CN106850028A - A kind of united beam forming method based on SWIPT systems - Google Patents

Abstract

In order to solve the problems, such as that the rate of information throughput and collection of energy speed are slower in SWIPT systems, the present invention provides a kind of united beam forming method based on SWIPT systems, and belonging to wireless taking can the communications field.Comprise the following steps：Step one：Transmitting terminal sends separate information Wave beam forming vector sum energy velocity of wave and forms vector simultaneously；Step 2：Receiving terminal enters row information and receives and collection of energy, including information receives time slot and energy acceptance time slot：Time slot is received in information, using information Wave beam forming vector information resources are carried out with the alignment of specific user, realize the error free transport of information；In energy acceptance time slot, forming vector using energy beam carries out the alignment of specific user to energy resource, by inter-user interference as collection of energy source.The present invention realizes the independent control of information transmission and energy transmission, increases the energy that collection of energy termination is received, while reasonably using the interference between user, to obtain more excellent R E curves.

Description

A kind of united beam forming method based on SWIPT systems

Technical field

The energy communications field is taken the invention belongs to wireless, and in particular to a kind of united beam side of being formed based on SWIPT systems Method.

Background technology

With the development of Scientific Strategy, people are devoted to finding a kind of new energy always, rather than depending on simply The natural resources such as wind energy, solar energy.In real application systems, for some small-sized devices, such as sensor node, intelligence Mobile phone, artificial intelligence heart etc., it is difficult to design energy collecting circuit utilizes the solar energy and wind energy to charge it, to keep Permanent steady operation.In order to overcome this practical problem, people start to seek new energy source, i.e. radio frequency (Radio Frequency, RF) signal.In traditional wireless communication system, radiofrequency signal role is transmission information, But it now appear that, radiofrequency signal can not only carry information, it is also possible to carry energy.Therefore, receiving terminal can be radiofrequency signal Get up as collection of energy, originated as the energy of receiving terminal, to extend the life-span of communication network.So, receiving terminal is not Only can be with demodulating information, it is also possible to which energy is collected by the design of receiving circuit simultaneously.

SWIPT (wireless portable communications system, Simultaneous Wireless Information and Power Transfer concept) was proposed that he points out SWIPT strategies not only in response to green by L.R.Varshney first in 2008 The call of color communication, and the service life of energy constraint cordless communication network can be extended, then he is also based on SISO Information is passed during (Simple Input Simple Output, single-input single-output) flat fading channel have studied SWIPT systems The compromise problem of defeated speed and energy transmission rate, i.e., usually said R-E curves, curve can clearly express SWIPT Information and the utilization power of energy two spike-type cultivars, if the utilization rate of information resources is lifted, are embodied in information rate increasing in system Greatly, if the utilization rate lifting of energy resource, is embodied in the increase of electricity collection amount.Further, since the power dissipation of free space And the conversion efficiency of power conversion circuits is relatively low, cause efficiency of energy collection not high, and the presence at collection of energy end influences The rate of information throughput of communication system.Therefore, both at home and abroad communication scholar for electricity collection amount is not enough and how is R-E curves Accomplishing the problem of optimal compromise has carried out extensive concern and research.

Wave beam forming (Beamforming, BF) strategy serves positive effect to optimization R-E, and the strategy can increase day The signal intensity of linear array column direction output, reduces the intensity of interference signal, so as to improve the reachable information rate of communication network.Ripple Beam is formed in terms of research of the strategy in SWIPT systems is divided into following four, the Wave beam forming strategy respectively under different channels Research, the Wave beam forming strategy study under different receivers strategy, the Wave beam forming strategy study under Different Optimization target and Intelligence receiver (Information Receiver, IR) and energy receiver (Energy Receiver, ER) do not hold situation altogether Under united beam formed strategy research.

But the above-mentioned research to Wave beam forming strategy is more single, it is limited only to system model, receives tactful or excellent Change the simple change of target.The change of system model, reception strategy or optimization aim can not be denied can obtain more excellent R-E songs The fact that line, but do not have to the research of Wave beam forming strategy and propose some innovative points with reference to the characteristics of SWIPT systems itself. Such as, existence information and energy two spike-type cultivars in SWIPT systems, and traditional wireless communication system only exists information resources one Composition is planted, and is still indiscriminately imitated from legacy wireless communication system for application of the Wave beam forming strategy in SWIPT systems, do not considered The otherness of information resources and energy resource.Therefore, can not be optimal using the R-E curves of traditional Wave beam forming strategy 's.

In traditional wireless communication system, radiofrequency signal role is transmission information, due to radiofrequency signal in itself Energy can also be carried, so can both carry out information transfer using radiofrequency signal as the wireless communication system of medium, it is also possible to Electric energy transmission is carried out, the communication system with this dual-use function is called wireless portable communications system, i.e. SWIPT systems.

In multiple antennas SWIPT systems, traditional Wave beam forming strategy has been obtained for being widely applied, and the strategy can increase The intensity of big aerial array direction useful signal, reduces the intensity of interference signal, to realize the directional transmissions of information and energy.But Traditional Wave beam forming strategy only with a Wave beam forming vector, not in view of informational content in SWIPT systems and energy into Point otherness, and have ignored inter-user interference can also as energy the fact, as a result cause the resource in system not obtain Rational to utilize, the energy electricity that collection of energy termination is received is very few so that information-speed (Rate-Energy, R-E) curve does not have Have and reach good effect.

The content of the invention

The invention aims to solve in SWIPT systems the rate of information throughput and collection of energy speed is slower asks Topic, the present invention provides a kind of united beam forming method based on SWIPT systems.

A kind of united beam forming method based on SWIPT systems of the invention, methods described comprises the following steps：

Step one：Transmitting terminal sends separate information Wave beam forming vector sum energy velocity of wave and forms vector simultaneously；

Step 2：Receiving terminal enters row information and receives and collection of energy, including information receives time slot and energy acceptance time slot：

Time slot is received in information, using information Wave beam forming vector information resources are carried out with the alignment of specific user, realized The error free transport of information；

In energy acceptance time slot, forming vector using energy beam carries out the alignment of specific user to energy resource, will use The source as collection of energy is disturbed between family.

Preferably, the step 2 includes：

Time slot and energy acceptance time slot are received according to information, object function is set up：

The bound for objective function：

Wherein, v_kIt is the information Wave beam forming vector of k-th user of transmitting terminal；

M is the quantity of user；

w_kFor the energy beam of k-th user of transmitting terminal forms vector；

α is time slot dividing ratio；W is transmission channel bandwidth；It is h_kConjugate transposition, h_kFor k-th channel of user increases Beneficial matrix；

It is k-th user antenna noise power；ζ_kIt is k-th energy conversion efficiency of user；γ_kWhen being received for information The minimum Signal to Interference plus Noise Ratio thresholding of k-th user of gap；e_kMinimum for k-th user of energy acceptance time slot collects energy threshold；

Optimal solution is sought to setting up parameter in object function；

Enter row information using the object function for trying to achieve optimal solution to receive and collection of energy.

Preferably, the Model Parameter of described pair of foundation asks the optimal solution to be：

Optimal solution is asked to parameter in object function using Lagrange relaxation method.

Preferably, the utilization Lagrange relaxation method asks parameter in object function the method for optimal solution to be：

The object function of foundation is converted to using Lagrange relaxation method：

And

L_k() is Lagrangian, and λ is Lagrange multiplier；

Bound for objective function after conversion is：

Optimal value is asked for object function after conversion.

Preferably, information Wave beam forming vector v_kOptimal valueFor：

Wherein,Φ_k Expression information Wave beam forming vector v_kDomain of definition；N_tRepresent the quantity of transmitting terminal transmission antenna.

Preferably, energy beam forms vectorial w_kOptimal valueFor：

According to L=A^-1Λ, obtains

ForOrthonormal basis；

N_tRepresent the quantity of transmitting terminal transmission antenna.

Preferably, the optimal value of the α

The initial value of α is α_start, the final value of α is α_end, using formulaValue to α is updated, when full FootWhen α values, be exactly the optimal value of the α

Preferably, the optimal value of the λ

The upper bound of Lagrange multiplier is λ_max, the lower bound of Lagrange multiplier is λ_min, using formula Value to λ is updated, when meeting P_m-α|v_k|²-|w_k|²+α|w_k|²λ value when=0, is exactly the optimal value of the λWhereinP is the maximum transmit power thresholding of transmitting terminal.

Above-mentioned technical characteristic can in any suitable manner be combined or substituted by equivalent technical characteristic, as long as can reach To the purpose of the present invention.

The beneficial effects of the present invention are in order to realize the independent control of information transmission and energy transmission, increase energy is received The energy that collection termination is received, while reasonably using the interference between user, to obtain more excellent R-E curves, the present invention is proposed Single Wave beam forming vector is extended to two independences by a kind of united beam forming method based on SWIPT systems, the method Wave beam forming vector.The stage is received in information, the right of specific user is carried out to information resources using information Wave beam forming vector Standard, now in order to eliminate user's interference, realizes the error free transport of information, exists between information Wave beam forming vector and channel gain Orthogonality relation；In the energy acceptance stage, forming vector using energy beam carries out the alignment of specific user to energy resource, this When in view of inter-user interference can as the source of collection of energy, and for energy acceptance in the absence of interference factor, So energy beam is formed need not meet orthogonality relation between vector and channel gain.The present invention has obtained one by mathematical modeling The individual most value Solve problems containing inequality constraints, can be solved using method of Lagrange multipliers to it.Last emulation experiment Show that strategy proposed by the present invention can optimize R-E curves.

Brief description of the drawings

Fig. 1 is the schematic diagram of the SWIPT system models of foundation in specific embodiment.

Fig. 2 is the schematic diagram of conventional beams forming method and TS reception patterns in specific embodiment.

Fig. 3 is the schematic diagram of united beam forming method of the invention and TS reception patterns in specific embodiment.

Fig. 4 be specific embodiment in respectively to use conventional beams forming method and the united beam side of being formed of the invention Method is emulated, the R-E curves of acquisition.

Specific embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on Embodiment in the present invention, those of ordinary skill in the art obtained on the premise of creative work is not made it is all its His embodiment, belongs to the scope of protection of the invention.

It should be noted that in the case where not conflicting, the embodiment in the present invention and the feature in embodiment can phases Mutually combination.

The invention will be further described with specific embodiment below in conjunction with the accompanying drawings, but not as limiting to the invention.

A kind of united beam forming method based on SWIPT systems described in present embodiment, methods described includes as follows Step：

Step one：Transmitting terminal sends separate information Wave beam forming vector sum energy velocity of wave and forms vector simultaneously；

Step 2：Receiving terminal enters row information and receives and collection of energy, including information receives time slot and energy acceptance time slot：

Time slot is received in information, using information Wave beam forming vector information resources are carried out with the alignment of specific user, realized The error free transport of information；

In energy acceptance time slot, forming vector using energy beam carries out the alignment of specific user to energy resource, will use The source as collection of energy is disturbed between family.

Time slot is received in information, using information Wave beam forming vector information resources is carried out with the alignment of specific user, now In order to eliminate user's interference, realize the error free transport of information, information Wave beam forming vector with existed between channel gain it is orthogonal Relation；In energy acceptance time slot, forming vector using energy beam carries out the alignment of specific user to energy resource, now considers As the source of collection of energy, and can not exist the factor disturbed for energy acceptance to inter-user interference, so energy Need not meet orthogonality relation between amount Wave beam forming vector and channel gain.In the case where same information rate can be reached, The electricity collection amount of lifting system, so as to obtain optimal R-E curves, solves information biography in SWIPT systems to a certain extent Defeated speed and the slower problem of collection of energy speed.

In preferred embodiment, step 2 includes：

Step 2 one：Time slot and energy acceptance time slot are received according to information, object function is set up：

The bound for objective function：

Wherein, v_kIt is the information Wave beam forming vector of k-th user of transmitting terminal；

M is the quantity of user；

w_kFor the energy beam of k-th user of transmitting terminal forms vector；

α is time slot dividing ratio；W is transmission channel bandwidth；It is h_kConjugate transposition, h_kIt is k-th channel of user Gain matrix；

It is k-th user antenna noise power；ζ_kIt is k-th energy conversion efficiency of user；γ_kWhen being received for information The minimum Signal to Interference plus Noise Ratio thresholding of k-th user of gap；e_kMinimum for k-th user of energy acceptance time slot collects energy threshold；

Step 2 two：Optimal solution is sought to setting up parameter in object function；

Step 2 three：Enter row information using the object function for trying to achieve optimal solution to receive and collection of energy.

Present embodiment with system and speed as optimization aim, with the minimum Signal to Interference plus Noise Ratio and energy acceptance of intelligence receiver The minimum of machine collects energy for restrictive condition to realize sending Wave beam forming vector and the combined optimization for receiving TS ratios, foundation Object function.In addition to saving the communication resource, the transmit power to base station is also required to do certain limitation.

In preferred embodiment, in step 2 two, the Model Parameter to setting up asks the method for optimal solution to be：

Optimal solution is asked to parameter in object function using Lagrange relaxation method.

The bound for objective function of above-mentioned foundation is more, in order to seek the optimized parameter in object function, will constrain bar Part is absorbed in object function, reduces some constraints so that the solution difficulty of problem is substantially reduced.

In preferred embodiment, the method for optimal solution is asked parameter in object function to be using Lagrange relaxation method：

The object function of foundation is converted to using Lagrange relaxation method：

And

L_k() is Lagrangian, and λ is Lagrange multiplier；

Bound for objective function after conversion is：

Optimal value is asked for object function after conversion.

Present embodiment show in particular object function and constraints after simplifying using Lagrange relaxation method.

In preferred embodiment, information Wave beam forming vector v_kOptimal valueFor：

Wherein,Φ_k Expression information Wave beam forming vector v_kDomain of definition；N_tRepresent the quantity of transmitting terminal transmission antenna.

In preferred embodiment, energy beam forms vectorial w_kOptimal valueFor：

According to L=A^-1Λ, obtains

ForOrthonormal basis；

N_tRepresent the quantity of transmitting terminal transmission antenna.

Preferably, the optimal value of α

The initial value of α is α_start, the final value of α is α_end, using formulaValue to α is updated, when full FootWhen α values, be exactly the optimal value of the α

In preferred embodiment, the optimal value of λ

The upper bound of Lagrange multiplier is λ_max, the lower bound of Lagrange multiplier is λ_min, using formula Value to λ is updated, when meeting P_m-α|v_k|²-|w_k|²+α|w_k|²λ value when=0, is exactly the optimal value of the λWhereinP is the maximum transmit power thresholding of transmitting terminal.

Present embodiment is following to be given using the united beam forming method based on SWIPT systems of the invention and general respectively The embodiment of logical Beamforming Method, and give performance comparison simulated effect：

First, SWIPT system models are set up：

The SWIPT systems are made up of a base station and M user.Base station end is furnished with N_tRoot antenna, user terminal is respectively furnished with 1 Antenna.Only consider the situation of downlink information transmission and collection of energy, the schematic diagram of SWIPT system models is as shown in Figure 1.

Transmitting terminal is respectively adopted conventional beams forming method and united beam forming method of the invention is entered to transmission information Row precoding processing, in order to contrasted, embodies advantage part of the invention.Receiving terminal uses TS patterns, that is, receive Either hold the signal for receiving to be collected by collection of energy end, or demodulating information is used for by information demodulating end.Now, the present embodiment pair One normalization time quantum carries out time slot division, is divided into 2 time slots：Time slot 1 and time slot 2, in the time slot 1 and time slot of receiving terminal 2 are respectively information receives time slot and energy acceptance time slot.Time slot is received in information, user enters row information demodulation；In energy acceptance Time slot, user carries out collection of energy.

Conventional beams forming method is with TS reception patterns schematic diagram as shown in Fig. 2 united beam forming method of the invention It is as shown in Figure 3 with TS reception pattern schematic diagrams.

Knowable to Fig. 2 and Fig. 3, the transmitting terminal of conventional beams forming method, only transmission information Wave beam forming is vectorial, and this hair Bright united beam forming method needs transmission information Wave beam forming vector sum energy beam to form vector.

2nd, receiving terminal receives the mathematical problem description that time slot and energy acceptance time slot are formed in information：

In receiving terminal, receive time slot user and enter row information in information and receive, conventional beams forming method and of the invention Base station end is by the information signal sent after Wave beam forming vector weighting in closing Beamforming Method：

Wherein, M is user's number；v_kIt is N_t× 1 Wave beam forming vector,s_kIt is k-th use The information at family, obeys circle symmetric complex distribution, and average is 0, and variance is 1.

The information signal that k-th user receives is：

Wherein, h_kIt is k-th user N_t× 1 channel gain matrix；The conjugate transposition of superscript H representing matrixs；n_kIt is The k antenna noise of user's receiving terminal, obeys circle symmetric complex distribution, and average is zero, and variance is

In order to suppress the interference between user, the error free transport of information is realized, it is possible to use Wave beam forming vector sum channel Relation between gain matrix, i.e.,：

Then, in the case where no user disturbs, the information signal that k-th user receives is：

The Signal to Interference plus Noise Ratio of information receiving end is represented by：

Wherein,It is the noise power of antenna.

Can be obtained according to Shannon's theorems, the maximum average information rate for receiving signal is：

Wherein, W is transmission channel bandwidth；α is the time slot dividing ratio into row information demodulation.

In sum, the maximum of all K users is average and information rate is in SWIPT systems：

The base of collection of energy, conventional beams forming method and united beam forming method is carried out in energy acceptance time slot user End energy signal of transmission after Wave beam forming vector is weighted of standing is respectively：

Wherein, M is user's number；v_kIt is N_t× 1 Wave beam forming vector,w_kIt is N_t× 1 Wave beam forming vector,s_kIt is k-th information of user, obeys circle symmetric complex distribution, average It is 0, variance is 1.

The energy signal that k-th user of conventional beams forming method and united beam forming method receives is respectively：

Wherein, h_kIt is k-th user N_t× 1 channel gain matrix；The conjugate transposition of superscript H representing matrixs；n_kIt is The k antenna noise of user's receiving terminal, obeys circle symmetric complex distribution, and average is zero, and variance is

Conventional beams forming method due to being controlled to information and energy using same Wave beam forming vector simultaneously, institute Equally abolished with the interference between second stage user, and united beam forming method then can be between user Interference is collected as energy, therefore conventional beams forming method and final k-th user of united beam forming method receive Energy signal be respectively：

Due to being the relation of quadratic sum between signal and power, so, conventional beams forming method and united beam are formed K-th user of method is respectively in the energy that energy acceptance time slot is collected into

Wherein,It is the noise power of antenna；1- α are the time for carrying out collection of energy；ζ_kIt is energy conversion efficiency.

Within the unit interval, total transmission work(of the SWIPT systems of conventional beams forming method and united beam forming method Rate is respectively：

For beam-forming technology in SWIPT systematic researches, the purpose of present embodiment is by information and energy Directive sending come realize system and speed and collect electricity increase.Because SWIPT systems can simultaneously realize radio communication With wireless charging, it is necessary to objectively weigh the importance of the two to determine final optimization aim.In SWIPT systems, no matter How much is the electricity of collection, is all finally for information transmission is serviced.So when intelligence receiver and energy receiver can be normal Under conditions of work, system and speed are bigger, and the performance of SWIPT systems is more excellent.Therefore, with system and speed as optimization aim, Realize sending wave beam with the minimum collection energy of the minimum Signal to Interference plus Noise Ratio of intelligence receiver and energy receiver as restrictive condition Form vector and receive the combined optimization of TS ratios.In addition to saving the communication resource, the transmit power to base station is also required to do Certain limitation.Therefore it is respectively using the problem description of conventional beams forming method and united beam forming method of the invention Formula (18) and formula (19)：

3rd, the mathematical problem for being formed is described to solve：

(1) mathematical problem of conventional beams forming method is solved

Observation formula (18), the Wave beam forming vector for finding each user in base-station transmitting-power constraint is to be coupled , in order to be M simple subproblem a former problem reduction, first with Lagrangian relaxation technology the-d of formula (18)) inhale Receive in object function (a), a generalized lagrange func- tion can be obtained：

Wherein λ >=0, In order to set up generalized lagrange func- tion with original The relation of object function, is obtained by formula (20)

During formula (21) is substituted into formula (18), a new problem for former problem equivalent is obtained：

Formula (22) is still the most value Solve problems containing complicated inequality constraints, if first solving object function λ's Optimal value, L ({ v_k, λ, α) contain more unknown quantity, solve inconvenient.Accordingly, it is considered to pass through the glug of solution formula (22) Bright day dual problem solves former problem, that is, exchange two orders for being most worth solution

OrderNow, former problem becomes two and is most worth Solve problems, and one is to solve forAnother is to solve for

Solve firstSimple mathematical distortions are carried out to formula (20), is madeFormula (20) it is decomposed into M independent incoherent minor：

Wherein, and k-th minor can be expressed as：

Because above-mentioned M minor is separate,：

One new problem can be obtained by formula (26)：

Object function L_k({v_k, λ, α) most value point can be divided into three kinds of situations, and one is that object function has extreme point, and extreme value In domain of definition, now object function obtains maximum to point at extreme point；Two is that object function has extreme point, but extreme point Not in domain of definition, now object function obtains maximum in boundary；Three is that object function does not exist extreme point, now target Function is same to obtain maximum in boundary.

First determine whether object function L_k({v_k, λ, α) whether there is extreme value.When there is extreme value, multi-targets function L_k Each rank partial derivative of ({ vk }, λ, α) must simultaneously meetWith

Solve firstAssuming that λ and α is known quantity.This is a derivation process on vector, Present embodiment is applied to following mathematical knowledge.Assuming that x, y are n dimensional vectors, thenAccording to upper Formula is stated, formula (27-a) derivation can be obtained：

V at extreme point_kMeetObtained by formula (28)：

H in formula (29)_kAnd v_kRelationship Comparison it is complicated, it is impossible to v_kIndividually put forward, if the two can be combined Solution can then be simplified, both members premultiplication simultaneously is made：

OrderWherein G_kIt is complex constant, carrying out abbreviation to formula (30) obtains：

v_kContain N_tIndividual unknown number, only cannot obtain v by (31) equations of formula_kAccurate numerical value is, it is necessary to find pass In v_kOther equations come jointly to v_kSolved.To formula (27-c), (31) andCombined, can ：

Order：

Formula (33-b) is a matrix equation, wherein,So in the basic course laboratory of formula (33-b) Contain an orthonormal basis.For matrix equation, present embodiment is directly asked it using the built-in function in matlab Solution, obtains its orthonormal basisThen the solution of formula (33-b) is：

Wherein, l is any non-zero real,

If there is extreme point, extreme point v ' in object function_kIt is same to meet：

In order to solve the occurrence of l, present embodiment is obtained according to formula (33-a)

As available from the above equation：

Wherein, α needs to meet：

V can be obtained by formula (35)_kExtreme point v '_kFor：

L can be obtained by formula formula (39)_k({v_k, λ, α) on variable v_kExtreme point v '_k, then in conjunction with formula (27- B) extreme point v ' is judged with formula (27-c)_kWhether in domain of definition, v_kDomain of definition Ψ_kCan be expressed as：

If v_k′∈Ψ_k, then L can be determined_k({v_k, λ, α) maximum, v are obtained at extreme point_kOptimal value Can be expressed as

IfBecause L_k({v_k, λ, α) it is on v_kFirst increasing after subtraction function, so L_k({v_k, λ, α) under Boundary obtains maximum, v_kLower border value v_k" can be expressed as：

Now, v_kBoundary value v_k" also need to meet formula (33-b), i.e.,：

Formula (42) is substituted into formula (43) to obtain：

Formula is carried out to deform：

Therefore, v_kOptimal valueCan be expressed as：

Wherein, α must is fulfilled for：

In sum, Wave beam forming vector v_kOptimal valueCan be expressed as：

Then the optimal value of α is solved, α has clear and definite span, i.e., 0<α<1, can be obtained by formula (22) and formula (31)：

α≥max(Θ₁, Θ₂)=Θ_max (49)

Therefore, the span of α can be expressed as：

Judge that α, with the presence or absence of extreme point, seeks local derviation, it is assumed that v to formula (25) on α_kIt is known quantity with λ, obtains：

In order to solve the optimal value of α, order：

Due toAndSoFormula (52) does not exist extreme value without solution, i.e. α Point.L_k({v_k, λ, α) function is incremented by α, therefore, boundary obtains optimal value to α on top.Obtained by formula (18-b)：

Aggregative formula (50)：

Therefore, the optimal value of αFor：

Next second Constrained and Unconstrained Optimization is solved, i.e.,Wherein, Now, g (λ) takes minimum value equally three kinds of situations, and one is that object function has extreme point, and extreme point is in domain of definition, this When object function maximum is obtained at extreme point；Two is that object function has extreme point, but extreme point is not in domain of definition, this When object function boundary obtain maximum；Three is that object function does not exist extreme point, and now object function is same on border Place obtains maximum.λ does not have clear and definite boundary in present embodiment, if so λ is present most is worth, it is inevitable at extreme point.

G (λ) is first determined whether with the presence or absence of extreme value, the derivative of g (λ) on λ is solved and is obtained：

Derivative is zero at extreme point, therefore：

P_m=| v_k|² (57)

When formula (57) are met, g (λ) can obtain minimum value, but can not obtain the occurrence of λ.Then, this implementation Mode is solved with dichotomy, and dichotomy is a kind of alternative manner of Fast Convergent, can be obtained by λ's with reference to formula (41) Optimal valueDichotomy solve detailed process be：The upper bound λ of Lagrange multiplier is set_maxWith lower bound λ_min, using formulaValue to λ is updated so that the value of the λ that formula (57) is set up, is exactly the optimal value of λ

(2) mathematical problem of united beam forming method is solved：

Formula (19-d) is absorbed in object function first with Lagrangian relaxation technology, a broad sense can be obtained Lagrangian：

Wherein λ >=0, from formula (58),So,：

Formula (59) is substituted into formula (19), a new problem of equal value with formula (19) can be obtained：

Formula (60) is still an extreme value Solve problems for the inequality constraints of complexity, it is considered to by solution formula (60) Lagrange duality problem solves former problem, i.e.,：

OrderNow, former problem becomes two and is most worth Solve problems, one It is to solve forAnother is to solve for

Solve firstOrderSimple mathematical distortions are carried out to formula (28), ：

Formula (62) is decomposed into M independent incoherent minor.Wherein, and k-th minor can be expressed as：

Because above-mentioned M minor is separate, can be obtained by formula (32) and formula (33)：

Then, formula (60-b) and formula (60-c) are deformed, and formula (60-d) is substituted into formula (60-b), so A new problem can be obtained with reference to formula (64) afterwards：

Object function L_k({v_k},{w_k, λ, α) most value point can be divided into three kinds of situations, and one is that object function has extreme point, And extreme point is in domain of definition, now object function obtains maximum at extreme point；Two is that object function has extreme point, but Not in domain of definition, now object function obtains maximum to extreme point in boundary；Three is that object function does not exist extreme point, this When object function is same obtains maximum in boundary.

First determine whether object function L_k({v_k},{w_k, λ, α) whether there is extreme value.When there is extreme value, multi-targets letter Number L_k({v_k},{w_k, λ, α) each rank partial derivative must simultaneously meet With

Solve firstMathematical problem of the solution procedure similar to conventional beams forming method Solution.It is final to obtain v_kExtreme point v '_kFor：

Judge extreme point whether in domain of definition then in conjunction with formula (65-b), v_kDomain of definition can be expressed as：

If the v at extreme point_k′∈Φ_k, then L can be determined_k({v_k},{w_k, λ, α) obtain maximum at extreme point It is worth, then v_kOptimal valueFor：

If at extreme pointL_k({v_k},{w_k, λ, α) it is on v_kFirst increasing after subtraction function, so L_k ({v_k},{w_k, λ, α) maximum, v are obtained at lower boundary_kLower border value v_k" can be expressed as：

v_k" solution procedure similar to the solution of the mathematical problem in conventional beams forming method, obtain v_kBoundary value v_k″ For：

Now, v_kOptimal valueFor：

In sum, information Wave beam forming vector v_kOptimal valueCan be expressed as：

Then w is solved_kOptimal value, in order to simplify the representation of formula (65), make g_k=| v_k|²,Its In, by solution above, g_kAnd b_kAll it is known quantity, now formula (65) can be equivalent to：

Wherein, object function can be expressed as：

Can be obtained by formula (73-c), w_kDomain of definition I_kCan be expressed as：

Therefore, formula (73) can be rewritten as

Wherein,

Ψ_k(w_k) most value point can equally be divided into three kinds of situations, one is that object function has extreme point, and extreme point in definition In domain, now object function obtains maximum at extreme point；Two is that object function has extreme point, but extreme point not in definition In domain, now object function obtains maximum in boundary；Three is that object function does not exist extreme point, and now object function is same Maximum is obtained in boundary.

First determine whether Ψ_k(w_k) extreme point whether there is.To Ψ_k(w_k) derivation obtains

First assume Ψ_k(w_k) exist, in order to solve w_kExtreme point, according to taking graceful theorem, makeBy formula (77)：

2(α-1)λw_k=0 (78)

Formula (78) solve, w_k=0 or α=1.Obviously it is invalid, so Ψ_k(w_k) most value point belongs to Three kinds of situations.It is on w that can obtain object function by formula (76)_kSubtraction function, so w_kMaximum is obtained at lower boundary, by Formula (75) can learn w_kDomain of definition be I_k, its lower boundary I_L_kCan be expressed as

When formula (80) are met：

Because on w_kConstraint it is less, so its value is more wide in range, formula (79) and (80) are w_kMeet domain of definition One of which value, it is but not unique.

Formula (79) contains a matrix equationWherein,So containing in the basic course laboratory of equation There is N_t- 1 orthonormal basis.For matrix equation, present embodiment can utilize the built-in function in matlab directly to enter to it Row is solved, and obtains its N_t- 1 orthonormal basis, respectivelyEquationSolutionFor

Wherein,It is any non-zero real.

OrderObtained by formula (81)

Because each vector in basic course laboratory is mutually orthogonal, formula (82) can be rewritten as

Formula (83) is write as matrix form, is obtained

AL=Λ (84)

Wherein,

Λ=[Λ₁ ²,…,Λ_k-1 ²,Λ_k+1 ²,…,Λ_M ²]^H∈C^(M-1)×1 (87)

A and Λ are known quantity, and A is square formation, then the solution of formula (84) is

L=A^-1Λ (88)

Then open the element in vectorial L radical sign and take on the occasion of obtaining

In sum, w_kOptimal valueFor

Wherein,WithCan be obtained by matlab built-in functions and formula (88) respectively.

Then the optimal value of α is solved, α has clear and definite span, i.e., 0<α<1, can be obtained by formula (80)

In sum, the span of α can be expressed as

Judge that α, with the presence or absence of extreme point, local derviation is sought to formula (63) on α, obtain

In order to solve the optimal value of α, order

When formula (93) are met, object function obtains maximum, but can not obtain the occurrence of α.Then, this implementation Mode is solved with dichotomy, it is assumed that the initial value of α is α_start, the final value of α is α_end, using formulaTo α Value be updated, meet the value of the α of formula (93), be exactly the optimal value of α

Next second Constrained and Unconstrained Optimization is solved, i.e.,Wherein, G (λ) takes minimum value equally three kinds of situations, and one is that object function has extreme point, and extreme point is in domain of definition, now target Function obtains maximum at extreme point；Two is that object function has extreme point, but extreme point is not in domain of definition, now target Function obtains maximum in boundary；Three is that object function does not exist extreme point, and now object function is same obtains in boundary Maximum.λ does not have clear and definite boundary in present embodiment, if so λ is present most is worth, it is inevitable at extreme point.

The derivative of g (λ) on λ is solved first to obtain

Then it can be seen from graceful theorem is taken, derivative is zero at extreme point, therefore

P_m-α|v_k|²-|w_k|²+α|w_k|²=0 (95)

When formula (95) are met, g (λ) can obtain minimum value, but can not obtain the occurrence of λ.Also with two points Method is solved, and sets the upper bound λ of Lagrange multiplier_maxWith lower bound λ_min, using formulaValue to λ is carried out Update, meet the value of the λ of formula (79), be exactly the optimal value of λ

Four：Performance comparison is emulated

Formula can be seen that during using conventional beams forming method and united beam forming method of the invention, Yong Huping As being with information rate expression formula, but due to using same Wave beam forming vector v in conventional beams forming method_k Information and energy are controlled, cause in energy acceptance time slot originally can using inter-user interference as the source of energy, But due to the control formed to close-to zero beam so that collection of energy end only have collected the energy included in own user information. And united beam forming method of the invention then uses two Wave beam forming vector vs_kAnd w_kInformation and energy are controlled respectively System, receiving time slot in information can remove user's interference, and more energy can be collected into again in energy acceptance time slot, here it is this The advantage part of invention.

Assuming that base station is equipped with N in the MISO SWIPT systems_tRoot antenna, receiving terminal has M user.Transmitting terminal uses wave beam Formation is oriented transmission to user profile.What the present invention was respectively adopted be Wave beam forming strategy and united beam formed slightly to Family information carries out alignment transmission, and in order to the interference between user is completely eliminated, present embodiment makes N_t=M.In order to reduce calculating Complexity, it is assumed that the parameters of each user are identicals, i.e. the Signal to Interference plus Noise Ratio thresholding γ of each user_k, antenna makes an uproar Acoustical powerThe energy conversion efficiency ζ of user_kAnd the energy threshold e of user_kIt is equal, i.e. γ_k=γ,ζ_k =ζ, e_k=e.Simulation parameter sets as shown in table 1.

The simulation parameter of table 1

From fig. 4, it can be seen that when SWIPT systems and speed increase, the energy of SWIPT systematic collections reduces, this be because For transmit power thresholding P is certain, for the more of information demodulation, then for the fewer of collection of energy.When SWIPT systems and When speed is equal, the energy that the energy ratio application conventional beams forming method collected using united beam forming method of the present invention is collected Amount is more, because in energy time slot, united beam forming method of the present invention forms vector by independent energy beam It is controlled, have collected unnecessary user's interference as energy, and conventional beams forming method is in collection of energy time slot, use Interference between family is eliminated, but is completely unnecessary for collection of energy.When the energy collected is identical, Using the SWIPT systems and speed of SWIPT systems and speed than application conventional beams forming method of united beam forming method Rate is big, because when collecting identical energy, united beam forming method is accounted for due to have collected unnecessary user's interference Energy time slot is smaller, is that information receives the more time slots of residue, therefore SWIPT systems and speed are bigger.

Although have references to specific implementation method in the present embodiment to describe the present invention, it should be understood that, These embodiments are only the example of principles and applications.It should therefore be understood that can be to exemplary implementation Example carries out many modifications, and can be designed that other arrangements, without departing from the present invention that appended claims are limited Spirit and scope.It should be understood that can be by way of different from described by original claim come with reference to different Feature described in dependent claims and present embodiment.It will also be appreciated that the spy with reference to described by separate embodiments Levying can use in other embodiments.

Claims (8)

1. a kind of united beam forming method based on SWIPT systems, it is characterised in that methods described comprises the following steps：

Step one：Transmitting terminal sends separate information Wave beam forming vector sum energy velocity of wave and forms vector simultaneously；

Step 2：Receiving terminal enters row information and receives and collection of energy, including information receives time slot and energy acceptance time slot：

Time slot is received in information, using information Wave beam forming vector information resources are carried out with the alignment of specific user, realize information Error free transport；

In energy acceptance time slot, forming vector using energy beam carries out the alignment of specific user to energy resource, by between user Disturb the source as collection of energy.

2. the united beam forming method based on SWIPT systems according to claim 1, it is characterised in that the step Two include：

Time slot and energy acceptance time slot are received according to information, object function is set up：

$\underset{\{v_k,w_k,\mathrm{\&alpha;}\}}{max}\underset{k=1}{\overset{M}{\&Sigma;}}{\mathrm{\&alpha;Wlog}}_2(1+\frac{|h_k^H{v}_k{|}^2}{{\mathrm{\&sigma;}}_k^2});$

The bound for objective function：

$\left\{\begin{array}{c}\frac{|h_k^H{v}_k{|}^2}{\underset{i\&NotEqual;k}{\&Sigma;}|h_k^H{v}_i{|}^2+{\mathrm{\&sigma;}}_k^2}\&GreaterEqual;{\mathrm{\&gamma;}}_k,\&ForAll;k\\ (1-\mathrm{\&alpha;}){\mathrm{\&zeta;}}_k\left(\underset{i=1}{\overset{M}{\&Sigma;}}\right|h_k^H{w}_i{|}^2+{\mathrm{\&sigma;}}_k^2)\&GreaterEqual;e_k,\&ForAll;k\\ \mathrm{\&alpha;}\underset{k=1}{\overset{M}{\&Sigma;}}|v_k{|}^2+(1-\mathrm{\&alpha;}\left)\underset{k=1}{\overset{M}{\&Sigma;}}\right|w_k{|}^2\&le;P\\ h_i^H{v}_j=0,\&ForAll;i\&NotEqual;j\\ \mathrm{\&alpha;}\&Element;(0,1)\end{array}\right.$

Wherein, v_kIt is the information Wave beam forming vector of k-th user of transmitting terminal；

M is the quantity of user；

w_kFor the energy beam of k-th user of transmitting terminal forms vector；

α is time slot dividing ratio；W is transmission channel bandwidth；It is h_kConjugate transposition, h_kIt is k-th channel gain square of user Battle array；

It is k-th user antenna noise power；ζ_kIt is k-th energy conversion efficiency of user；γ_kFor information receives time slot kth The minimum Signal to Interference plus Noise Ratio thresholding of individual user；e_kMinimum for k-th user of energy acceptance time slot collects energy threshold；

Optimal solution is sought to setting up parameter in object function；

Enter row information using the object function for trying to achieve optimal solution to receive and collection of energy.

3. the united beam forming method based on SWIPT systems according to claim 2, it is characterised in that described pair is built Vertical Model Parameter asks the optimal solution to be：

Optimal solution is asked to parameter in object function using Lagrange relaxation method.

4. the united beam forming method based on SWIPT systems according to claim 2, it is characterised in that the utilization Lagrange relaxation method asks parameter in object function the method for optimal solution to be：

The object function of foundation is converted to using Lagrange relaxation method：

And

L_k() is Lagrangian, and λ is Lagrange multiplier；

Bound for objective function after conversion is：

$\left\{\begin{array}{c}|h_k^H{v}_k{|}^2\&GreaterEqual;{\mathrm{\&gamma;}}_k{\mathrm{\&sigma;}}_k^2,\&ForAll;k\\ \underset{i=1}{\overset{M}{\&Sigma;}}|h_k^H{w}_i{|}^2\&GreaterEqual;\frac{e_k}{(1-\mathrm{\&alpha;}){\mathrm{\&zeta;}}_k}-{\mathrm{\&sigma;}}_k^2,\&ForAll;k\\ h_k^H{v}_i=0,\&ForAll;k\&NotEqual;i\\ 0<\mathrm{\&alpha;}<1\end{array}\right.;$

Optimal value is asked for object function after conversion.

5. the united beam forming method based on SWIPT systems according to claim 4, it is characterised in that

Information Wave beam forming vector v_kOptimal valueFor：

${\stackrel{\&OverBar;}{v}}_k=\left\{\begin{array}{cc}{v}_k^{\&prime;},& if{v_k}^{\&prime;}\&Element;{\mathrm{\&Phi;}}_k\\ v_k^{\&prime;\&prime;},& if{v_k}^{\&prime;}\&NotElement;{\mathrm{\&Phi;}}_k\end{array}\right.;$

Wherein,Φ_kRepresent Information Wave beam forming vector v_kDomain of definition；N_tRepresent the quantity of transmitting terminal transmission antenna.

6. the united beam forming method based on SWIPT systems according to claim 4, it is characterised in that

Energy beam forms vectorial w_kOptimal valueFor：

${\stackrel{\&OverBar;}{w}}_k=l_1{\tilde{w}}_k^1+l_2{\tilde{w}}_k^2+...+l_{N_t-1}{\tilde{w}}^{N_t-1};$

$L={\&lsqb;{l_1}^2,{l_2}^2,...,{l_{N_t-1}}^2\&rsqb;}^H\&Element;C^{(N_t-1)\&times;1}$

Λ=[Λ₁ ²,…,Λ_k-1 ²,Λ_k+1 ²,…,Λ_M ²]^H∈C^(M-1)×1

$A=\left[\begin{array}{cccc}|h_1^H{\tilde{w}}_k^1{|}^2& |h_1^H{\tilde{w}}_k^2{|}^2& \mathrm{...}& |h_1^H{{\tilde{w}}_k}^{N_t-1}{|}^2\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ |h_{M-1}^H{\tilde{w}}_k^1{|}^2& |h_{M-1}^H{\tilde{w}}_k^2{|}^2& \mathrm{...}& |h_{M-1}^H{{\tilde{w}}_k}^{N_t-1}{|}^2\\ |h_M^H{\tilde{w}}_k^1{|}^2& |h_M^H{\tilde{w}}_k^2{|}^2& \mathrm{...}& |h_M^H{{\tilde{w}}_k}^{N_t-1}{|}^2\end{array}\right]\&Element;C^{(M-1)\&times;(N_t-1)}$

According to L=A^-1Λ, obtains

ForOrthonormal basis；

N_tRepresent the quantity of transmitting terminal transmission antenna.

7. the united beam forming method based on SWIPT systems according to claim 4, it is characterised in that the α is most The figure of merit

The initial value of α is α_start, the final value of α is α_end, using formulaValue to α is updated, and works as satisfactionWhen α values, be exactly the optimal value of the α

8. the united beam forming method based on SWIPT systems according to claim 4, it is characterised in that the λ is most The figure of merit

The upper bound of Lagrange multiplier is λ_max, the lower bound of Lagrange multiplier is λ_min, using formulaTo λ's Value is updated, when meeting P_m-α|v_k|²-|w_k|²+α|w_k|²λ value when=0, is exactly the optimal value of the λWhereinP is the maximum transmit power thresholding of transmitting terminal.