CN106357315A - Energy efficiency resource distribution method for large-scale multi-antenna network with incomplete CSI (channel state information) - Google Patents

Abstract

The invention discloses an energy efficiency resource distribution method for a large-scale multi-antenna network with incomplete CSI (channel state information). User equipment is charged by adoption of an energy collection method and transmits data to a base state by the aid of collected energy; N antennas are matched with a multi-antenna system to construct a block fading channel model, and the block fading channel model is subjected to channel estimation to obtain an estimation channel capacity between the base station and the user equipment; on the basis of comprehensive consideration of circuit loss and power consumption of a transceiving terminal, analysis is performed to obtain an antenna selection algorithm which enables optimization of energy efficiency, so that energy efficiency of the multi-antenna system is greatly improved. According to the energy efficiency resource distribution method, an energy collection technique, an antenna selection algorithm and a resource distribution optimization algorithm are adopted under incomplete channel estimation to effectively improve energy efficiency of the multi-antenna network.

Description

A kind of large-scale multiple antennas network energy efficiency resource allocation methods with incomplete csi

Technical field

The present invention relates to wireless communication technology field, more particularly, to a kind of large-scale multiple antennas network with incomplete csi Efficiency resource allocation methods.

Background technology

The explosive increase of smart mobile phone industry, multimedia service and Mobile solution is highly dependent on high-speed data wireless network The development of network.Although our life of continuous change of the high-data-rate wireless system of rapid growth, or even the day changing us Chang Hangwei, but be still nervous to the demand of high data rate service in current network, the battery consumption of mobile terminal also than from Front a lot of soon.On the other hand, for the demand of contrast high data rate service and fast wireless network, the development of battery technology is Slow and progressive.This mismatch necessarily leads to research worker to need, to improving, the mobile terminal that continuously runs in wireless network The research interest of life cycle and enthusiasm.

Limited battery capacity is to improve one of major obstacle of life cycle, and it is also to improve user's high data speed simultaneously The major obstacle of rate multimedia service.The method of the life cycle of one prolongation mobile terminal is to provide energy supply.However, giving Battery charges or replaces the battery of mobile terminal and may result in high cost, sometimes even inconvenient or can not possibly be real Existing.Under currently this environment, energy collection technology can extend the Life Cycle of energy constraint network on a continuing basis Phase.Energy collection technology makes the receiving device in wireless network collect energy from the environment of surrounding to be possibly realized.Meanwhile, Most energy source be since distance, such as solar energy and wind energy etc..However, solar energy can hardly be accessed for those Or for the wireless device of these energy sources of wind energy, how to provide energy supply to be that problem is located again.Recently, except from the sun Obtain outside the technology of energy in energy, wind energy, vibration, pyroelectric effect or other physical phenomenons, within a wireless communication network, wireless work( Rate transmission becomes the focus of research, and it provides a promising solution for the network of traditional energy constraint.Because Radiofrequency signal not only can carry information, can also transmit energy, and this is extending the battery statement cycle and is improving system energy efficiency side Face is made that huge contribution.

Mimo (multiple-in multi-out, multiple-input and multiple-output) technology is to realize information in future communications network One of key technology of high-speed transfer, it without increase in bandwidth, can exponentially improve the mutual information of communication system And spectrum efficiency.The benefit of large-scale multiple antennas not only contains the advantage of mimo, also includes when number of antennas tends to infinite, and one A little declines and incoherent effect of noise all can disappear, so that spectrum efficiency is unrelated with bandwidth.

Because the transmission antenna number in large-scale multiple antenna communication is a lot, and mobile terminal quantity is far smaller than bs's Transmission antenna quantity is although system intermediate frequency spectrum efficiency significantly improves, but the problem of the energy expenditure in communication is also further serious. Transmission antenna number is more, and radio frequency (rf) link being used is also more, and corresponding digital to analog converter in rf chain, frequency mixer, The power attenuation of transmitting filter, frequency synthesizer, low-noise amplifier, audio frequency amplifier, wave filter and analog-digital converter The more, energy consumption is the more.Therefore system carries out day line options is necessary, improves system while reducing power attenuation Energy efficiency.

Due to the presence of channel estimation errors, transmitting terminal is difficult to obtain complete channel status information (csi), channel training mistake Journey is that system is known necessary to csi, and the incompleteness of this process can directly result in the decline of systematic function.Therefore, with complete Standby channel contrast, the research of problem under incomplete channel then more meets the research conditions of reality.

Content of the invention

Present invention aim at providing, a kind of system energy efficiency is high, calculate the large-scale multiple antennas simply with incomplete csi Network energy efficiency resource allocation methods.

For achieving the above object, the method for the invention comprises the following steps:

Step 1, base station (bs) is trained to bulk nanometer materials estimating by channel estimator, the letter after being estimated Road；

Step 2, using the mode of wireless power transfer, user equipment collects energy from base station；

Step 3 it is considered in channel incomplete csi situation, calculate multiaerial system in when transmission antenna number be n when letter Road capacity；

Step 4 it is considered in channel incomplete csi situation, calculate multiaerial system in add Antenna Selection Technology after Excellent antenna number is channel capacity during l；

Step 5, is set up power attenuation model, and is built based on the channel capacity after above-mentioned sky line options and power attenuation model Vertical energy efficiency model；

Step 6, using the Antenna Selection Algorithem based on binary chop and optimization resource allocation algorithm, by Lagrange Multiplier method carries out the combined optimization of time and transmit power to above-mentioned energy efficiency model, obtains antenna number and the optimum of optimum Can valid value.

Step 7, obtains the transmit power of optimum according to above-mentioned optimal energy efficiency.

Further, in step 1, the channel after estimation is:

$\hat{h}=h-\mathrm{\δ}e$

Wherein,For the channel matrix between the base station after estimating and user；H is complete between base station and user equipment Channel；δ e represents estimation difference, and has It is the variance of channel estimation errors.

Further, in step 2, the communication between base station and user equipment is divided into two time periods, first In the individual time period, base station is user's charging by wireless power transfer (wpt), and in second time period, user is using collecting Energy transfers information to base station.According to law of conservation of energy, the energy that user collects in first time period is:

$e_u={\mathrm{\η\αp}}_t\left|{\hat{h}}^h\mathrm{\β}\right|\mathrm{\τ}$

Wherein, η represents user and the power of collection is converted into the transformation efficiency that power storage is got up；α represents base station and use The path loss that distance produces is depended between the equipment of family；p_tRepresent the transmit power of base station end；β is energy beam shaping square Battle array, hasτ is the time of first paragraph in a time block.

Further, in step 3, on the premise of incomplete channel estimation, calculate and in multiple antennas, work as transmission antenna Quantity be n, and n > > 1 when up-link mutual channel capacity:

$c={\log }_2(1+n\frac{\frac{e_u}{t-\mathrm{\τ}}\mathrm{\α}}{{\mathrm{\σ}}_n^2+\frac{e_u\mathrm{\α}}{t-\mathrm{\τ}}{\mathrm{\σ}}_e^2})$

Wherein, t represents the total time of one period of uplink and downlink link,It is the variance of additive white Gaussian noise, It is the variance of channel estimation errors, δ h obeys multiple Gauss distribution (complex normal distribution), that is,

Further, in step 4, on the premise of incomplete channel estimation, calculate and in multiaerial system, work as transmission Antenna sum be n and n > > 1, select optimal antenna number be l when up-link channel capacity:

$c={\log }_2(1+(1+ln\frac{n}{l}\left)\frac{e_u\mathrm{\α}/(t-\mathrm{\τ})}{{\mathrm{\σ}}_n^2+e_u{\mathrm{\α\σ}}_e^2/(t-\mathrm{\τ})}l\right).$

Further, the power attenuation model described in step 5 is:

u(p_t, τ, l)=(lp_bs+p_user)t+p_tτ

In formula, p_bsIt is the power attenuation of every antenna of base station end, p_userIt is the power attenuation of user side, respectively by following two Individual formula represents,

p_user=2p_syn+p_lna+p_mix+p_ifa+p_filr+p_adc

p_bs=p_dac+p_mix+p_filt

In formula, p_dac,p_mix,p_filt,p_syn,p_lna,p_ifa,p_filr,p_adcRepresent digital to analog converter, frequency mixer respectively, send filter The power attenuation of ripple device, frequency synthesizer, low-noise amplifier, audio frequency amplifier, wave filter and analog-digital converter；

Based on above-mentioned power attenuation model, calculate described energy efficiency model:

$ee(p_t,\mathrm{\τ},l)=\frac{c}{u(p_t,\mathrm{\τ},l)}.$

Further, in step 7, acquisition optimum efficiency corresponding optimum transmit power such as following formula:

$p_t^{*}=\frac{1}{2}\frac{\sqrt{{(x_3-x_4)}^2{x}_2^2+4\underset{i=1}{\overset{5}{\π}}{x}_i}-x_3{x}_2-x_4{x}_2}{x_3{x}_4}$

Wherein, χ₁～χ₅It is respectively as follows:

$x_1=\mathrm{\η}\mathrm{\α}\left|\right|\hat{h}|{|}^2\mathrm{\τ}(l+lln\frac{l}{n}),x_2=(t-\mathrm{\τ}){\mathrm{\σ}}_n^2,x_3=\mathrm{\η}\mathrm{\α}|\left|\hat{h}\right|{|}^2{{\mathrm{\τ\σ}}_e}^2,$

$x_4=\mathrm{\η}\mathrm{\α}\left|\right|\hat{h}|{|}^2\mathrm{\τ}(l+lln\frac{l}{n})+\mathrm{\η}\mathrm{\α}|\left|\hat{h}\right|{|}^2{{\mathrm{\τ\σ}}_e}^2,x_5=\frac{x_2^2+x_2{x}_3{p}_t+x_4{x}_2{p}_t+x_4{x}_3{p}_t^2}{x_1{x}_2}.$

Work process approximately as:

Under incomplete channel estimation, the efficiency resource allocation methods of mimo days line options of massive, are received by energy Collection, massive mimo, sky line options, the technology such as beam shaping, the energy efficiency model needing to optimize is proposed, by target problem Solved by method of Lagrange multipliers with the conditions of constraint, obtain optimum energy efficiency, draw the optimum transmit power in bs end.

Compared with prior art, the inventive method has the advantage that

1st, antenna selecting method is quickly found out the transmission antenna number of optimum based on a binary chop algorithm, and uses energy Measure beam forming technique to maximize system energy efficiency.

2nd, combine wireless power transfer and the second period wireless information transfer considering for the first period, present a resource Allocation algorithm, obtains system maximum efficiency by the transmit power of combined optimization bs and the time of first paragraph.

Brief description

Fig. 1 is the system model of the present invention.

Fig. 2 is one of embodiment of the present invention particular flow sheet.

Fig. 3 is the optimization algorithm software flow pattern in the present invention with day line options.

Specific embodiment

The present invention will be further described below in conjunction with the accompanying drawings:

Fig. 1 is the system model of the present invention.Fig. 1 presents has two period communication protocols under an incomplete channel estimation Radio communications system.In system structure, there are 1 base station (bs), 1 single user (user), bs comprises a channel and estimates Gauge (ce, channel estimator).Wherein, bs assembling n root antenna, user assembles 1 antenna.In model, the effect of bs It is to charge for user, carry out the transmission of energy；And user is a single user of receiving terminal, its effect is that storage base station sends Energy so that information transmission is gone back to base station.

Fig. 1 also describes the agreement of two periods.There is a continuous energy supply base station it is therefore an objective to be not required to user Other energy to support.Therefore, in first time period, transmitted by infinite energy, the energy that base station is transmitted is carried out by user Collect, and store during the energy of collection rechargeable battery to one；In second time period, user will using the energy collected Information transmission is to base station.The time of whole transmitting procedure is t, and this is referred to as a time block.In each piece of beginning, in the τ time Interior (the first period), between base station and user, carry out wireless energy transfer (wpt, wireless power transfer)；? In the τ-τ time (the second period), between user and base station, carry out wireless information transfer (wit, wireless information transfer).

Fig. 2 is the concrete grammar flow process of one specific embodiment of the present invention, mainly comprises the steps that

Step 1: estimate the channel model of bulk nanometer materials, the channel after being estimated by training.

Step 2: using the mode of wireless power transfer, user equipment collects energy from base station.

Step 3: consider the situation of incomplete csi in channel, calculate the letter when transmission antenna number is n in multiaerial system Road capacity.

Step 4: consider the situation of incomplete csi in channel, after addition Antenna Selection Technology in calculating multiaerial system Excellent antenna number is channel capacity during l.

Step 5: set up power attenuation model, and based on the channel capacity after above-mentioned sky line options and described power attenuation mould Energy efficiency model set up by type.

Step 6: with reference to Antenna Selection Algorithem and resource allocation optimization algorithm, above-mentioned energy model is carried out with the time and sends out Send the combined optimization of power, obtain the transmission antenna number of optimum and optimum efficiency.

Step 7: obtain the transmit power of optimum according to above-mentioned optimal energy efficiency.

Fig. 3, after algorithm starts, judges whether the ee calculating in step 7 restrains, if convergence, end loop, if do not receive Holding back, then change Lagrange multiplier value, obtain power and the time of circulation next time, if meeting Rule of judgment, continuing to obtain Optimized antenna number and efficiency, thus draw optimum transmit power, end loop.

Embodiment described above is only that the preferred embodiment of the present invention is described, the not model to the present invention Enclose and be defined, on the premise of without departing from design spirit of the present invention, the technical side to the present invention for the those of ordinary skill in the art Various modifications and improvement that case is made, all should fall in the protection domain of claims of the present invention determination.

Claims (7)

1. a kind of large-scale multiple antennas network energy efficiency resource allocation methods with incomplete csi, mainly include user equipment, base Stand, antenna is it is characterised in that the method comprising the steps of:

Step 1, base station (bs) is trained to bulk nanometer materials estimating by channel estimator, the channel after being estimated；

Step 2, using the mode of wireless power transfer, user equipment collects energy from base station；

Step 3 it is considered in channel incomplete csi situation, calculate multiaerial system in when transmission antenna number be n when channel hold Amount；

Step 4 it is considered in channel incomplete csi situation, calculate and in multiaerial system, add optimum sky after Antenna Selection Technology Line number is channel capacity during l；

Step 5, sets up power attenuation model, and sets up energy based on the channel capacity after above-mentioned sky line options and power attenuation model Amount efficiency model；

Step 6, using the Antenna Selection Algorithem based on binary chop and optimization resource allocation algorithm, by Lagrange multiplier Method carries out the combined optimization of time and transmit power to above-mentioned energy efficiency model, obtains the antenna number of optimum and optimum efficiency Value.

Step 7, obtains the transmit power of optimum according to above-mentioned optimal energy efficiency.

2. a kind of large-scale multiple antennas network energy efficiency resource allocation methods with incomplete csi according to claim 1, its It is characterised by: in step 1, the channel after estimation is:

$\hat{h}=h-\mathrm{\δ}e$

Wherein,For the channel matrix between the base station after estimating and user；H is complete letter between base station and user equipment Road；δ e represents estimation difference, and has It is the variance of channel estimation errors.

3. a kind of large-scale multiple antennas network energy efficiency resource allocation methods with incomplete csi according to claim 1, its It is characterised by: in step 2, the communication between base station and user equipment is divided into two time periods, first time period Interior, base station is passed through wireless power transfer (wpt) and is charged for user, and in second time period, user will be believed using the energy collected Breath is sent to base station；According to law of conservation of energy, the energy that user collects in first time period is:

$e_u={\mathrm{\η\αp}}_t\left|{\hat{h}}^h\mathrm{\β}\right|\mathrm{\τ}$

Wherein, η represents user and the power of collection is converted into the transformation efficiency that power storage is got up；α represents base station and user sets The path loss that distance produces is depended between standby；p_tRepresent the transmit power of base station end；β is energy beamforming matrix, hasτ is the time of first paragraph in a time block.

4. a kind of large-scale multiple antennas network energy efficiency resource allocation methods with incomplete csi according to claim 1, its It is characterised by: in step 3, on the premise of incomplete channel estimation, calculate when transmission antenna quantity is n in multiple antennas, And n > > 1 when up-link mutual channel capacity:

$c={\log }_2(1+n\frac{\frac{e_u}{t-\mathrm{\τ}}\mathrm{\α}}{{\mathrm{\σ}}_n^2+\frac{e_u\mathrm{\α}}{t-\mathrm{\τ}}{\mathrm{\σ}}_e^2})$

Wherein, t represents the total time of one period of uplink and downlink link,It is the variance of additive white Gaussian noise,It is letter The variance of channel estimation error, δ h obeys multiple Gauss distribution (complex normal distribution), that is,

5. a kind of large-scale multiple antennas network energy efficiency resource allocation methods with incomplete csi according to claim 1, its It is characterised by: in step 4, on the premise of incomplete channel estimation, calculate in multiaerial system when transmission antenna sum For n and n > > 1, select optimal antenna number be l when up-link channel capacity:

$c={\log }_2(1+(1+ln\frac{n}{l}\left)\frac{e_u\mathrm{\α}/(t-\mathrm{\τ})}{{\mathrm{\σ}}_n^2+e_u{\mathrm{\α\σ}}_e^2/(t-\mathrm{\τ})}l\right).$

6. a kind of large-scale multiple antennas network energy efficiency resource allocation methods with incomplete csi according to claim 1, its It is characterised by, the power attenuation model described in step 5 is:

u(p_t, τ, l)=(lp_bs+p_user)t+p_tτ

In formula, p_bsIt is the power attenuation of every antenna of base station end, p_userIt is the power attenuation of user side, public by following two respectively Formula represents,

p_user=2p_syn+p_lna+p_mix+p_ifa+p_filr+p_adc

p_bs=p_dac+p_mix+p_filt

In formula, p_dac,p_mix,p_filt,p_syn,p_lna,p_ifa,p_filr,p_adcRepresent digital to analog converter, frequency mixer respectively, send filtering The power attenuation of device, frequency synthesizer, low-noise amplifier, audio frequency amplifier, wave filter and analog-digital converter；

Based on above-mentioned power attenuation model, calculate described energy efficiency model:

$ee(p_t,\mathrm{\τ},l)=\frac{c}{u(p_t,\mathrm{\τ},l)}.$

7. a kind of large-scale multiple antennas network energy efficiency resource allocation methods with incomplete csi according to claim 1, its It is characterised by: in step 7, acquisition optimum efficiency corresponding optimum transmit power such as following formula:

$p_t^{*}=\frac{1}{2}\frac{\sqrt{{(x_3-x_4)}^2{x}_2^2+4\underset{i=1}{\overset{5}{\π}}{x}_i}-x_3{x}_2-x_4{x}_2}{x_3{x}_4}$

Wherein, χ₁～χ₅It is respectively as follows:

$x_1=\mathrm{\η}\mathrm{\α}\left|\right|\hat{h}|{|}^2\mathrm{\τ}(l+lln\frac{l}{n}),x_2=(t-\mathrm{\τ}){\mathrm{\σ}}_n^2,x_3=\mathrm{\η}\mathrm{\α}|\left|\hat{h}\right|{|}^2{{\mathrm{\τ\σ}}_e}^2,$

$x_4=\mathrm{\η}\mathrm{\α}\left|\right|\hat{h}|{|}^2\mathrm{\τ}(l+lln\frac{l}{n})+\mathrm{\η}\mathrm{\α}|\left|\hat{h}\right|{|}^2{{\mathrm{\τ\σ}}_e}^2,x_5=\frac{x_2^2+x_2{x}_3{p}_t+x_4{x}_2{p}_t+x_4{x}_3{p}_t^2}{x_1{x}_2}.$