their with to paul is defined by these white or and
0:06:05these can
0:06:06so the original problem can be the by finding the maximum summary
0:06:11with given
0:06:12bits to poll
0:06:13and then finding the optimal with to collect or
0:06:16here
0:06:17yeah know we use the bisection method or the to search the optimal but the bait
0:06:20and then we have to solve this problem
0:06:24yeah
0:06:25and that so that problem we use the thing approach in this paper
0:06:28and we use the bisection method that to search the maximum summary
0:06:33which satisfies all these constraint
0:06:35we can't be
0:06:36some are eyes and to test a wider it satisfies all that constraint if you yes we try not a
0:06:41one
0:06:42you if know which files more one
0:06:44and the full
0:06:45for a tree
0:06:46summary
0:06:47we comes so this problem to see why there eight
0:06:51satisfies is five these constraints here or not
0:06:53if the minimize
0:06:55mm no here is the lower than one we state
0:06:58the people not rise is feasible
0:07:00and
0:07:01this problem after some mathematical derivation
0:07:04can be
0:07:05we write into this these form they say is the standard
0:07:08code that collect constrained the code that the core problem
0:07:11uh
0:07:12which can be rewritten we which can become what had to be uh
0:07:15semi-definite a problem with a rank one constraint
0:07:18and so we start to the
0:07:19why to use the semi-definite to jen to solve
0:07:22to so wait
0:07:23the procedure from here to here can be found in the journal paper a by provides a
0:07:30and then we and the base station transmitter
0:07:32oh when the really station
0:07:34just he is fixed
0:07:36the base station trans
0:07:37the base station transmitter meter are only affects the town read
0:07:40therefore we
0:07:41maximise the that only grid here
0:07:43a a it is also not comics summary max
0:07:47maximization problem
0:07:48we can use the with to pull here
0:07:51original problem can be
0:07:53can be so by
0:07:55finding the maximum downlink three
0:07:57uh with the people will be to people and since search open more with to pull or
0:08:03and
0:08:04uh a to find is the maximum
0:08:06well to find the maximal maximum only agreed with that be it to pull we can of this problem
0:08:11and we also use the bisection method is two
0:08:14find is a mark or more D which satisfies all these constraint
0:08:18and it these screen
0:08:19um forms a second order cone people read in
0:08:23there for each feed but it the problem can be reading into a second order cone problem the costs
0:08:29a a late we optimize the base station receiver was based a C what only a pink read of the
0:08:35of all we only might the having lead here
0:08:38and the from the a three expression here we found that
0:08:41each user ping
0:08:43read is only a function
0:08:45of the
0:08:46i score along with a of this matrix
0:08:48therefore this problem can be decoupled into
0:08:51the and use up problem
0:08:53each problem optimize one column of this matrix
0:08:57and this problem can be easily be right in into the form of a
0:09:00really racial maximization problem
0:09:02that so
0:09:04hmmm finally i want to talk about the convergence of the alternating optimization
0:09:10uh a to me
0:09:11in that summary increase ease
0:09:13by iterations
0:09:14so the alternating optimization will surely come work
0:09:17but
0:09:18since the original problem is not common
0:09:21therefore of the come to the result depends on the initial value
0:09:25hmmm
0:09:26we cannot guarantee a uh
0:09:28baltimore optimal out but
0:09:29we can perform alternating optimization with mode all different initial is and is and choose the best one which can
0:09:36words to the
0:09:36i is the maximum
0:09:38uh uh which which can to the high the summary
0:09:41and the by doing so we can increase the probability to achieve optimal
0:09:44so
0:09:46here is the some some simulation results first i want to show the
0:09:50a bidirectional sum rate was as uh
0:09:53each region number
0:09:54uh the in this speaker the release a number at a number is that has full the base station and
0:10:00the number and the use number instead
0:10:02a set that's two
0:10:03the
0:10:04in the power of each user is no that's one and the
0:10:07and those of the base station the it's it has set has two
0:10:11this three blue curve
0:10:13uh the convergence performance of you
0:10:16by using the you need to let was as the uh
0:10:19and is this again uh
0:10:21zero for the a game and the signal i gay
0:10:23and the
0:10:24to read curve
0:10:25uh the convergence performance by using a me use initial value
0:10:29the lower lower use these only one initial values and
0:10:32a problem right curve use
0:10:34use it time a random initial values and the choose is the best one
0:10:38we can see that by using different in usual is the come were to the result out E
0:10:43and a by using multiple initial values and the choose the best one we can increase the performance
0:10:49and is a single galatians we found that
0:10:51by using more and time take initial values
0:10:54the
0:10:55performance gain or were these right curve
0:10:57is my to or four we can take the
0:11:00right curve as a new optimal solution and sorrow as a bit performance but file
0:11:06and in this curve
0:11:07in this speaker are we compare as a sum rate performance of different at time zero
0:11:11it's
0:11:12uh okay as the base station the you base station and a number and the use a number as that
0:11:17has to
0:11:18uh the how uh
0:11:20note chance or a set has is
0:11:22uh you know a the bidirectional sum rate words as the relay station on a number
0:11:26because see that the
0:11:27traditional zero forcing to game
0:11:29performs performs bad out when the at than and when the
0:11:33and than a number at the relay station
0:11:35is
0:11:36is a low
0:11:36and it out performs the signal i'm and
0:11:40when the when when the relay station and the number is a lot
0:11:43and are well form our proposed to balance this again outperformed performs ball
0:11:48and as the
0:11:48but curve here
0:11:50is the alternating an eighteen optimisation
0:11:52we can see that the performance gap between the
0:11:55alternating optimization and a lot from our proposed the balance gain is quite small
0:12:00therefore by four by proposing this performance but of we can see that our form proposed a scheme is the
0:12:06new optimal solution
0:12:08yeah i'd like to conclude my talk
0:12:10uh uh this work we employed of alternating optimization
0:12:13these to design a base station and a station transceiver us
0:12:16you know a two we will a system
0:12:19the in at to we aim at maximizing the bidirectional sum rate on the interference free tree
0:12:25and we use a multiple initial values to perform alternating my station and select like the best one
0:12:30by doing so we can treat the probability to achieve the global optimal solution
0:12:35and the and the of the a the formants can be taken as a performance a by back for the
0:12:39can them
0:12:41and that we found that performance gap between the bit to our form a proposed the it that's again and
0:12:46alternating optimization is small
0:12:48it indicates that the balance this game is then the optimal solution
0:12:52these uh some reference you use the in these talk
0:12:55uh in this paper as you can find of the signal and scheme
0:12:59and that they is our uh and of the scheme
0:13:02and they
0:13:03uh uh you this paper as you can send is uh
0:13:06the concept of a it to pull
0:13:08and they they say a in this it provides a lot paper you can fans them
0:13:12techniques the a convex optimization
0:13:14that's source and Q
0:13:28sorry um
0:13:29so you you mentioned a your that the is for to really a your systems
0:13:34but made it is really thing with a so
0:13:36just or and can you comment on
0:13:38just just to about so interference "'cause" it seems like it's going to
0:13:42um um
0:13:43drastically in the to roots so you gonna achieve
0:13:46uh yeah no i you insist can we only consider a single cell scenario
0:13:51and is the intra sell you and the so so there is no inter self interference
0:13:56and i was interest out interference a handled by
0:13:59these
0:14:00by these
0:14:01no interference a constraint
0:14:03the you
0:14:04a three constraint
0:14:06i is that the we will
0:14:07we we don't have the rip we there will be no interest that interference
0:14:12but uh and they
0:14:13that's a
0:14:14yeah
0:14:21what do you to just three constraints
0:14:24so this just a a a a uh a
0:14:27the shows use six
0:14:29sorry what we have been to students G constraint well don't to just
0:14:33the max most the rate
0:14:34oh yeah why don't use you can do as you have two reasons for was
0:14:38for was it for the constraint
0:14:39the first one we can well oh
0:14:41because there's that if the
0:14:43i so nice highly nap
0:14:44the therefore for
0:14:46by doing by doing so the sum the we will be
0:14:49uh the same as to what you does the same we be to is don't to call you pick can
0:14:53see that the just to maximise the sum rate
0:14:56right
0:14:56and the otherwise that the back can see didn't that can screen
0:14:59and uh
0:15:00we will achieve them
0:15:02good mathematical result
0:15:05because the
0:15:07summary here are we'll be where simple
0:15:22yeah