CN110311719A - A kind of beam selection method and its device applied to the extensive mimo system of millimeter wave - Google Patents
A kind of beam selection method and its device applied to the extensive mimo system of millimeter wave Download PDFInfo
- Publication number
- CN110311719A CN110311719A CN201910705982.7A CN201910705982A CN110311719A CN 110311719 A CN110311719 A CN 110311719A CN 201910705982 A CN201910705982 A CN 201910705982A CN 110311719 A CN110311719 A CN 110311719A
- Authority
- CN
- China
- Prior art keywords
- bird
- nest
- user
- mimo system
- millimeter wave
- 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
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/0413—MIMO systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/0413—MIMO systems
- H04B7/0456—Selection of precoding matrices or codebooks, e.g. using matrices antenna weighting
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
- H04B7/0686—Hybrid systems, i.e. switching and simultaneous transmission
- H04B7/0695—Hybrid systems, i.e. switching and simultaneous transmission using beam selection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/08—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the receiving station
- H04B7/0868—Hybrid systems, i.e. switching and combining
- H04B7/088—Hybrid systems, i.e. switching and combining using beam selection
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Radio Transmission System (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The present invention discloses a kind of beam selection method and its device applied to the extensive mimo system of millimeter wave, the beam selection method includes: step S1, first define Bird's Nest number, Bird's Nest probability of detection, binary coding control parameter, maximum number of iterations, reinitialize multiple Bird's Nests, finally calculates the fitness of multiple Bird's Nests；Step S2 carries out binary coding mixing and updates, and repairs improper coding, calculates the fitness of newly generated multiple Bird's Nests, by retaining the biggish Bird's Nest of fitness to screen Bird's Nest；Bird's Nest probability of detection is compared with random number, replicates the Bird's Nest of globally optimal solution to replace the Bird's Nest that one of them is found, and change the position for the Bird's Nest that remaining is found at random by step S3；Step S4, judges whether the number of iterations reaches maximum number of iterations, is, exports globally optimal solution, no to then follow the steps S2.The relatively digital precoding algorithms of the present invention not will cause significant performance loss, reduce the complexity of algorithm, obtain nearly excellent system performance.
Description
Technical field
The present invention relates to a kind of beam selection methods of mobile communication technology field, more particularly to one kind to be applied to millimeter wave
The beam selection method of extensive mimo system, be related to the beam selection method is applied to the extensive mimo system of millimeter wave
Beam selection device.
Background technique
In the rapid development with Internet service, demand of the people to each application field of wireless network is increasing,
The frequency spectrum resource of growing tension can no longer meet demand of the people to communication.The extensive input and output of millimeter wave can be by wider
Signal bandwidth realize higher data rate and higher spectrum efficiency, it is considered to be the crucial skill of the following 5G wireless communication
Art.Traditional digital Wave beam forming scheme requires the corresponding independent radio frequency link of each antenna, with antenna for base station number
With being continuously increased for community user number, required RF number of links is also constantly rising, although haveing excellent performance, increases hard
Part cost and realization difficulty.
The discrete lens array ignored in the prior art by using performance loss converts traditional space channel
For beam space channel, to obtain the channel degree of rarefication under millimeter-wave frequency.But since beam space mimo channel is with dilute
Property is dredged, need to only select a small amount of suitable antenna just to can be reduced RF number of links, and not will cause apparent performance loss.Together
When, scheme used at present further includes that selection elimination causes the smallest wave beam of capacitance loss, selection to be contributed in terms of power system capacity
Maximum wave beam, amplitude peak beam selection algorithm, interference perception beam selection scheme, the scheme based on ant group optimization.Wherein,
The selection scheme for leading to capacitance loss minimum wave beam is eliminated in selection and the choosing of maximum wave beam is contributed in selection in terms of power system capacity
Scheme is selected, both schemes require successively to be searched for, and complexity is excessively high.This scheme of amplitude peak beam selection algorithm is most
Simply, but there is multi-user interferences and different RF chains to select what RF chain caused by identical wave beam was wasted to ask
Topic.Interference perception beam selection scheme is that interference user reselects wave beam, but its complexity is disturbed the influence of number of users.
It is similar with interference perception beam selection scheme based on the scheme of ant group optimization, it is all based on the standard of amplitude maximization, rather than
Directly optimization and rate, lifting system limited capacity.Therefore, there are complexity or energy losses for these existing beam selection schemes
It is excessively high, the problem of calculating overlong time, not being suitable for real system.
Summary of the invention
Problem in view of the prior art, the present invention provide a kind of beam selection applied to the extensive mimo system of millimeter wave
Method and device thereof solve these existing beam selection schemes there are complexity or energy loss are excessively high, calculate overlong time,
Not the problem of not being suitable for real system.
The present invention is implemented with the following technical solutions: a kind of beam selection side applied to the extensive mimo system of millimeter wave
Method comprising following steps:
Step S1 first defines Bird's Nest number, the Bird's Nest probability of detection, binary coding of an extensive mimo system of millimeter wave
Control parameter, maximum number of iterations, antenna number and number of users, reinitialize multiple Bird's Nests, and each Bird's Nest is that user selects letter
Road range value is maximum and without duplicate wave beam, calculates the fitness of multiple Bird's Nests, and finally with the maximum adaptation degree of current Bird's Nest
For globally optimal solution；
Step S2 first carries out binary coding mixing to multiple Bird's Nests and updates after the fitness for calculating multiple Bird's Nests, and
Repair improper coding, then calculate the fitness of newly generated multiple Bird's Nests, finally by retain the biggish Bird's Nest of fitness with
Screen Bird's Nest；
Step S3, after all Bird's Nests screen, by the Bird's Nest probability of detection and obey equally distributed random number into
Row compares, if the random number is greater than the Bird's Nest probability of detection, replicates the Bird's Nest of the globally optimal solution to replace wherein
A Bird's Nest being found, and change the position for the Bird's Nest that remaining is found at random, retain globally optimal solution to accelerate algorithm receipts
Speed is held back, determines Bird's Nest position and the optimal value of current optimal solution；
Step S4 judges whether the number of iterations reaches the greatest iteration time after determining Bird's Nest position and optimal value
Number, is to export the globally optimal solution, no to then follow the steps S2.
As being described in further detail for above scheme, in step s 2, it includes following step that binary coding, which mixes update method,
It is rapid:
Step S21, judges whether the random number of system is not more than the binary coding control parameter；
When the random number is not more than the binary coding control parameter, step S22 is executed, using sigmoid letter
Real number is mapped to discrete binary data by number, is calculated by the following formula k-th of the m+1 times iteration, n-th of user antenna
State value
When the random number is greater than the binary coding control parameter, step S23 is executed, judges Levy hop path
It whether is positive number；
When Levy hop path is not positive number, step S24 is executed, is calculated by the following formula the m+1 times iteration k-th
The state value of n-th of antenna of user
When hop path is positive number, step S25 is executed, is calculated by the following formula the m+1 times iteration, k-th of user the
The state value of n antenna
Wherein, rand is the random number.
Further, the improper coding restorative procedure the following steps are included:
Step S26 successively calculates the state value x of k-th of m-th of Bird's Nest, n-th of user antennank, that is, calculate m-th
The state value of matrix kth column line n, and judge state value xnkIt whether is 1；
In state value xnkWhen being 1, step S27 is executed, corresponding channel gain g is calculatednk；
All state value x in each column for judging the binary coded matrixnkAfterwards, step S28 is executed, by following
The wave beam and beam collection of formula selection user:
Wherein, βkFor the wave beam of k-th of user,The beam collection chosen for user.
As being described in further detail for above scheme, the beam selection method is further comprising the steps of:
Step S0 constructs the extensive mimo system of the millimeter wave；Wherein, the structure of the extensive mimo system of the millimeter wave
Construction method the following steps are included:
Step S01, the expression formula of the reception signal of k-th of user of preliminary definition；
Traditional Space channel is converted to wave by spatial Fourier transform matrix using discrete lens array by step S02
Beam space channel, and according to the beam space channel define again k-th of user reception signal expression formula；
Step S03 only needs selected part wave beam according to the sparsity of beam space mimo channel, establishes the mesh that antenna is selected
Scalar functions and constraint condition.
Further, in the extensive mimo system of the millimeter wave of the present embodiment, the expression of the reception signal of k-th of user
Formula is first is defined as:
Y=HHWs+n
In formula, H is channel matrix, H=[h1,h2,...,hK], hkFor the channel vector between k-th of user and base station；s
For original signal vector, s ∈ CK×1, and normalized power E (ssH)=IK；W is the pre-coding matrix having a size of N × K, and is met
tr(WWH)≤ρ, ρ are total emission power；N is the additive white Gaussian noise having a size of K × 1, and n~CN (0, σ2IK)。
Still further, the calculation formula of the channel vector are as follows:
In formula,For the los path of k-th of user,For the non line of sight road of k-th of user
Diameter；GkFor complex gain, ψkFor dimensional orientation,For array response vector.
Still further, the spatial Fourier transform matrix are as follows:
In formula,
As a result, in the extensive mimo system of millimeter wave, the expression formula of the reception signal of k-th of user defines again
Are as follows:
Wherein,For beam space channel, and calculation formula are as follows:
For the beam space channel vector of k-th of user, and k=1 ..., K.
Still further, the expression formula of the downlink received signal are as follows:
In formula,For the channel of selected wave beam composition, Wr∈CK×KAnd to have reduced the digital precode matrix of dimension.
Still further, the objective function are as follows:
In formula, xnkFor the state value of k-th of user, n-th of antenna；RkFor the achievable Mean Speed of k-th of user；
Wherein, the calculation formula of the Mean Speed are as follows:
In formula, σ2For noise power,Wr∈CK×K；α is zoom factor and satisfactionρ is total emission power.
The present invention also provides a kind of beam selection devices applied to the extensive mimo system of millimeter wave, using above-mentioned
Any beam selection method applied to the extensive mimo system of millimeter wave comprising:
Fitness computing module is used to first define Bird's Nest number, the Bird's Nest discovery of an extensive mimo system of millimeter wave
Probability, binary coding control parameter, maximum number of iterations, antenna number and number of users, reinitialize multiple Bird's Nests, makes each
Bird's Nest calculates the fitness of multiple Bird's Nests, and finally for user's selection channel magnitude maximum and without duplicate wave beam with current bird
The maximum adaptation degree of nest is globally optimal solution；Wherein, the calculation formula of the fitness of multiple Bird's Nests are as follows:
In formula, xnkFor the state value of k-th of user, n-th of antenna；RkFor the achievable Mean Speed of k-th of user；
Wherein, the calculation formula of the Mean Speed are as follows:
In formula, σ2For noise power；α is zoom factor and satisfactionρ is total emission power, and K is to use
Amount；H is channel matrix, hkFor the channel vector between k-th of user and base station；For the channel of selected wave beam composition；Wr∈CK×KAnd to have reduced the digital precode matrix of dimension；
Bird's Nest screening module is used for after the fitness computing module calculates the fitness of multiple Bird's Nests, first to more
A Bird's Nest carries out binary coding mixing and updates, and repairs improper coding, then calculate the fitness of newly generated multiple Bird's Nests,
Finally by the reservation biggish Bird's Nest of fitness to screen Bird's Nest；
Bird's Nest position replacement module is used for the Bird's Nest screening module after screening all Bird's Nests, the Bird's Nest is sent out
Existing probability is compared with equally distributed random number is obeyed；If the random number is greater than the Bird's Nest probability of detection, the bird
Nest position replacement module replicates the Bird's Nest of the globally optimal solution then to replace wherein be found Bird's Nest, and changes at random
The position of remaining Bird's Nest being found, with the Bird's Nest position of the current optimal solution of determination and optimal value；And the number of iterations judgement
Module is used in Bird's Nest position replacement module after determining Bird's Nest position and optimal value, whether judges the number of iterations
Reach the maximum number of iterations, be, export the globally optimal solution, the Bird's Nest screening module is otherwise driven to work.
Compared to existing beam selection scheme, the beam selection applied to the extensive mimo system of millimeter wave of the invention
Method and device thereof have the advantages that
The beam selection method for being applied to the extensive mimo system of millimeter wave sees beam selection as solution { 0-1 } back
Packet problem is simultaneously solved, and rf chain number needed for reducing can be avoided without causing apparent performance loss using total
A large amount of radio frequency links needed for word coding cause energy loss excessively high, and nearly excellent system can be obtained under the premise of reducing complexity
Performance.Due to being easy to appear improper coding in solution procedure, the present invention repairs improper solution, ensure that calculating gained
Solution be feasible solution.Meanwhile the present invention replicates the Bird's Nest of global optimum to replace the Bird's Nest being wherein found, and makes optimal Bird's Nest
It preserves, accelerates convergence, otherwise without the interference considered between user, directly using optimization and rate as target,
Therefore realize compared to existing several beam selection schemes and rate capability is best.In addition, the present embodiment is discrete cuckoo
Searching algorithm increases application scenarios.
In the solution of the present invention, beam selection method sees the wave beam of selection as selected item and is packed into knapsack, will be
The maximum and rate of system regard the maximum capacity problem that is filled of knapsack that solves as, and using the discrete cuckoo searching algorithm of improvement come
Near-optimum solution is obtained, for the improper coding generated in cuckoo algorithm Levy flight discretization results, using heuristic greediness
Algorithm is repaired.Meanwhile the duplication in genetic algorithm is introduced into DCS algorithm by the present invention, the Bird's Nest for replicating global optimum comes
The Bird's Nest being wherein found is replaced, accelerates algorithm the convergence speed, and mentioning by analytical proof beam selection scheme of the invention
It is better than existing beam selection scheme in terms of high system summation rate.
Detailed description of the invention
Fig. 1 is the flow chart of the beam selection method applied to the extensive mimo system of millimeter wave of the embodiment of the present invention 1.
Fig. 2 is constructed by the beam selection method applied to the extensive mimo system of millimeter wave of the embodiment of the present invention 2
System model figure.
Fig. 3 is the algorithmic statement curve graph of beam selection method provided by the embodiment of the present invention 2.
Fig. 4 be the embodiment of the present invention 2 provided by beam selection method under Sparse System with different beams selection algorithm
Realize the analogous diagram compared with rate.
Fig. 5 be the embodiment of the present invention 2 provided by beam selection method under congestion system with different beams selection algorithm
Realize the analogous diagram compared with rate.
Fig. 6 is beam selection method provided by the embodiment of the present invention 2 under different user number and rate compares analogous diagram.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not
For limiting the present invention.
Embodiment 1
Referring to Fig. 1, a kind of beam selection method applied to the extensive mimo system of millimeter wave is present embodiments provided,
The beam selection method can be used for selecting wave beam in an extensive mimo system of millimeter wave.In the present embodiment, for wave
Beam preference pattern optimization problem proposes to see beam selection problem as solution { 0-1 } knapsack problem, the wave beam of selection is regarded as
For article is packed into knapsack, regard the maximum and rate of system as maximum capacity problem that knapsack is filled, using discrete cuckoo
Algorithm is come the model that solves.CS algorithm is based on following three hypothesis:
(1) every cuckoo randomly chooses a Bird's Nest and only produces next cuckoo bird egg；
(2) best Bird's Nest will be retained to the next generation；
(3) quantity of Bird's Nest is fixed, and the probability that the cuckoo bird egg in Bird's Nest is found by host bird is Pa∈[0,1]。
The formula of Levy flight hop path, the basic cuckoo algorithmic formula of Levy flight are simulated using Mantegna
It is as follows:
Discrete cuckoo algorithm require to fly to the Levy hop path of each location updating carries out binary code transformation.
Based on this, the beam selection method applied to the extensive mimo system of millimeter wave of the present embodiment includes the following steps (step S1-
S4)。
Step S1: Bird's Nest number, the Bird's Nest probability of detection, binary coding of an extensive mimo system of millimeter wave are first defined
Control parameter, maximum number of iterations, antenna number and number of users, reinitialize multiple Bird's Nests, and each Bird's Nest is that user selects letter
Road amplitude is maximum and without duplicate wave beam, finally calculates the fitness of multiple Bird's Nests, and be with the maximum adaptation degree of current Bird's Nest
Globally optimal solution.In the present embodiment, the parameter of beam selection method: Bird's Nest number M, Bird's Nest detection probability P is first seta, two into
Coding control parameters p processedr, maximum number of iterations Tmax, antenna number N, number of users K.Then, M Bird's Nest, each Bird's Nest are initialized
It selects channel magnitude value maximum for user and is then this if there is the duplicate wave beam of multiple users selection without duplicate wave beam
Multiple users randomly choose unduplicated wave beam from remaining wave beam.Finally, calculating the fitness size of M Bird's Nest, that is, calculate
The objective function that antenna is selected is globally optimal solution with current Bird's Nest fitness maximum value.Objective function is indicated in constraint item
System and rate are maximized under part.
Step S2: after the fitness for calculating multiple Bird's Nests, first carrying out binary coding mixing to multiple Bird's Nests and update, and
Repair improper coding, then calculate the fitness of newly generated multiple Bird's Nests, finally by retain the biggish Bird's Nest of fitness with
Screen Bird's Nest.The present embodiment carries out binary coding under the Parameter Conditions that above-mentioned steps S1 gives, by the path of Levy flight
Mixing updates, and is repaired using heuristic greedy algorithm to improper coding.Then, the suitable of new M Bird's Nest of generation is calculated
Response, if than former Bird's Nest fitness greatly if replace original Bird's Nest, abandoned if smaller than former Bird's Nest fitness；If current Bird's Nest is suitable
Response is bigger than globally optimal solution, then using current Bird's Nest fitness as globally optimal solution.In the present embodiment, binary coding mixes
Update method includes the following steps (step S21-25):
Step S21, judges whether the random number of system is not more than binary coding control parameter；
When random number is not more than binary coding control parameter, step S22 is executed, is reflected real number using sigmoid function
Discrete binary data is penetrated into, the state value of k-th of the m+1 times iteration, n-th of user antenna is calculated by the following formula
When random number is greater than binary coding control parameter, step S23 is executed, judges whether Levy hop path is positive
Number；
When Levy hop path is not positive number, step S24 is executed, is calculated by the following formula the m+1 times iteration k-th
The state value of n-th of antenna of user
When hop path is positive number, step S25 is executed, is calculated by the following formula the m+1 times iteration, k-th of user the
The state value of n antenna
Wherein, rand is random number, pr∈[0,1]。prBigger, then the global diversity of discrete cuckoo algorithm is stronger；pr
Smaller, then discrete cuckoo convergence is stronger.
In addition, the binary coding method that Levy flight updates is easy to generate improper coding, example under constraint condition
As k-th of user selects more than one wave beam in m-th of Bird's Nest of i-th iteration.In order to guarantee to be feasible solution, it is necessary to use one
Fixed coding correcting strategy, set forth herein use heuristic greedy algorithm to repair improper coding.The present embodiment is first right
Levy updated 1 item of being encoded to that flies carries out k-1 nonoptional and width before asking amplitude size, k-th of user to select
The maximum wave beam of angle value, successively selects, until all users have selected a wave beam.Before reparation, input are as follows: having a size of N
The binary coded matrix of × K exports after reparation are as follows: the size of compound constant condition is the binary coded matrix of N × K.Tool
For body, the restorative procedure of improper coding includes the following steps (step S26-28).
Step S26 successively calculates the state value x of k-th of m-th of Bird's Nest, n-th of user antennank, that is, calculate m-th
The state value of matrix kth column line n, and judge state value xnkIt whether is 1；
In state value xnkWhen being 1, step S27 is executed, corresponding channel gain g is calculatednk；
All state value x in each column for judging binary coded matrixnkAfterwards, step S28 is executed, following formula is passed through
Select the wave beam and beam collection of user:
Wherein, βkFor the wave beam of k-th of user,The beam collection chosen for user.
Step S3: after all Bird's Nests screen, Bird's Nest probability of detection is compared with equally distributed random number is obeyed
Compared with, if random number is greater than Bird's Nest probability of detection, replicate the Bird's Nest of globally optimal solution to replace wherein be found Bird's Nest,
And change the position for the Bird's Nest that remaining is found at random, with the Bird's Nest position of the current optimal solution of determination and optimal value.In this reality
It applies in example, with the detection probability P of external eggaIt is compared with equally distributed random number R ∈ [0,1] is obeyed, if R > Pa, then
Using the thought replicated in genetic algorithm (GA), the Bird's Nest of global optimum is replicated to replace the Bird's Nest that one of them is found,
The remaining Bird's Nest being found changes the position of Bird's Nest at random, so as to preserve optimal Bird's Nest, determines current optimal Bird's Nest
Position and optimal value.
Step S4: after determining Bird's Nest position and optimal value, judge whether the number of iterations reaches maximum number of iterations
(Tmax), it is to export globally optimal solution, it is no to then follow the steps S2.
In conclusion the beam selection method applied to the extensive mimo system of millimeter wave of the present embodiment is with following excellent
Point:
The beam selection method for being applied to the extensive mimo system of millimeter wave sees beam selection as solution { 0-1 } back
Packet problem is simultaneously solved, and rf chain number needed for reducing can be avoided without causing apparent performance loss using total
A large amount of radio frequency links needed for word coding cause energy loss excessively high, and nearly excellent system can be obtained under the premise of reducing complexity
Performance.Due to being easy to appear improper coding in solution procedure, the present invention repairs improper solution, ensure that calculating gained
Solution be feasible solution.Meanwhile the present embodiment replicates the Bird's Nest of global optimum to replace the Bird's Nest being wherein found, and makes optimal bird
Nest preserves, and accelerates convergence, otherwise without the interference considered between user, directly using optimization and rate as mesh
Mark, thus it is best with rate capability compared to what existing several beam selection schemes were realized.In addition, the present embodiment is discrete cuckoo
Bird searching algorithm increases application scenarios.
In the present embodiment, beam selection method sees the wave beam of selection as selected item and is packed into knapsack, by system
Maximum and rate regards the maximum capacity problem that knapsack is filled that solves as, and is obtained using discrete cuckoo searching algorithm is improved
Near-optimum solution, for the improper coding generated in cuckoo algorithm Levy flight discretization results, using heuristic greedy algorithm
It is repaired.Meanwhile the duplication in genetic algorithm is introduced into DCS algorithm by the present embodiment, replicates the Bird's Nest of global optimum to replace
The Bird's Nest being wherein found is changed, accelerates algorithm the convergence speed, and improving by analytical proof beam selection scheme of the invention
It is better than existing beam selection scheme in terms of system summation rate.
Embodiment 2
Referring to Fig. 2, the beam selection method applied to the extensive mimo system of millimeter wave is present embodiments provided,
Step (step S0) is increased on the basis of embodiment 1.Step S0 are as follows: the building extensive mimo system of millimeter wave.?
In the present embodiment, the mono- cell system of the extensive MIMO of millimeter wave is considered, it is assumed that it is N, RF link that base station end, which is equipped with antenna number,
Number is NRF, meet N > NRF.Base station services K single-antenna subscriber simultaneously.In order to without loss of generality, present embodiment assumes that K=NRF。
Therefore, the construction method of the extensive mimo system of millimeter wave includes the following steps (step S01-S04).
Step S01, the expression formula of the reception signal of k-th of user of preliminary definition.In the big rule of the millimeter wave of the present embodiment
In mould mimo system, the expression formula of the reception signal of k-th of user is first is defined as:
Y=HHWs+n
In formula, H is channel matrix, H=[h1,h2,...,hK], hkFor the channel vector between k-th of user and base station；s
For original signal vector, s ∈ CK×1, and normalized power E (ssH)=IK；W is the pre-coding matrix having a size of N × K, and is met
tr(WWH)≤ρ, ρ are total emission power；N is the additive white Gaussian noise having a size of K × 1, and n~CN (0, σ2IK).In this reality
It applies in example, the Saleh-Valenzuela channel model being widely used using millimetre-wave attenuator, the channel vector of user k are as follows:
In formula,For the los path of k-th of user,For the non line of sight road of k-th of user
Diameter；GkFor complex gain, ψkFor dimensional orientation,For array response vector.
Step S02, using discrete lens array (DLA), by spatial Fourier transform matrix by traditional space channel
Beam space channel is converted to, and defines the expression formula for receiving signal of k-th of user again according to beam space channel.It is empty
Between Fourier transform matrix be one group of given orthogonal basis:
In formula,As a result, in the extensive mimo system of millimeter wave, k-th
The expression formula of the reception signal of user is again is defined as:
Wherein,For beam space channel, and calculation formula are as follows:
For the beam space channel vector of k-th of user, and k=1 ..., K.Here each element representation is predefined
The channel gain that wave beam provides.
Step S03 only needs selected part wave beam according to the sparsity of beam space mimo channel, establishes the mesh that antenna is selected
Scalar functions and constraint condition.Due to the sparsity of channel, when calculating transmission rateIn there is only several essential elements, only
A small amount of suitable wave beam need to be chosen can reduce the dimension of mimo system under the premise of not causing significant performance to lose.Therefore, K
The reception signal of a user's downlink may be expressed as:
In formula,For the channel of selected wave beam composition, Wr∈CK×KAnd to have reduced the digital precode matrix of dimension.
Step S04 establishes objective function and constraint condition that antenna is selected.In the premise for not causing significant performance to lose
Under K antenna is picked out from N number of antenna, need to establish it objective function and constraint condition.Due to be analog domain into
Row beam selection may be expressed as: using close-to zero beam (zero-foring, ZF) precoding
In formula, α is zoom factor and satisfactionρ is total emission power.
Assuming that distribute equal power to each user in base station end, then the achievable Mean Speed of k-th of user are as follows:
In formula, σ2For noise power.
Further, it is determined that objective function, are as follows:
xnkThe state value of n-th of antenna is chosen for k-th of user；RkFor the achievable Mean Speed of k-th of user.
Moreover, each user only selects the expression formula of a wave beam are as follows:
Each wave beam is at most selected by a user, it is also possible to not chosen by any user, corresponding expression formula are as follows:
xnkValue range be that 0 or 1, the 0 expression antenna is not selected, 1 indicates that the antenna is selected, specific to express
Formula are as follows:
xnk∈ { 0,1 }, n=1,2 ..., N k=1,2 ..., K
On the basis of the above, beam preference pattern optimization can be carried out, that is, carries out the work of embodiment 1.
In the present embodiment, emulation experiment has been carried out, wherein the specific facilities of emulation experiment parameter are as follows: assuming that the space of user k
Channel has a LoS component and has L=2 NLoS component, wherein WithDistributed area is followed to existBe uniformly distributed.The present embodiment considers two kinds of millis of Sparse System and congestion system
The extensive mimo system of metric wave, wherein Sparse System base station, which is equipped with, has N=256 antenna and K=32 user, congestion system
Base station is equipped with antenna number N=64 and K=32 user, transmission power ρ=32mW, Bird's Nest number M=20, Pa=0.75, Pr=0.5.
First the degree of convergence of algorithm mentioned to the present embodiment is analyzed.Maximum number of iterations T is setmax=500, SNR=
30dB, with Sparse System antenna number N=256, for number of users K=32, referring to Fig. 3, it can be seen that the mentioned base of the present embodiment
Maximum and rate is substantially had been obtained at iteration 100 times or so in DCS beam selection algorithm, in order to reduce calculation amount,
With maximum number of iterations T in subsequent simulationmax=100.
Referring to Fig. 4, the figure is calculated under conditions of Sparse System based on DCS beam selection algorithm and digital precoding
Method, IA algorithm and MM algorithm are compared.It can be with the MM-2 wave beam of IA algorithm as seen from Figure 4 and 2 wave beams of each user
Selection and rate capability it is substantially consistent, it is good more many than the MM-1 beam selection and rate performance of 1 wave beam of each user, this be because
The interference between user is not considered for MM-1 scheme, much causes higher energy to damage although MM-2 is realized and rate improves
Consumption.And rate capability more better than above-mentioned three kinds of algorithms is realized based on DCS beam selection algorithm, also closer to total type families
System.
Referring to Fig. 5, the figure is real based on DCS beam selection algorithm and existing algorithm institute under conditions of congestion system
Now with the comparison of rate.Very maximum probability can have the different identical antennas of user's selection to IA algorithm under congestion system, by dry
The complexity of the more IA algorithms of the number of users disturbed is bigger.Do not have to consider the interference between user based on DCS beam selection algorithm, directly
Using optimization and rate as target, therefore realize best with rate capability.Fig. 6 be under conditions of SNR=30dB, N=100 not
With the comparison of number of users and rate, optimum performance can be obtained based on DCS beam selection algorithm as seen from the figure.
Embodiment 3
Present embodiments provide a kind of beam selection device applied to the extensive mimo system of millimeter wave, application implementation
The beam selection method applied to the extensive mimo system of millimeter wave of example 1 or embodiment 2.Wherein, the beam selection device packet
Include fitness computing module, Bird's Nest screening module, Bird's Nest position replacement module and the number of iterations judgment module.
Fitness computing module is general for the first Bird's Nest number of one extensive mimo system of millimeter wave of definition, Bird's Nest discovery
Rate, binary coding control parameter, maximum number of iterations, antenna number and number of users, reinitialize multiple Bird's Nests, makes each bird
Nest selects channel magnitude maximum and without duplicate wave beam, calculates the fitness of multiple Bird's Nests, and finally with the maximum of current Bird's Nest
Fitness is globally optimal solution；Wherein, the calculation formula of the fitness of multiple Bird's Nests are as follows:
In formula, xnkThe state value of n-th of antenna is chosen for k-th of user；RkFor the achievable average speed of k-th of user
Rate；Wherein, the calculation formula of Mean Speed are as follows:
In formula, σ2For noise power；α is zoom factor and satisfactionρ is total emission power, and K is to use
Amount；H is channel matrix, hkFor the channel vector between k-th of user and base station；For the channel of selected wave beam composition；Wr∈CK×KAnd to have reduced the digital precode matrix of dimension.
Bird's Nest screening module be used for after the fitness that fitness computing module calculates multiple Bird's Nests, first to multiple Bird's Nests into
The mixing of row binary coding updates, and repairs improper coding, then calculate the fitness of newly generated multiple Bird's Nests, finally by
Retain the biggish Bird's Nest of fitness to screen Bird's Nest.
Bird's Nest position replacement module after screening all Bird's Nests, by Bird's Nest probability of detection and is obeyed for Bird's Nest screening module
Equally distributed random number is compared；If random number is greater than Bird's Nest probability of detection, Bird's Nest position replacement module replicates the overall situation
The Bird's Nest of optimal solution changes the position for the Bird's Nest that remaining is found to replace wherein be found Bird's Nest at random, with true
The Bird's Nest position of settled preceding optimal solution and optimal value.
The number of iterations judgment module is used for the judgement in Bird's Nest position replacement module after determining Bird's Nest position and optimal value
Whether the number of iterations reaches maximum number of iterations, is, exports globally optimal solution, otherwise drives the work of Bird's Nest screening module.
Embodiment 4
Present embodiments provide a kind of terminal comprising memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor.What processor realized embodiment 1 or embodiment 2 when executing program is applied to millimeter
The step of beam selection method of the extensive mimo system of wave.
The beam selection method applied to the extensive mimo system of millimeter wave of embodiment 1 or embodiment 2 is in use, can
It is applied in the form of software, is such as designed to independently operated program, on computer terminals, terminal can be with for installation
It is computer, smart phone, control system and other internet of things equipment etc..Embodiment 1 or embodiment 2 to be applied to millimeter wave big
The beam selection method of scale mimo system can also be designed to the program of embedded operation, install on computer terminals, such as
It is mounted on single-chip microcontroller.
Embodiment 5
A kind of computer readable storage medium is present embodiments provided, computer program is stored thereon with.Program is processed
When device executes, the step of the beam selection method applied to the extensive mimo system of millimeter wave of embodiment 1 or embodiment 2 is realized
Suddenly.
The beam selection method applied to the extensive mimo system of millimeter wave of embodiment 1 or embodiment 2 is in use, can
Applied in the form of software, be such as designed to computer readable storage medium can independently operated program, it is computer-readable to deposit
Storage media can be USB flash disk, be designed to U-shield, be designed to start the program of entire method by external triggering by USB flash disk.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of beam selection method applied to the extensive mimo system of millimeter wave, which is characterized in that itself the following steps are included:
Step S1 first defines Bird's Nest number, Bird's Nest probability of detection, binary coding the control ginseng of the extensive mimo system of millimeter wave
Number, maximum number of iterations, antenna number and number of users, reinitialize multiple Bird's Nests, and each Bird's Nest user is made to select channel width
Angle value is maximum and without duplicate wave beam, finally calculates the fitness of multiple Bird's Nests, and be complete with the maximum adaptation degree of current Bird's Nest
Office's optimal solution；
Step S2 first carries out binary coding mixing to multiple Bird's Nests and updates, and repair after the fitness for calculating multiple Bird's Nests
Improper coding, then the fitness of newly generated multiple Bird's Nests is calculated, finally by the reservation biggish Bird's Nest of fitness to screen
Bird's Nest；
Step S3 compares the Bird's Nest probability of detection with equally distributed random number is obeyed after all Bird's Nests screen
Compared with if the random number replicates the Bird's Nest of the globally optimal solution and is wherein sent out with replacing greater than the Bird's Nest probability of detection
An existing Bird's Nest, and change the position for the Bird's Nest that remaining is found at random, retain globally optimal solution to accelerate algorithmic statement speed
Degree, determines Bird's Nest position and the optimal value of current optimal solution；
Step S4 judges whether the number of iterations reaches the maximum number of iterations after determining Bird's Nest position and optimal value, is
The globally optimal solution is then exported, it is no to then follow the steps S2.
2. being applied to the beam selection method of the extensive mimo system of millimeter wave as described in claim 1, which is characterized in that
In step S2, binary coding mix update method the following steps are included:
Step S21, judges whether the random number of system is not more than the binary coding control parameter；
When the random number is not more than the binary coding control parameter, step S22 is executed, it will using sigmoid function
Real number is mapped to discrete binary data, is calculated by the following formula the shape of k-th of the m+1 times iteration, n-th of user antenna
State value
When the random number is greater than the binary coding control parameter, step S23 is executed, whether judges Levy hop path
For positive number；
When Levy hop path is not positive number, step S24 is executed, the m+1 times iteration, k-th of user is calculated by the following formula
The state value of n-th of antenna
When hop path is positive number, step S25 is executed, k-th of the m+1 times iteration, n-th of user is calculated by the following formula
The state value of antenna
Wherein, rand is the random number.
3. being applied to the beam selection method of the extensive mimo system of millimeter wave as claimed in claim 2, which is characterized in that institute
State the restorative procedure of improper coding the following steps are included:
Step S26 successively calculates the state value x of k-th of m-th of Bird's Nest, n-th of user antennank, that is, calculate m-th of matrix
The state value of kth column line n, and judge state value xnkIt whether is 1；
In state value xnkWhen being 1, step S27 is executed, corresponding channel gain g is calculatednk；
All state value x in each column for judging the binary coded matrixnkAfterwards, step S28 is executed, following formula is passed through
Select the wave beam and beam collection of user:
Wherein, βkFor the wave beam of k-th of user,The beam collection chosen for user.
4. being applied to the beam selection method of the extensive mimo system of millimeter wave as described in claim 1, which is characterized in that institute
It is further comprising the steps of to state beam selection method:
Step S0 constructs the extensive mimo system of the millimeter wave；Wherein, the building side of the extensive mimo system of the millimeter wave
Method the following steps are included:
Step S01, the expression formula of the reception signal of k-th of user of preliminary definition；
Traditional Space channel is converted to wave beam sky by spatial Fourier transform matrix using discrete lens array by step S02
Between channel, and according to the beam space channel define again k-th of user reception signal expression formula；
Step S03 only needs selected part wave beam according to the sparsity of beam space mimo channel, establishes the target letter that antenna is selected
Several and constraint condition.
5. being applied to the beam selection method of the extensive mimo system of millimeter wave as claimed in claim 4, which is characterized in that
In the extensive mimo system of the millimeter wave of the present embodiment, the expression formula of the reception signal of k-th of user is first is defined as:
Y=HHWs+n
In formula, H is channel matrix, H=[h1,h2,...,hK], hkFor the channel vector between k-th of user and base station；S is original
Beginning signal phasor, s ∈ CK×1, and normalized power E (ssH)=IK；W is the pre-coding matrix having a size of N × K, and meets tr
(WWH)≤ρ, ρ are total emission power；N is the additive white Gaussian noise having a size of K × 1, and n~CN (0, σ2IK)。
6. being applied to the beam selection method of the extensive mimo system of millimeter wave as claimed in claim 5, which is characterized in that institute
State the calculation formula of channel vector are as follows:
In formula,For the los path of k-th of user,For the obstructed path of k-th of user；GkFor
Complex gain, ψkFor dimensional orientation,For array response vector.
7. being applied to the beam selection method of the extensive mimo system of millimeter wave as claimed in claim 6, which is characterized in that institute
State spatial Fourier transform matrix are as follows:
In formula,
As a result, in the extensive mimo system of millimeter wave, the expression formula of the reception signal of k-th of user is again is defined as:
Wherein,For beam space channel, and calculation formula are as follows:
For the beam space channel vector of k-th of user, and k=1 ..., K.
8. the use as claimed in claim 7 in the beam selection method of the extensive mimo system of millimeter wave, which is characterized in that institute
State the expression formula of downlink received signal are as follows:
In formula,For the channel of selected wave beam composition, Wr∈CK×KAnd to have reduced the digital precode matrix of dimension.
9. the use as claimed in claim 7 in the beam selection method of the extensive mimo system of millimeter wave, which is characterized in that institute
State objective function are as follows:
In formula, xnkFor the state value of k-th of user, n-th of antenna；RkFor the achievable Mean Speed of k-th of user；Wherein,
The calculation formula of the Mean Speed are as follows:
In formula, σ2For noise power,And Wr∈CK×K；α is zoom factor and satisfaction
ρ is total emission power.
10. a kind of beam selection device applied to the extensive mimo system of millimeter wave, using any in such as claim 1-9
It is applied to the beam selection method of the extensive mimo system of millimeter wave described in one, characterized in that it comprises:
Fitness computing module, be used to first define the Bird's Nest number of an extensive mimo system of millimeter wave, Bird's Nest probability of detection,
Binary coding control parameter, maximum number of iterations, antenna number and number of users, reinitialize multiple Bird's Nests, makes every in Bird's Nest
A user selects channel magnitude maximum and without duplicate wave beam, calculates the fitness of multiple Bird's Nests, and finally with current Bird's Nest
Maximum adaptation degree is globally optimal solution；Wherein, the calculation formula of the fitness of multiple Bird's Nests are as follows:
In formula, xnkFor the state value of k-th of user, n-th of antenna in m-th of Bird's Nest；RkIt is put down for k-th of the achievable of user
Equal rate；Wherein, the calculation formula of the Mean Speed are as follows:
In formula, σ2For noise power, α is zoom factor and satisfactionρ is total emission power, and K is user
Number；H is channel matrix, hkFor the channel vector between k-th of user and base station；For the channel of selected wave beam composition；Wr∈CK×KAnd to have reduced the digital precode matrix of dimension；
Bird's Nest screening module is used for after the fitness computing module calculates the fitness of multiple Bird's Nests, first to multiple birds
Nest carries out binary coding mixing and updates, and repairs improper coding, then calculate the fitness of newly generated multiple Bird's Nests, finally
By retaining the biggish Bird's Nest of fitness to screen Bird's Nest；
Bird's Nest position replacement module is used for the Bird's Nest screening module after screening all Bird's Nests, the Bird's Nest is found general
Rate is compared with equally distributed random number is obeyed；If the random number is greater than the Bird's Nest probability of detection, the Bird's Nest position
It sets replacement module and replicates the Bird's Nest of the globally optimal solution then to replace wherein be found Bird's Nest, and change remaining at random
The position for the Bird's Nest being found, with the Bird's Nest position of the current optimal solution of determination and optimal value；And the number of iterations judgment module,
It is used in Bird's Nest position replacement module after determining Bird's Nest position and optimal value, judges whether the number of iterations reaches institute
Maximum number of iterations is stated, is, exports the globally optimal solution, the Bird's Nest screening module is otherwise driven to work.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910705982.7A CN110311719B (en) | 2019-07-31 | 2019-07-31 | Beam selection method and device applied to millimeter wave large-scale MIMO system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910705982.7A CN110311719B (en) | 2019-07-31 | 2019-07-31 | Beam selection method and device applied to millimeter wave large-scale MIMO system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110311719A true CN110311719A (en) | 2019-10-08 |
| CN110311719B CN110311719B (en) | 2022-03-29 |
Family
ID=68082847
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910705982.7A Active CN110311719B (en) | 2019-07-31 | 2019-07-31 | Beam selection method and device applied to millimeter wave large-scale MIMO system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110311719B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112261728A (en) * | 2020-12-22 | 2021-01-22 | 之江实验室 | Beam selection matrix design method based on lens array |
| CN113114319A (en) * | 2021-04-20 | 2021-07-13 | 安徽大学 | Joint optimization method based on beam selection and interference elimination and application thereof |
| CN113824478A (en) * | 2021-10-11 | 2021-12-21 | 北京邮电大学 | Broadband millimeter wave multi-user large-scale MIMO uplink spectrum efficiency optimization method assisted by discrete lens antenna array |
| CN114024589A (en) * | 2021-11-11 | 2022-02-08 | 深圳大学 | MISO communication system design method and device |
| CN115102586A (en) * | 2022-06-18 | 2022-09-23 | 河南工业大学 | Wave beam selection method and device based on cuckoo search and ant colony optimization |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110279321A1 (en) * | 2010-03-11 | 2011-11-17 | Nec Laboratories America, Inc. | Wireless Multicasting with Beamforming Antennas |
| CN109299570A (en) * | 2018-10-25 | 2019-02-01 | 中国人民解放军陆军工程大学 | Array aerial direction figure self-repair method |
-
2019
- 2019-07-31 CN CN201910705982.7A patent/CN110311719B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110279321A1 (en) * | 2010-03-11 | 2011-11-17 | Nec Laboratories America, Inc. | Wireless Multicasting with Beamforming Antennas |
| CN109299570A (en) * | 2018-10-25 | 2019-02-01 | 中国人民解放军陆军工程大学 | Array aerial direction figure self-repair method |
Non-Patent Citations (2)
| Title |
|---|
| GENG SUN等: "Thinning of Concentric Circular Antenna Arrays Using Improved Discrete Cuckoo Search Algorithm", 《2017 IEEE WIRELESS COMMUNICATIONS AND NETWORKING CONFERENCE (WCNC)》 * |
| 冯登科等: "二进制布谷鸟搜索算法", 《计算机应用》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112261728A (en) * | 2020-12-22 | 2021-01-22 | 之江实验室 | Beam selection matrix design method based on lens array |
| CN113114319A (en) * | 2021-04-20 | 2021-07-13 | 安徽大学 | Joint optimization method based on beam selection and interference elimination and application thereof |
| CN113114319B (en) * | 2021-04-20 | 2022-02-15 | 安徽大学 | Joint optimization method based on beam selection and interference elimination and application thereof |
| CN113824478A (en) * | 2021-10-11 | 2021-12-21 | 北京邮电大学 | Broadband millimeter wave multi-user large-scale MIMO uplink spectrum efficiency optimization method assisted by discrete lens antenna array |
| CN113824478B (en) * | 2021-10-11 | 2023-07-04 | 北京邮电大学 | Broadband millimeter wave multi-user large-scale MIMO uplink spectrum efficiency optimization method assisted by discrete lens antenna array |
| CN114024589A (en) * | 2021-11-11 | 2022-02-08 | 深圳大学 | MISO communication system design method and device |
| CN114024589B (en) * | 2021-11-11 | 2024-02-06 | 深圳大学 | MISO communication system design method and device |
| CN115102586A (en) * | 2022-06-18 | 2022-09-23 | 河南工业大学 | Wave beam selection method and device based on cuckoo search and ant colony optimization |
| CN115102586B (en) * | 2022-06-18 | 2024-04-05 | 河南工业大学 | Beam selection method and device based on cuckoo search and ant colony optimization |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110311719B (en) | 2022-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110311719A (en) | A kind of beam selection method and its device applied to the extensive mimo system of millimeter wave | |
| Maksymyuk et al. | Deep learning based massive MIMO beamforming for 5G mobile network | |
| KR102277466B1 (en) | Methods for linear rf beam search in millimeter wave communication system with hybrid beam-forming | |
| CN107135024A (en) | A kind of mixed-beam figuration Iterative Design method of low complex degree | |
| CN109104225A (en) | A kind of optimal extensive MIMO Beam Domain multicast transmission method of efficiency | |
| Yang et al. | Machine learning enabling analog beam selection for concurrent transmissions in millimeter-wave V2V communications | |
| CN108449118B (en) | Hybrid precoding method and device in large-scale MIMO system | |
| US11025317B2 (en) | Adaptation of hierarchical codebooks | |
| CN110233653A (en) | Blind multipath recognition methods and system based on the mimo system for weighting integrated clustering algorithm | |
| WO2022174418A1 (en) | Wireless telecommunications network | |
| CN105991167B (en) | The millimeter wave MIMO analog beam shaping method of low complex degree | |
| CN112367105B (en) | Low-overhead topological interference alignment and power optimization method, system, equipment and application | |
| CN112564754B (en) | Wave beam selection method based on self-adaptive cross entropy under millimeter wave Massive MIMO system | |
| CN113261016A (en) | Single-shot multi-user multiple-input multiple-output (MU-MIMO) resource pairing using Deep Q Network (DQN) based reinforcement learning | |
| Chen et al. | 5G air-to-ground network design and optimization: A deep learning approach | |
| CN113315552A (en) | Energy efficiency optimization method of compact planar antenna array Massive MIMO system | |
| CN102104406B (en) | Method and system for switching beam forming and MIMO (Multi-Input Multi-Output) beam forming | |
| US11881916B2 (en) | Precoding method and apparatus | |
| CN115133969A (en) | Performance improving method of millimeter wave large-scale MIMO-NOMA system | |
| CN112242860B (en) | Beam forming method and device for self-adaptive antenna grouping and large-scale MIMO system | |
| CN103765805B (en) | A kind of method for multi-user pre-coding and device | |
| Akbarpour-Kasgari et al. | Deep Reinforcement Learning in mmW-NOMA: Joint Power Allocation and Hybrid Beamforming | |
| Bouchibane et al. | Upgraded-abc algorithm for antenna selection in energy efficient massive MIMO system | |
| Chen et al. | Planning optimization of the distributed antenna system in high-speed railway communication network based on improved cuckoo search | |
| Patil et al. | Hybrid meta-heuristic algorithms based optimal antenna selection for large scale MIMO in LTE network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |