CN101646211A - Vertical-switching control method based on frequency spectrum mapping - Google Patents

Vertical-switching control method based on frequency spectrum mapping Download PDF

Info

Publication number
CN101646211A
CN101646211A CN200910184548A CN200910184548A CN101646211A CN 101646211 A CN101646211 A CN 101646211A CN 200910184548 A CN200910184548 A CN 200910184548A CN 200910184548 A CN200910184548 A CN 200910184548A CN 101646211 A CN101646211 A CN 101646211A
Authority
CN
China
Prior art keywords
mapping
spectral bandwidth
user
network
wlan
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN200910184548A
Other languages
Chinese (zh)
Other versions
CN101646211B (en
Inventor
潘甦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Post and Telecommunication University
Nanjing University of Posts and Telecommunications
Original Assignee
Nanjing Post and Telecommunication University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing Post and Telecommunication University filed Critical Nanjing Post and Telecommunication University
Priority to CN2009101845485A priority Critical patent/CN101646211B/en
Publication of CN101646211A publication Critical patent/CN101646211A/en
Application granted granted Critical
Publication of CN101646211B publication Critical patent/CN101646211B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Mobile Radio Communication Systems (AREA)

Abstract

The invention relates to a vertical-switching control method based on frequency spectrum mapping; radio resources occupied in a CDMA system by a user and QoS requirements thereof are mapped into mapping spectral bandwidth, and then radio resources occupied in a WLAN system by a user and QoS requirements thereof are mapped into mapping spectral bandwidth, and then WLAN/CDMA united resource space isconstructed, service-quality revenue function is defined by the quantity of obtained mapping spectrum on the space, next, a vertical-switching process is controlled, the control target is to lead each calling to have highest total revenue within a duration time. The QoS of the user and the occupied resources in WLAN and CDMA network can be mapped into the mapping spectral bandwidth in a unified way, the mapping spectral bandwidth is used to define the vertical-switching revenue function, and the target of the highest revenue function is used to control switching, thereby improving the utilization efficiency of radio resources while ensuring the service quality of the user.

Description

Vertical-switching control method based on the frequency spectrum mapping
Technical field
The present invention relates to a kind of vertical-switching control method, belong to communication technical field based on the frequency spectrum mapping.
Background technology
Following wireless access system can merge the Radio Access Network of multiple isomery on IP backbone.The Radio Access Network of isomery refers to and adopts different wireless technology systems to carry out user's wireless access.Different wireless technology systems comprises WCDMA system, WLAN system etc.One of key issue that the heterogeneous wireless network convergence system faces is exactly to support user's (or connection) promptly keeping in the switching between heterogeneous network to be transformed into another kind of wireless network from a kind of wireless network when the user connects.This conversion is referred to as vertical switching.Conversion between the different networks must be controlled someway in employing, and the purpose of changing between Control Network is when guaranteeing QoS of customer, and improves the efficient that Radio Resource utilizes as far as possible.
In the vertical-switching control method that has existed, all exist two basic problems: at first, do not consider the unified expression and the quantification of heterogeneous network resource.In existing method for handover control, all with the usable spectrum bandwidth of heterogeneous networks as the user or call out the key parameter of considering when selecting network, yet in CDMA net and WLAN net, the Radio Resource of CU is not expressed with spectral bandwidth, in above-mentioned two networks, the user occupies the whole network bandwidth in the mode of spread spectrum or competition, and available resources are expressed and reserved to the form that therefore is difficult to direct dedicated bandwidth for switching the user.From this aspect, existing method for handover control generally with bandwidth as parameter, and do not provide the corresponding relation of bandwidth and each Internet resources, therefore too shielded the resource characteristics of physical layer and made that method for handover control is difficult to use.Secondly existing method for handover control is only considered with current network state to lack in call duration calling out the global optimization of switching income as handover triggering condition.
We think that Radio Resource according to different dividing mode (in fact also being multi-access mode), has different expression in different systems.Can be expressed as channel (TDMA), code channel (CDMA) or medium (WLAN) as it.Yet because transmission medium all is unified electromagnetic wave, therefore above-mentioned resource is the expression of electromagnetic spectrum when cutting apart in the different orthogonal space.Can be mapped as taking to taking of above-mentioned resource in essence to frequency spectrum.
Summary of the invention
Technical problem: the purpose of this invention is to provide a kind of vertical-switching control method based on the frequency spectrum mapping, the resource that the user takies in said system is unified with the mapping frequency spectrum designation, unified reference when being switched between heterogeneous network, the user is arranged on resource, thereby when guaranteeing QoS of customer, improve wireless resource utility efficiency.
Technical scheme: the present invention earlier is mapped as the mapping spectral bandwidth with Radio Resource and the qos requirement thereof that the user takies when the cdma system, Radio Resource and the qos requirement thereof that the user is taken when the wlan system is mapped as the mapping spectral bandwidth again, then construct WLAN/CDMA federated resource space, how much with acquisition mapping frequency spectrum on this space is switch call definition service quality revenue function, then control vertical handover procedure, the target of control is to make each call out total revenue maximum in the duration.
1.CDMA the map bandwidth of CU in the network
In cdma system, the signal noise ratio of quality of services for users (QoS) and received signal is corresponding, and this corresponding relation is relevant with system modulation mode and signal processing mode, and given quality of service requirement is also with regard to given signal noise ratio thresholding.The signal noise ratio of receiver received signal is function (the different cdma system of bit rate, spreading gain, channel parameter and the overall system bandwidth of signal number, each signal power of arriving receiver, each signal, the expression of signal noise ratio is different, in many code systems, overall system bandwidth can be expressed as the long-pending of chip speed and spreading gain, so overall system bandwidth can not occur in the signal noise ratio expression formula).When the multi-user, this function has constituted the equation with many unknowns group.For certain variable vector (as vector power), exist the condition of separating to be subjected to the common restriction of overall system bandwidth, number of users and signal noise ratio, we can find each user's the weight of signal noise ratio in the mapping spectral bandwidth from this restricting relation, promptly find the mapping relations of service quality to spectral bandwidth.
2.WLAN user's mapping spectral bandwidth in the network
As in time division multiple access (TDMA), expressing Radio Resource with time slot, in wlan system, resource was expressed by the channel occupancy time, yet a user data package is successfully transmitted needed channel time and is not equal to the interchannel that takies when packet really transmits, and this is different with TDMA.Because will wait for channel idle a period of time before the data packet transmission, and transmitted and also do not meant that certain success, therefore this paper time of successfully transmitting the required usefulness of packet from the research wlan network sets about analyzing resource, and and then find out relation between it and the frequency spectrum resource, finally draw the mapping spectral bandwidth and express.
This method is mapped as the mapping spectral bandwidth with quality of services for users QoS and the resource unification that takies thereof in WLAN and cdma network, with the vertical revenue function that switches of mapping spectral bandwidth definition, be target to the maximum with revenue function and control switching; Specifically comprise: the resource that the user takies in WLAN and cdma network is to the mapping method of mapping spectral bandwidth; Vertical switching revenue function and maximization control changing method thereof with the definition of mapping spectral bandwidth;
The first step is calculated the mapping frequency spectrum of user in the multi-code cdma system:
∂ m = ( 1 + 2 v 2 ) C m R m W ( 1 + 2 v 2 ) C m R m + 3 2 G [ 1 - F ( G ) R m ]
Wherein G is a spreading gain, and F (G) depends on covariance function relevant with attenuation process in the WSSUS channel, and parameter v is the transmission coefficient of fading channel, C mBe a Φ mService class is called out used code channel number, R mBe Φ mProfessional required minimum signal noise interference, W is a system bandwidth;
Second step was calculated the mapping frequency spectrum of user in wlan system:
β n = λ n ( 2 - p n 2 - 2 p n ) P S
λ mBe packet generation rate, p nBe φ nThe data packet collision probability of category node, P SIt is the packet size;
In the 3rd step, structuring user's is obtainable normalized mapping spectral bandwidth space B in CDMA and wlan network 1And B 2Can be expressed as respectively:
B 1 = 0 , 1 , 2 , . . . , [ W ∂ 1 ] B 2 = 0 , 1 , 2 , . . . , [ B β 1 ]
1 is a mapping spectral bandwidth unit in the formula, at B 1Representative be that can to obtain to shine upon spectral bandwidth be one
Figure G2009101845485D00034
At B 2In representative be that can to obtain to shine upon spectral bandwidth be a β 1, [] expression rounds;
The 4th step is with mapping spectral bandwidth definition revenue function
r ( s , a ) = f ( s , a ) - μ ( s , a ) - 1 , i ≠ j f ( s , a ) - μ ( s , a ) , i = j
In the formula: f ( s , a ) = f b ( s , a ) = 1 , b a &GreaterEqual; U B ( b a - L B ) / ( U B - L B ) , L B < b a < U B 0 , b a &le; L B Be the link income, L BAnd U BBe respectively needed minimum of user and maximum mapping spectral bandwidth;
&mu; ( s , a ) = b a M a , L B < b a < U B U B M a , b a &GreaterEqual; U B Be network billing function, M aThe price of the unit mapping spectral bandwidth of a action back in network has as a result been taked in expression,
The 5th step. switch with the maximum next controlled switching controls of income, make v &pi; ( s ) = E &pi; s [ E N { &Sigma; t = 1 N r ( s t , A ( s t ) ) } ] Maximum;
In the formula: v π(s) be given initial condition s (the current state s when also being t=1 1) expectation total revenue during from t=1 usage policy π; s tExpression t current state constantly, A (s t) be illustrated in s tThe action of choosing under the state, E NFor on connect hours N, total revenue being asked expectation, E π sFor to given initial condition s, on the tactful π that selects, total revenue is asked expectation;
VIA realizes v with the value iterative algorithm π(s) maximum, process prescription is as follows:
1) right &ForAll; s &Element; S , v is established in initialization 0(s)=0, iterations k=0, ε>0
2), calculate to each state s v k + 1 ( s ) = max a &Element; A { r ( s , a ) + &Sigma; s &prime; &Element; S &lambda;P [ s &prime; | s , a ] v k ( s &prime; ) }
3) if || v K+1-v k||<ε (1-λ)/2 λ, jumped to for the 4th step, otherwise make k=k+1, returned for second step
4) obtain best static policies
&delta; ( s ) = arg max a &Element; A { r ( s , a ) + &Sigma; s &prime; &Element; S &lambda;P [ s &prime; | s , a ] v k + 1 ( s &prime; ) }
In the above-mentioned formula, S represents the residing state of system, and a represents the behavior taked, and ε is an iteration step length, and the average of call duration is 1/1-λ; P (s ' | s, a) be illustrated under the condition that state s taked action a, etching system is in the probability of state s ' when next is made a strategic decision; A is an action space, promptly is the set of behavior a.
Beneficial effect: be expressed as map bandwidth by Radio Resource unification with heterogeneous system, solve the resource in the heterogeneous networks and unified expression problem, and then can define revenue function by map bandwidth, and having provided under the state of heterogeneous networks, how the user switches to obtain maximum total revenue.Therefore when guaranteeing QoS of customer, improved the Radio Resource utilization ratio.
Embodiment
One. at first consider the mapping of multi-service multi-code cdma system frequency spectrum
The first step: list Φ mThe signal noise interference ratio of service class is as follows:
r m = [ ( &Sigma; j = 1 , j &NotEqual; m M C j N j P j + N m C m P m - C m P m ) 1 + 2 v 2 3 2 GP m + F ( G ) + N 0 E bm ] - 1 m=1,…,M(1)
Here P mBe Φ mService class is called out the received power at the every code channel of base station end, and G is a spreading gain, N mBe Φ mThe number of calling out in the service class, F (G) depends on covariance function relevant with attenuation process in the WSSUS channel, parameter v is the transmission coefficient of fading channel, N 0Be the spectrum density of Gaussian noise, E BmBe energy per information bit, it can be expressed as E Bm=P m/ B.C mBe a Φ mService class is called out used code channel number, second step: separate above-mentioned equation group and can get minimum power
P m = ( 1 + 2 v 2 ) C m R m W ( 1 + 2 v 2 ) C m R m + 3 2 G [ 1 - F ( G ) R m ] 3 2 N 0 BG ( 1 + 2 v 2 ) ( W - &Sigma; j = 1 M N j ( 1 + 2 v 2 ) C j R j W ( 1 + 2 v 2 ) C j R j + 3 2 G [ 1 - F ( G ) R j ] ) C m
R wherein mBe Φ mProfessional required minimum signal noise interference .W is a system bandwidth.
The 3rd step was separated P from power mThe necessary and sufficient condition that exists draws the mapping relations of Radio Resource (code channel number), user's qos requirement (error rate requirement) and the frequency spectrum of CU:
&PartialD; m = ( 1 + 2 v 2 ) C m R m W ( 1 + 2 v 2 ) C m R m + 3 2 G [ 1 - F ( G ) R m ]
Relation between every CU frequency spectrum and the total frequency spectrum:
&Sigma; j = 1 M &PartialD; j N j < W
Clearly, in fact above-mentioned inequality " has converted " the CDMA multi-access mode to FDMA mode that different user occupies the different mappings bandwidth.In this way, the reservation of Radio Resource can be expressed with frequency spectrum is unified.
Two then carry out the frequency spectrum mapping of wlan system ground
Φ mA packet of class user's generation can be through the not busy detection of channel busy, transmission wait and possible repeatedly collision until successfully send.Our definition is " equivalent busy channel time " during this period of time, uses T mExpression.During this period of time, network has following situation and takes place: 1) the network free time, this user is in and surveys the stand-by period; 2) network busy, other users have successfully sent packet; 3) network busy, data bump; 4) network busy, this user successfully sends packet.It is as follows that the first step is calculated the average latency:
W m &OverBar; = &Sigma; k = 1 m r + 1 p m k - 1 ( 1 - p m ) &Sigma; n = 1 k CW m ( n ) - 1 2
CW M, minBe Φ mThe minimum competition window of class business is.m rBe the maximum retransmit restriction.CW m(k)=min (CW M, max, 2 K-1CW M, min), k=1 ... m rThe+1st, Φ mThe competition window of class packet when retransmitting for the k time. p m = 1 - ( 1 - &lambda; m 1 - p m ) N m - 1 &Pi; n = 1 , n &NotEqual; m S ( 1 - &lambda; n 1 - p n ) N n , M=1 ... S is Φ mClass user's data bag collision probability
Second step was calculated other users' average transmitting time
At T mIn time, except being investigated user, each Φ mThe data packet number of class user's average generation is λ mT mmBe the packet generation rate).If system is in stable state, so at T mIn time, the packet that each user produces should be sent out on statistics, and except being investigated the user, the total time that every other user expends in the whole system is: ( N m - 1 ) &lambda; m T m T S + T m T S &Sigma; n = 1 , n &NotEqual; m S N n &lambda; n .
The 3rd step was calculated mean collision time
Therefore at T mIn time, the mean collision time of being investigated the user can be expressed as:
T c m &OverBar; = T s &Sigma; k = 1 m r + 1 ( k - 1 ) p m k - 1 ( 1 - p m )
p mBe φ mThe data packet collision probability of category node.At T mIn time, other users of system because of the temporal summation that collision takies are 1 2 [ ( 1 + ( N m - 1 ) &lambda; m T m ) T c m &OverBar; + T m &Sigma; n = 1 , n &NotEqual; m S N n &lambda; n T c n &OverBar; ] .
Equivalence busy channel time T mBe above-mentioned three time sums.
The frequency spectrum mapping of the 4th step
For Φ mClass user, the equivalent busy channel time of successfully transmitting a packet is T m, so the equivalent rate of data packet transmission is 1/T mThe stable condition of wlan system is
Figure G2009101845485D00061
That is:
&lambda; m T m = &lambda; m ( 2 T S + 2 W m &OverBar; + T c m &OverBar; ) 2 + &lambda; m ( 2 T S + T C m &OverBar; ) - &Sigma; n = 1 S N n &lambda; n ( 2 T S + T c n &OverBar; ) < 1
Put in order:
&Sigma; n = 1 S N n &lambda; n ( 2 - p n 2 - 2 p n ) P S < B
p nBe φ nThe data packet collision probability of category node, P SIt is the packet size.The 5th step drew the mapping relations of Radio Resource, user's qos requirement and the frequency spectrum of CU from the stable condition of above-mentioned wlan system:
&beta; n = &lambda; n ( 2 - p n 2 - 2 p n ) P S
Relation between every CU frequency spectrum and the total frequency spectrum is as follows:
&Sigma; n = 1 S N n &beta; n < B .
Equally, in fact above-mentioned inequality " has converted " the statistics time division multiple access way to FDMA mode that different user occupies different spectral.In this way, the reservation of Radio Resource can be expressed with frequency spectrum is unified.
Three. switch the obtainable spectral bandwidth of user
The user the available mapping frequency spectrum of heterogeneous networks what, be that the user selects one of key factor of network.Switching the user can be expressed as this network total spectrum bandwidth in the available mapping spectral bandwidth of certain network and deduct the mapping spectral bandwidth that network Central Plains has the user to take.For the purpose of simplifying statement, consider to support in the network two class business: speech and data,
Figure G2009101845485D00066
The mapping spectral bandwidth that each voice user takies in the expression cdma network,
Figure G2009101845485D00067
Be the bandwidth that each data user takies in the cdma network, β 1, β 2Be respectively each speech and the shared bandwidth of data user in the wlan network.Switch user's obtaining in cdma network and shine upon spectral bandwidth b a = W - &Sigma; j = 1 2 &PartialD; j N j , Obtaining in WLAN shone upon spectral bandwidth b a = B - &Sigma; j = 1 2 &beta; j N j , Because the user's order N in the network jBecome when being, according to N jDifference, switch user's obtainable maximum mapping spectral bandwidth in two networks and be respectively W and B, minimum is 0.
For Markovian decision process being introduced vertical switching controls, must be able to obtain to shine upon the spectral bandwidth discretization and express, so with
Figure G2009101845485D000610
β 1Be respectively radix to CDMA and wlan network obtain shine upon spectral bandwidth and carry out normalization.User's obtainable normalized mapping spectral bandwidth space B in CDMA and wlan network then 1And B 2Can be expressed as respectively:
B 1 = 0 , 1 , 2 , . . . , [ W &PartialD; 1 ] B 2 = 0 , 1 , 2 , . . . , [ B &beta; 1 ]
1 is a mapping spectral bandwidth unit in the formula, at B 1Representative be that can to obtain to shine upon spectral bandwidth be one
Figure G2009101845485D00073
At B 2In representative be that can to obtain to shine upon spectral bandwidth be a β 1, [] expression rounds.
Four. with the link maximum revenue is that target is carried out switching controls
First step definition revenue function; The link income has reflected the degree that user QoS is met, and is defined as:
f ( s , a ) = f b ( s , a ) = 1 , b a &GreaterEqual; U B ( b a - L B ) / ( U B - L B ) , L B < b a < U B 0 , b a &le; L B
In the formula: L BAnd U BBe respectively needed minimum of user and maximum mapping spectral bandwidth.When network can obtain to shine upon spectral bandwidth greater than the required maximum mapping of connection spectral bandwidth, the link income was 1; In the time can obtaining to shine upon spectral bandwidth between user's minimum and greatest requirements, spectral bandwidth is linear to be increased the link income with obtaining to shine upon; Income is 0 in the time can obtaining to shine upon spectral bandwidth less than the required minimum mapping of connection spectral bandwidth.
In order to make the user select factum more reasoningly, the present invention adopts the mode that CU mapping spectral bandwidth is chargeed of pressing.Make M aThe price of the unit mapping spectral bandwidth of a action back in network has as a result been taked in expression, and define grid charging function is:
&mu; ( s , a ) = b a M a , L B < b a < U B U B M a , b a &GreaterEqual; U B
Then revenue function is
r ( s , a ) = f ( s , a ) - &mu; ( s , a ) - 1 , i &NotEqual; j f ( s , a ) - &mu; ( s , a ) , i = j
Second step. obtain switching control strategy with the income maximum
Make v π(s) be given initial condition s (the current state s when also being t=1 1) expectation total revenue during from t=1 usage policy π, it can be expressed as
v &pi; ( s ) = E &pi; s [ E N { &Sigma; t = 1 N r ( s t , A ( s t ) ) } ]
In the formula: S tExpression t current state constantly; A (s t) be illustrated in s tThe action of choosing under the state, i.e. switching behavior; E NFor on connect hours N, total revenue being asked expectation; E π sFor to given initial condition s, on the tactful π that selects, total revenue is asked expectation.
Our purpose requires out to make v exactly π(s) Zui Da switchover policy.
The present invention realizes this goal with VIA (value iterative algorithm).
Algorithmic procedure is described below:
1) right &ForAll; s &Element; S , V is established in initialization 0(s)=0, iterations k=0, ε>0
2), calculate to each state s v k + 1 ( s ) = max a &Element; A { r ( s , a ) + &Sigma; s &prime; &Element; S &lambda;P [ s &prime; | s , a ] v k ( s &prime; ) }
3) if || v K+1-v k||<ε (1-λ)/2 λ, jumped to for the 4th step, otherwise make k=k+1, returned for second step
4) obtain best static policies
&delta; ( s ) = arg max a &Element; A { r ( s , a ) + &Sigma; s &prime; &Element; S &lambda;P [ s &prime; | s , a ] v k + 1 ( s &prime; ) }

Claims (1)

1. vertical-switching control method based on frequency spectrum mapping, it is characterized in that this method is mapped as the mapping spectral bandwidth with quality of services for users QoS and the resource unification that takies thereof in WLAN and cdma network, with the vertical revenue function that switches of mapping spectral bandwidth definition, be target to the maximum with revenue function and control switching; Specifically comprise: the resource that the user takies in WLAN and cdma network is to the mapping method of mapping spectral bandwidth; Vertical switching revenue function and maximization control changing method thereof with the definition of mapping spectral bandwidth;
The first step is calculated the mapping frequency spectrum of user in the multi-code cdma system:
&PartialD; m = ( 1 + 2 v 2 ) C m R m W ( 1 + 2 v 2 ) C m R m + 3 2 G [ 1 - F ( G ) R m ]
Wherein G is a spreading gain, and F (G) depends on covariance function relevant with attenuation process in the WSSUS channel, and parameter v is the transmission coefficient of fading channel, C mBe a Φ mService class is called out used code channel number, R mBe Φ mProfessional required minimum signal noise interference, W is a system bandwidth;
Second step was calculated the mapping frequency spectrum of user in wlan system:
&beta; n = &lambda; n ( 2 - p n 2 - 2 p n ) P S
λ mThe packet generation rate, p nBe φ nThe data packet collision probability of category node, P SIt is the packet size;
In the 3rd step, structuring user's is obtainable normalized mapping spectral bandwidth space B in CDMA and wlan network 1And B 2Can be expressed as respectively:
B 1 = 0 , 1 , 2 , . . . , [ W &PartialD; 1 ] B 2 = 0 , 1 , 2 , . . . , [ B &beta; 1 ]
1 is a mapping spectral bandwidth unit in the formula, at B 1Representative be that can to obtain to shine upon spectral bandwidth be one
Figure A2009101845480002C5
At B 2In representative be that can to obtain to shine upon spectral bandwidth be a β 1, [] expression rounds;
The 4th step is with mapping spectral bandwidth definition revenue function
r ( s , a ) = f ( s , a ) - &mu; ( s , a ) - 1 , i &NotEqual; j f ( s , a ) - &mu; ( s , a ) , i = j
In the formula: f ( s , a ) = f b ( s , a ) = 1 , b a &GreaterEqual; U B ( b a - L B ) / ( U B - L B ) , L B < b a < U B 0 , b a &le; L B
Be the link income, L BAnd U BBe respectively needed minimum of user and maximum mapping spectral bandwidth; &mu; ( s , a ) = b a M a , L B < b a < U B U B M a , b a &GreaterEqual; U B Be network billing function, M aThe price of the unit mapping spectral bandwidth of a action back in network has as a result been taked in expression,
The 5th step. with the maximum next controlled switching of income
Control is switched, and makes v &pi; ( s ) = E &pi; s [ E N { &Sigma; t = 1 N r ( s t , A ( s t ) ) } ] Maximum;
In the formula: v π(s) be given initial condition s (the current state s when also being t=1 1) expectation total revenue during from t=1 usage policy π; s tExpression t current state constantly, A (s t) be illustrated in s tThe action of choosing under the state, E NFor on connect hours N, total revenue being asked expectation, E π sFor to given initial condition s, on the tactful π that selects, total revenue is asked expectation;
VIA realizes v with the value iterative algorithm π(s) maximum, process prescription is as follows:
1) right &ForAll; s &Element; S , V is established in initialization 0(s)=0, iterations k=0, ε>0
2), calculate to each state s v k + 1 ( s ) = max a &Element; A { r ( s , a ) + &Sigma; s &prime; &Element; S &lambda;P [ s &prime; | s , a ] v k ( s &prime; ) }
3) if ‖ is v K+1-v k‖<ε (1-λ)/2 λ jumped to for the 4th step, otherwise makes k=k+1, returned for second step
4) obtain best static policies
&delta; ( s ) = arg max a &Element; A { r ( s , a ) + &Sigma; s &prime; &Element; S &lambda;P [ s &prime; | s , a ] v k + 1 ( s &prime; ) }
In the above-mentioned formula, S represents the residing state of system, and a represents the behavior taked, and ε is an iteration step length, and the average of call duration is 1/1-λ; P (s ' | s, a) be illustrated under the condition that state s taked action a, etching system is in the probability of state s ' when next is made a strategic decision; A is an action space, promptly is the set of behavior a.
CN2009101845485A 2009-08-31 2009-08-31 Vertical-switching control method based on frequency spectrum mapping Expired - Fee Related CN101646211B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2009101845485A CN101646211B (en) 2009-08-31 2009-08-31 Vertical-switching control method based on frequency spectrum mapping

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2009101845485A CN101646211B (en) 2009-08-31 2009-08-31 Vertical-switching control method based on frequency spectrum mapping

Publications (2)

Publication Number Publication Date
CN101646211A true CN101646211A (en) 2010-02-10
CN101646211B CN101646211B (en) 2011-07-06

Family

ID=41657874

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2009101845485A Expired - Fee Related CN101646211B (en) 2009-08-31 2009-08-31 Vertical-switching control method based on frequency spectrum mapping

Country Status (1)

Country Link
CN (1) CN101646211B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106465215A (en) * 2014-04-23 2017-02-22 皇家Kpn公司 Improved vertical handover
CN110519831A (en) * 2019-09-04 2019-11-29 华南师范大学 WIFI and small cell network switching method, system, storage medium and networking component
CN113784411A (en) * 2021-09-13 2021-12-10 展讯通信(上海)有限公司 Link quality evaluation method, link switching method, device and storage medium

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7042858B1 (en) * 2002-03-22 2006-05-09 Jianglei Ma Soft handoff for OFDM
CN101207611A (en) * 2006-12-20 2008-06-25 华为技术有限公司 System and method for processing of ensuring service quality in multi-standard communication network

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106465215A (en) * 2014-04-23 2017-02-22 皇家Kpn公司 Improved vertical handover
CN106465215B (en) * 2014-04-23 2022-05-31 皇家Kpn公司 Improved vertical handover
CN110519831A (en) * 2019-09-04 2019-11-29 华南师范大学 WIFI and small cell network switching method, system, storage medium and networking component
CN113784411A (en) * 2021-09-13 2021-12-10 展讯通信(上海)有限公司 Link quality evaluation method, link switching method, device and storage medium

Also Published As

Publication number Publication date
CN101646211B (en) 2011-07-06

Similar Documents

Publication Publication Date Title
JP4718269B2 (en) Subchannel assignment method in wireless network
CN105827273B (en) The extensive mimo system user double antenna Pilot Interference removing method of multiple cell
CN101754346B (en) Intercell interference suppression method based on channel coherence multi-subscriber dispatching
CN103166685B (en) A kind of interference alignment schemes based on joint Power distribution in LTE
CN103797725A (en) Systems and methods to exploit areas of coherence in wireless systems
CN102056325B (en) Multiple access method based on multiple-input multiple-output antenna
CN102769917B (en) Combinational iteration optimization-based resource allocation method for device-to-device system
CN108650006A (en) Non-orthogonal multiple cut-in method based on extensive MIMO
CN107466099A (en) A kind of interference management self-organization method based on non-orthogonal multiple access
CN102291842B (en) Virtual MIMO pairing method taking user QoS into account
CN106231665B (en) Resource allocation methods based on the switching of RRH dynamic mode in number energy integrated network
CN107302766A (en) Energy efficiency and the algorithm of spectrum efficiency balance optimization in a kind of distributing antenna system
CN101646211B (en) Vertical-switching control method based on frequency spectrum mapping
CN109451569A (en) A kind of resource allocation methods wirelessly taken in energy heterogeneous network
CN101562882B (en) Method and device for allocating power
CN101674118B (en) Weighted rate and maximization-based low-complexity multi-user MIMO scheduling algorithm and device
CN106102173A (en) Wireless backhaul based on multicast beam shaping and base station sub-clustering combined optimization method
CN106793111B (en) Joint virtual MIMO resource allocation methods are matched based on dynamic multiple cell user
CN102970683B (en) Graph theory based multi-cell two-layer network spectrum allocation method
CN112261662A (en) Transmission method for improving energy efficiency of NOMA cooperative communication system
Ferreira et al. Power and delay optimization based uplink resource allocation for wireless networks with device-to-device communications
CN106301501B (en) A kind of instant data transfer optimization method of combined coding modulation
CN107018576B (en) Accidental access method based on grid chart
CN102811189B (en) Probability control-based anti-interference method for cognitive code division multiple access (CDMA) communication system
Sum et al. Performance analysis of low duty FSK system for smart utility network

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20110706

Termination date: 20150831

EXPY Termination of patent right or utility model