CN101079660A - Downlink power distribution method of multi-user JT MIMO system - Google Patents

Abstract

The invention discloses a downward link power distributing method of multi-user JT MINO system in the wireless communicating technical domain, which is characterized by the following: finding a practicable sending power vector after satisfying every user stem lower than small rain threshold; minimizing the total transmitting power of all users; calculating power allocating vector of each user through minimum energy calculating formula; transmitting the power control for downward link user according to power allocating vector; controlling the transmitting quality of each user effectively; improving the energy efficiency; expanding the system capacitance.

Description

Downlink power distribution method in the multi-user JT mimo system

Technical field

The present invention relates to wireless communication technology field, relate in particular to the power division of channel in the multi-user system.

Background technology

Joint transmission JT technology is the multi-user's transmission technology that is applicable to the cdma system down link.It is a transmission precoding technique with respect to joint-detection JD, and it is finished transferring to the base station end in the complex calculation that terminal is finished in the JD system.At present, the JT technology is divided into linear and non-linear two big classes, non-linear joint transmission technology, as Tomlinson-Harashima precoding (THP) and the non-linear ZF of transmission (TxNZF) etc., can be when guaranteeing transmission quality with transmission power control a lower level, but its encoder complexity is higher, document (Hai Ping, Xie Xianzhong, Optimum Transmission SchemeJoint (JT) in TD-SCDMA Downlink. China post and telecommunications Journal in Colleges and Universities (English edition), 2004, Vol.11 (3): 45-48) disclose a kind of linear combined transmission technology, encoder complexity is lower, easily practicability.But there is bigger energy gain in precoding in the linear combined transmission technology, makes the multiple access between the user disturb increase, directly influences the total capacity of system.Simultaneously, because the influence of channel estimation errors and channel fading makes the communication quality of whole system that bigger reduction be arranged.

Summary of the invention

For overcoming above unfavorable factor, in linear joint transmission technology, power control is a requisite link.The present invention is directed to the above-mentioned deficiency of prior art, a kind of power Control Allocation method that is applicable to multi-user's multiple-input and multiple-output joint transmission (JT MIMO) system is proposed, under the prerequisite that satisfies each user's signal interference ratio SINR, find a feasible transmitted power vector, make all users' transmission gross power minimum simultaneously; By uniting the transmitted power of adjusting each user, realize control to each user's signal interference ratio.

The technical scheme that the present invention solves the problems of the technologies described above is, at each user's of downlink estimation self link gain and to other users' interference; Signal interference ratio threshold value γ according to the given user k of user's priority ^(k), making up with user's signal interference ratio threshold value is the diagonal matrix D that diagonal constitutes; Determine constraints: the signal interference ratio SINR each user is higher than under the prerequisite of threshold value, makes total transmission energy minimum of all users; Make up intersymbol interference (ISI), inter-user interference (MAI) obstacle gain matrix F and thermal noise interference matrix U; According to constraints, interference matrix, make up the least energy computing formula that satisfies under the constraints, i.e. the transmission vector E of each user's transmitted power composition _d=[I-DF] ^-1U, wherein I is a unit matrix; Call algoritic module, according to formula E _d=[I-DF] ^-1U rated output allocation vector.

Multiply each other and just can obtain the data symbol that we need send sending data with the user after the power allocation vector of gained and the modulation before the precoding.

Can make the transmission gross energy minimum of system according to the proposed method, and can control each user's transmission quality effectively, thereby improve energy efficiency, increase power system capacity.

Description of drawings

Figure 1 shows that the down channel system architecture diagram

Figure 2 shows that and determine the transmitted power flow chart according to the present invention

Embodiment

In JT mimo system multi-user downlink, with the frequency parallel transmission, each user's emitted energy all exerts an influence to other all users, is an interference limiting system on many antennas in each user data process modulation matrix M modulation back.Therefore, how reasonably the emitted energy of distributing user under the prerequisite of the transmission quality that guarantees the user, reduces transmitted power, improves energy efficiency, is the problem that the JT mimo system need solve.

Be illustrated in figure 1 as the down channel system construction drawing, between QPSK modulation module and spread spectrum module, insert a power division module, the power division module is carried out vector power and is distributed, user after the power allocation vector that obtains and the modulation is sent data multiplies each other, and then carry out spread spectrum, preliminary treatment is launched by antenna.

Figure 2 shows that the present invention determines the power allocation vector flow chart, by estimating each user's self link gain at down link with training sequence and, disturbing to comprise that intersymbol interference ISI, inter-user interference MAI, thermal noise disturb to other users' interference.In conjunction with concrete communication environment such as landform, user density, ambient noise or the like, and each user's priority, the signal interference ratio SINR threshold value of given user k is γ ^(k), be that diagonal makes up diagonal matrix D, the channel gain matrix G of estimating user with user's signal interference ratio threshold value _kWith interference matrix G _Kg, determine each user's signal interference ratio SINR to determine constraints: be higher than at each user's SINR under the prerequisite of threshold value, make total transmission energy of all users

Minimum, just $\min \underset{k}{\mathrm{\Σ}}{E}_d^{\left(k\right)}\mathrm{Subject\; to}{\mathrm{\γ}}_o^{\left(k\right)}\≥{\mathrm{\γ}}^{\left(k\right)},$ γ _o ^(k)Signal interference ratio for user's reality; Make up intersymbol interference (ISI), inter-user interference (MAI) gain matrix F and thermal noise interference matrix U,, determine the least energy computing formula, i.e. the most rational power allocation vector E according to above-mentioned constraints and interference matrix _d=[I-DF] ^-1U.

According to the system channel feature such scheme is specifically described below.Under channel circumstance, adopt linear ZF (TxZF) precoding algorithm to carry out preliminary treatment to sending data. adopt the maximum likelihood algorithm estimated channel impulse response.And at down channel by estimating to obtain coding, transmission gain and inter-user interference that is subjected to and the intersymbol interference of user k.

If the channel information that transmitting terminal sends is no inclined to one side, TxZF can eliminate intersymbol interference and inter-user interference fully so, but in practical communication, no inclined to one side channel information is non-existent, simultaneously because the limitation of some Preprocessing Algorithm, user's interference is from intersymbol interference ISI, inter-user interference MAI and thermal noise.So the signal interference ratio SINR of user k is γ _o ^(k):

${\mathrm{\γ}}_o^{\left(k\right)}=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\left|{\left[D^{\left(k\right)}{H}^{\left(k\right)}{M}^{\left(k\right)}\right]}_{n,n}\right|}^2{E}_d^{\left(k\right)}}{\underset{g=1}{\overset{K}{\mathrm{\Σ}}}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\left|\right|{\left[D^{\left(k\right)}{H}^{\left(k\right)}{M}^{\left(g\right)}\right]}_n\left|\right|}^2{E}_d^{\left(g\right)}-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\left|{\left[D^{\left(k\right)}{H}^{\left(k\right)}{M}^{\left(k\right)}\right]}_{n,n}\right|}^2{E}_d^{\left(k\right)}+\mathrm{trace}\left\{D^{\left(k\right)}{R}_n^{\left(k\right)}{D}^{\left(k{)}^H\right)}\right\}}$ Following formula is made normalized, and the SINR that can obtain each frame is ${\mathrm{\γ}}_o^{\left(k\right)}=\frac{E_d^{\left(k\right)}}{\underset{g=1}{\overset{K}{\mathrm{\Σ}}}\frac{G_{\mathrm{kg}}}{G_k}{E}_d^{\left(g\right)}-E_d^{\left(k\right)}+\frac{N_k}{G_k}}$

Wherein, user's channel gain matrix $G_k=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\left|{\left[D^{\left(k\right)}{H}^{\left(k\right)}{M}^{\left(k\right)}\right]}_{n,n}\right|}^{2}k$ And interference matrix $G_{\mathrm{kg}}=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\left|\right|{\left[D^{\left(k\right)}{H}^{\left(k\right)}{M}^{\left(g\right)}\right]}_n\left|\right|}^2$ Represent the transmission gain of user k and the interference that is subjected to respectively, comprise inter-user interference and intersymbol interference, wherein D ^(k)Be the demodulation matrix of user k, M ^(k)Be the modulation matrix of user k, H ^(k)Channel matrix for user k.

According to concrete communication environment, as landform, user density, ambient noises etc., and each user's priority are provided with the signal interference ratio threshold value γ of user k ^(k), making up with k user's signal interference ratio threshold value is the diagonal matrix that diagonal constitutes

Set constraints: be higher than at each user's SINR under the prerequisite of threshold value, make total transmission energy of all users

Minimum, just $\min \underset{k}{\mathrm{\Σ}}{E}_d^{\left(k\right)}\mathrm{Subject\; to}{\mathrm{\γ}}_o^{\left(k\right)}\≥{\mathrm{\γ}}^{\left(k\right)}.$

According to above-mentioned constraints, set up the power allocation vector matrix $E_d={(E_d^{\left(1\right)}\·\·\·E_d^{\left(K\right)})}^T,$

Determine ISI, MAI obstacle gain matrix according to user's channel gain matrix, interference matrix and user's signal interference ratio

Thermal noise interference matrix $U={({\mathrm{\γ}}^{\left(1\right)}\frac{N_1}{G_1}\·\·\·{\mathrm{\γ}}^{\left(K\right)}\frac{N_K}{G_K})}^T{N}_k,$ N wherein _kWhite Gaussian noise for user k.

Base station end control treatment module is called algoritic module, makes up the least energy computing formula that satisfies under the constraints, makes all users' vector power E _dSatisfy inequality E _d〉=DFE _d+ U, when above-mentioned inequality was got equal sign, user's gross power minimum can get least energy computing formula: E thus _d=[I-DF] ^-1U, i.e. the most rational power allocation vector.Wherein I is a unit matrix, power allocation vector matrix E _dIn each element represent a user's transmitted power.Above-mentioned matrix D, F, U substitution formula are promptly obtained rational power allocation vector, i.e. the transmitted power vector.

In multi-user JT mimo system, the power allocation vector that transmitter is determined according to said method multiplies each other with it and each user's modulated data symbol, and then carries out spread spectrum, and preliminary treatment is launched by antenna.By power distribution method of the present invention, realize downlink user is carried out power control, realize control simultaneously to each user's signal interference ratio, the communication quality of whole system is had greatly improved.

Claims (5)

1, the downlink power distribution method in the multi-user JT mimo system is characterized in that, at each user's of downlink estimation self link gain and to other users' interference; The signal interference ratio threshold value γ of given user k ^(k), be that diagonal makes up diagonal matrix D with user's signal interference ratio threshold value; Make up obstacle gain matrix F and thermal noise interference matrix U; Determine least energy computing formula E according to constraints and interference matrix _d=[I-DF] ^-1U; Processor controls is called the power allocation vector that least energy computing formula in the algoritic module is calculated each user; Transmitter is realized the power of downlink user is controlled according to power allocation vector.

2, downlink channel power distribution method according to claim 1 is characterized in that, described interference is: intersymbol interference ISI, inter-user interference MAI, thermal noise disturb.

3, downlink channel power distribution method according to claim 1 is characterized in that, described constraints is: the signal interference ratio each user is higher than under the prerequisite of threshold value, makes total transmission energy minimum of all users.

4, downlink channel power distribution method according to claim 1 is characterized in that, constructed obstacle gain matrix, thermal noise interference matrix are determined by user's channel gain matrix, interference matrix and user's signal interference ratio.

5, downlink channel power distribution method according to claim 1 is characterized in that, the signal interference ratio threshold value of user k is determined by communication environment and user's priority.

Images