CN101835265A - Resource optimal distribution method of non-regenerative cooperative multiband ultra-wideband system - Google Patents

Abstract

The invention discloses a resource optimal distribution method of a non-regenerative cooperative multiband ultra-wideband system, which is characterized in that: firstly, a closed expression of signal to noise ratio and symbol error rate (SER) of the system is deduced through the performance analysis on the non-regenerative cooperative multiband ultra-wideband system; on the basis of the closed expressions of the SER and the upper limit thereof, two viable resources of a sub-band and power distribution are distributed by using a resource optimal distribution algorithm proposed according to an optimization object of sending total power minimization and coverage area maximization, and a derivative of the upper limit of the symbol rate SER is inducted into the optimization object realization process based on a convex optimization theory; meanwhile, a method for solving to a univariate cubic equation is applied to the mathematical reasoning of the resource optimal distribution algorithm; and finally a sub-optimal resource utilization way is obtained. The distribution method ensures that the performance of the SER is improved, the total power consumption of the system can be reduced, and the coverage area is far longer than communication distance of a non-cooperative system.

Description

The resource optimal distribution method of non-regenerative cooperative multiband ultra-wideband system

Technical field

The present invention relates to a kind of cooperative multiband ultra-wideband communication system, relate in particular to a kind of resource optimal distribution method of non-regenerative cooperative multiband ultra-wideband system.

Background technology

It is RRM that resource optimization distributes, and is meant by the various resources in the wireless communication system (channel, power, frequency spectrum, speed etc.) being implemented effectively management, realizes set user class target or system-level target.For radio ultra wide band system, frequency spectrum and power division are the bottlenecks of design.Enough wide bandwidth safeguard frequency spectrum resource, so the distribution of power is had higher requirement.

The UWB system development a wireless channel with Gigahertz high power capacity, be particularly suitable for the place that indoor multi-path dense etc. needs high speed transmission data.The FCC of FCC has issued the preliminary standard of super broad band radio communication in 2002, stipulated that the present actual spendable spectral range of super broad band radio communication is 3.1-10.6GHz, and average transmit power is no more than-41.3dBm/MHz.Multi-band OFDM scheme (MB-OFDM) is as the best solution of UWB.It is divided into a lot of subbands to the UWB frequency spectrum, each subband 500MHz based on many band OFDM modulation techniques.Because radio ultra wide band system is the strict constrained system of transmitting power, how under the limited transmitting power condition, to realize farther transmission range and better transmission property, be the problem that the UWB system design faces.

Once there was a large amount of research that the MIMO technology is loaded into and further improved systematic function in the UWB system.But because the restriction of UWB equipment size and power is unpractiaca transmitting and receiving the many antennas of end device.Collaboration diversity is to develop a kind of rapidly diversity technique in recent years.It is based on the relay transmission principle, utilizes the cooperation between the user to reach space diversity gain, is the conventional means of antagonism multipath fading.So, can assist miscellaneous equipment to carry out message transmission the UWB equipment of idle condition as relaying owing to seldom the situation that equipment moves work simultaneously can occur in family and the office environment.Cooperative diversity technique is incorporated in the radio ultra wide band system, has crucial meaning.

And MB-OFDM radio ultra wide band system channel model reaches the restriction of having compared strictness all having done aspect speed and the power.How carry out resource allocation between the collaboration user and between each sub-band of radio ultra wide band system will be significant to improving systematic function.So be necessary very much the cooperative ultra-wide band system is carried out the research of resource allocation efficiently.At present the research about cooperative ultra-wide band system resource optimized distribution is not a lot, correlative study for example, the resource allocation research of IR-UWB system has been studied carefully by Mongolian Wuhan Research Institute; People such as W.P.Siriwongpairat have proposed the resource allocation methods that distributes speed, sub-band and power of uniting of regenerative cooperative radio ultra wide band system; Chen Fangni has proposed the method that a kind of sub-band under the regenerative repeating collaboration mode and power combined optimization distribute on this basis.IEEE 802.15.3a working group adopts time division multiple access (TDMA) to carry out user's access to pointing out to utilize time-frequency code (TFC) to distinguish different piconet (Piconet) in the physical layer of MB-OFDM UWB system and the motion of MAC layer in the identical piconet.The time-frequency code sequence is fixed in motion, and the subband allocation of promptly stipulating each user is changeless, rather than according to the channel condition flexible allocation, each user's power division also is identical.Obviously, the decline difference that each user experiences on different sub-bands, therefore according to channel condition and systematic function needs, distributing radio resource comes further conserve network resources to be necessary neatly.Fig. 1 is common cooperative multiband ultra-wideband system RRM content.Comprise: subband allocation, power division, access control, Qos demand, cooperation partner selection.The via node of non-regenerative relaying pattern does not need decoding, and the advantage with low complex degree is particularly suitable for the double bounce relay system.So in multimedia communication, non-regenerative cooperative multiband UWB system is a kind of important traffic model.Because the resource allocation methods of non-regenerative cooperative radio ultra wide band system is comparatively complicated, the research in this respect of rare always bibliographical information.

Summary of the invention

The present invention is directed to the proposition of above problem, and develop a kind of resource optimal distribution method of non-regenerative cooperative multiband ultra-wideband system.The technical solution used in the present invention is as follows:

A kind of resource optimal distribution method of non-regenerative cooperative multiband ultra-wideband system is characterized in that

At first, by non-regenerative cooperative multiband ultra-wideband system is carried out performance evaluation, derive the signal to noise ratio of this system and the closed expression of error sign ratio SER;

Secondly, on the basis of the closed expression of error sign ratio SER and the upper limit thereof, sub-band and two variable resources of power division are minimized the resource optimization allocation algorithm that proposes with the maximized optimization aim of coverage and distribute according to sending gross power; According to information theory and network theory above-mentioned optimization aim and variable are carried out modeling respectively;

Then, utilize the suboptimization strategy of proceeding in two phases to be optimized: the first step, adopt the suboptimum distribution method based on greedy algorithm to be optimized at above-mentioned optimization aim to sub-band; Second step, at above-mentioned optimization aim to power division based on protruding optimum theory, the derivative of the error sign ratio SER upper limit is updated in the optimization aim implementation procedure, utilizes the method for finding the solution simple cubic equation to be applied in the mathematical derivation of resource optimization allocation algorithm of the present invention simultaneously;

Finally, the suboptimization strategy of walking by above-mentioned two steps finally obtains the utilization of resources mode of suboptimization.

The closed expression of described error sign ratio SER is:

A wherein _NsBe the element in the allocation of subbands matrix, and

Wherein N represents the user, and s represents frequency band; b _nThe coefficient factor during for user's different rates, and

$b_n=\left\{\begin{array}{cc}2& R_n\&le;80\mathrm{<figure-callout\; id="1"\; label="Mbps"\; filenames="HSA00000067345300061.png"\; state="\{\{state\}\}">Mbps</figure-callout>}\\ 1& 80\mathrm{Mbps}<R_n<200\mathrm{<figure-callout\; id="1"\; label="Mbps"\; filenames="HSA00000067345300061.png"\; state="\{\{state\}\}">Mbps</figure-callout>}\\ 1/2& R_n>200\mathrm{Mbps}\end{array}\right\}$

δ wherein _{S, d} ^s(n), δ _{S, r} ^s(n), δ _{R, d} ^s(n) expression expression nth user centering respectively, source to destination node, source to relaying and via node to destination node average channel gain on s subband; P _S1kAnd P _S2kBe subcarrier k, the transmitting power of the middle double bounce of k '; σ in addition ²Be channel noise power.

The optimization mathematical expression of described sub-band is:

A wherein _NsBe the element in the allocation of subbands matrix; P _n ^sBe the equivalent power consumption on each subcarrier; G _n ^sThe equivalent channel gain of expression user n on sub-band s; R (ε, R _n) for the user to n for reaching R _nRate requirement and the lowest signal-to-noise demand of error rate demand when being ε; Finally obtain P _Nmin ^sBe that each user selects its power consumption minimum in each sub-frequency bands.

Owing to adopted technique scheme, the resource optimal distribution method of non-regenerative cooperative multiband ultra-wideband system provided by the invention, be on the basis of trying to achieve the error sign ratio SER upper limit, proposed respectively to realize on minimum power and the minimum power basis that the coverage maximum turns to sub-band, the optimized power allocation method of target, and the method that will find the solution simple cubic equation is applied in the mathematical derivation and proposes the effective ways that resource optimization distributes, after optimizing by this method.By the emulation experiment digital proof under radio ultra wide band system power limited condition, after using the resource optimization allocation algorithm of cooperative diversity technique and proposition, error rate of system SER performance makes moderate progress, the total power consumption of system consumption can reduce 3-4dBm, coverage has proved that considerably beyond the communication distance of non-cooperative system this optimized distribution method is applied to the superiority aspect performance improvement in the cooperative ultra-wide band system.

Description of drawings

Fig. 1 is common cooperative multiband ultra-wideband system RRM content;

The cooperative multiband ultra-wideband system resource optimization allocation flow figure that Fig. 2 proposes for the present invention;

Fig. 3 is the cooperative multiband ultra-wideband system model;

Situation comparison sheet when Fig. 4 gets different value for optimum distribution parameter in the embodiment of the invention;

Fig. 5 is the system's total power consumption after the resource optimization of target distributes and the comparison diagram of fixed allocation power for adopting with the minimum power;

The performance of error sign ratio SER compared (δ when Fig. 6 was optimum distribution power and power-sharing _Sd=1, δ _Sr=1, δ _Rd=10);

Fig. 7 is the resource allocation result that obtains among the substitution optimum distribution result when getting different value for signal to noise ratio.

Embodiment

Be illustrated in figure 2 as the processing procedure of the method that the non-regenerative cooperative multiband ultra-wideband system resource optimization distributes: i.e. derivation system signal to noise ratio at first, the closed expression of error sign ratio SER and the upper limit thereof have proposed respectively then to realize on minimum power and the minimum power basis that the coverage maximum turns to sub-band, the optimized power allocation method of target.Derivative with the error sign ratio SER upper limit is updated in the optimization aim implementation procedure of using method of Lagrange multipliers then, simultaneously in the mathematical derivation of this resource optimization allocation algorithm, also be applied to the method for finding the solution simple cubic equation, and finally obtained the utilization of resources mode of optimization.Following content is exactly the correlation step of specifically setting forth in the flow chart.

As shown in Figure 3, the cooperative ultra-wide band system is made up of source node S, via node R and destination node D.Multiband ultra-wideband system has S frequency band, has k subcarrier to use the OFDM modulation on each frequency band, can allow N that the user is communicated simultaneously.Source and via node carry out relaying with the form of subcarrier, and the information among the subcarrier k ' relaying subcarrier k is called subcarrier k and k ' pairing, make subcarrier k for simplifying unified note here.Among Fig. 3

Represent n respectively _iIndividual user's centering, source node-via node, source node-destination node, via node-destination node are at the k of s sub-frequency bands _iChannel frequency response on the individual subcarrier.

The topmost characteristics of multi-band OFDM scheme are to have adopted time-frequency (TFI) technology that interweaves.Be that the OFDM symbol can carry out the staggered transmission of time domain, frequency domain on a plurality of subbands in the same TFC frame.And utilize different TFC to select and distinguish.It is research object but not subcarrier that the This document assumes that user selects with the subband, and promptly the user has selected this subband and utilized whole subcarriers in this subband.Suppose that the time-frequency code cycle is two OFDM symbol times, definition allocation of subbands matrix A (N * s-matrix) shows that the user sends the OFDM symbolic number at a time-frequency code in the cycle on S subband, and the element in the matrix is a _NsAnd suppose that the sub-carrier power in the identical frequency band is evenly distributed, source and via node carry out relaying with the form of subcarrier, and the information among the subcarrier k ' relaying subcarrier k is called subcarrier k and k ' pairing.N represents the user herein, and k represents subcarrier, and s represents frequency band, σ ²Be channel noise power.

As seen $a_{\mathrm{ns}}\&Element;\left\{\begin{array}{c}\left\{\mathrm{0,1}\right\}{R}_n\&le;200\mathrm{Mbps}\\ \left\{\mathrm{0,1,2}\right\}{R}_n>200\mathrm{Mbps}\end{array}\right.$ $\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}=2,n=\mathrm{1,2},...,N,$ $\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}\&le;2,s=\mathrm{1,2},...,S$

Adopt non-regeneration (AF) cooperating relay process to be divided into two stages.Suppose subcarrier k, the transmitting power of the middle double bounce of k ' is respectively P _S1kAnd P _{S2k '}The information that two stages is received at the destination node place adopts high specific to merge (MRC), and the signal after the merging is: Y=ω ₁Y _S1k+ ω ₂Y _{S2k '}, ω wherein ₁And ω ₂It is merge coefficient.Its value is directly proportional with the average signal-to-noise ratio of corresponding route, and adopts maximum likelihood method to detect judgement.Instant signal to noise ratio according to MRC criterion: MRC output is the instant signal to noise ratio sum in independent path

When high s/n ratio, the nth user is at a time-frequency code cycle inner receiver average signal-to-noise ratio in the non-regeneration MB-OFDM-UWB system:

$\stackrel{\&OverBar;}{{\mathrm{\&Gamma;}}_n}=b_n\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}(\frac{P_{s1k}{\mathrm{\&delta;}}_{s,d}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}+\frac{\frac{P_{s1k}{\mathrm{\&delta;}}_{s,r}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}\&CenterDot;\frac{P_{s2k}{\mathrm{\&delta;}}_{r,d}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}}{\frac{P_{s1k}{\mathrm{\&delta;}}_{s,r}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}+\frac{P_{s2k}{\mathrm{\&delta;}}_{r,d}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}+1})$

Wherein, b _nThe coefficient factor during for user's different rates;

Expression expression nth user centering respectively, source to destination node, source to relaying and via node to destination node average channel gain on s subband;

Channel is the conditional code error probability that adopts the MPSK modulation under the condition of H and the pass of instant signal to noise ratio r is:

$P_{e\left\{h_{\mathrm{sd},}{h}_{\mathrm{sr},}{h}_{\mathrm{rd}}\right\}}=\mathrm{\&Psi;}\left(r\right)=\frac{1}{\mathrm{\&pi;}}{\&Integral;}_0^{(M-1)\mathrm{\&pi;}/M}\exp (-\frac{{\sin }^2(\mathrm{\&pi;}/4)r}{{\sin }^2\mathrm{\&theta;}})\mathrm{d\&theta;}$

When adopting the QPSK modulation to get M=4 here, all equate so suppose the power of the subcarrier in the sub-frequency bands.When big signal to noise ratio, remove 1 in the signal to noise ratio expression formula denominator, following formula is the average SER that on average obtains system respectively is on the channel of correspondence:

$P_e=\frac{1}{\mathrm{\&pi;}}{\&Integral;}_0^{3\mathrm{\&pi;}/4}\exp (-\frac{b_n\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}{P}_{s1k}{\mathrm{\&delta;}}_{s,d}^s+b_n\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}\frac{P_{s1k}{P}_{s2k}{\mathrm{\&delta;}}_{s,r}^s{\mathrm{\&delta;}}_{r,d}^s}{P_{s1k}{\mathrm{\&delta;}}_{s,r}^s+P_{s2k}{\mathrm{\&delta;}}_{r,\mathrm{<figure-callout\; id="2"\; label="d\; s"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">d</figure-callout>}}^s}}{2{\mathrm{\&sigma;}}^2{\sin }^2\mathrm{\&theta;}})\mathrm{d\&theta;}$

If

Then the closed solutions of MB-OFDM-UWB system average symbol error probability SER is:

$P_{e-\mathrm{cooperative}}=F\left((1+\frac{b_n\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}{P}_{s1k}{\mathrm{\&delta;}}_{s,\mathrm{<figure-callout\; id="2"\; label="d\; s"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">d</figure-callout>}}^s}{2{\mathrm{\&sigma;}}^2{\mathrm{<figure-callout\; id="2"\; label="sin"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">sin</figure-callout>}}^2\mathrm{\&theta;}})(1+\frac{b_n\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}\frac{P_{s1k}{P}_{s2k}{\mathrm{\&delta;}}_{s,d}^s{\mathrm{\&delta;}}_{r,d}^s}{P_{s1k}{\mathrm{\&delta;}}_{s,d}^s+P_{s2k}{\mathrm{\&delta;}}_{r,\mathrm{<figure-callout\; id="2"\; label="d\; s"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">d</figure-callout>}}^s}}{2{\mathrm{\&sigma;}}^2{\sin }^2\mathrm{\&theta;}})\right)$

Find the following formula more complicated, consider situation in a frequency band, and 1 in the denominator removed as hypothesis, desirable its upper limit, so abbreviation is:

$P_{e-\mathrm{upper}}=\frac{1}{\mathrm{\&pi;}}{\&Integral;}_0^{\frac{3}{4}\mathrm{\&pi;}}\frac{4{\mathrm{\&sigma;}}^4{\sin }^4\mathrm{\&theta;}\&CenterDot;(P_{s1k}{\mathrm{\&delta;}}_{\mathrm{sr}}+P_{s2k}{\mathrm{\&delta;}}_{\mathrm{rd}})}{b_n^2{P}_{s1k}^2{P}_{s2k}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{sr}}{\mathrm{\&delta;}}_{\mathrm{rd}}}\mathrm{d\&theta;}$

$=\frac{4{\mathrm{\&sigma;}}^4(P_{s1k}{\mathrm{\&delta;}}_{\mathrm{sr}}+P_{s2k}{\mathrm{\&delta;}}_{\mathrm{rd}})}{b_n^2{P}_{s1k}^2{P}_{s2k}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{sr}}{\mathrm{\&delta;}}_{\mathrm{rd}}}(\frac{9}{32}+\frac{1}{4\mathrm{\&pi;}})=\frac{{\mathrm{\&sigma;}}^4(P_{s1k}{\mathrm{\&delta;}}_{\mathrm{sr}}+P_{s2k}{\mathrm{\&delta;}}_{\mathrm{rd}})}{b_n^2{P}_{s1k}^2{P}_{s2k}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{sr}}{\mathrm{\&delta;}}_{\mathrm{rd}}}(\frac{9}{8}+\frac{1}{\mathrm{\&pi;}})$

The channel model of the ultra broadband standard proposals among the present invention and parameter reference and employing " CASSIOLI D; WIN M Z; VATALARO F; et al.Performance of low-complexity Rake reception in a realistic UWB channel.IEEE ICC[C]; New York:IEEE Press; 2002:763-767P ", guaranteed that theory analysis and Simulation Application are to actual reliability.

Here the power of supposing each subcarrier in the same sub-band is identical, and the nth user is on the s sub-frequency bands, and the power of source node and via node is designated as P respectively _N1 ^s, P _N2 ^sConsider that at first the cooperative ultra-wide band system under the non-regenerative relaying is the resource allocation of target with the minimum power, optimization problem is defined as:

$\min P=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}K(P_{n1}^s+P_{n2}^s)$

$s.t.\left\{\begin{array}{c}\stackrel{\&OverBar;}{{\mathrm{\&Gamma;}}_n}=b_n\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}(\frac{P_{n1}^s{\mathrm{\&delta;}}_{s,d}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}+\frac{\frac{P_{n1}^s{\mathrm{\&delta;}}_{s,r}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}\&CenterDot;\frac{P_{n2}^s{\mathrm{\&delta;}}_{r,d}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}}{\frac{P_{n1}^s{\mathrm{\&delta;}}_{s,r}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}+\frac{P_{n2}^s{\mathrm{\&delta;}}_{r,d}^s\left(n\right)}{{\mathrm{\&sigma;}}^2}})\&GreaterEqual;\mathrm{\&gamma;}(\mathrm{\&epsiv;},R_n),\&ForAll;n\\ \underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}=2,n=\mathrm{1,2},...,N\\ \underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}\&le;2,s=\mathrm{1,2},...,S\end{array}\right.$

A subband allocation algorithm

Is a user's direct transmission course with a user to equivalence, P in the collaboration communication process that the nth user is right _n ^sBe defined as the equivalent power consumption on each subcarrier, but not actual power consumption can be carried out the distribution of sub-band with this.When channel gain satisfies rayleigh distributed, under the AF trunking scheme, can use H _e ^ARepresent equivalent channel gain.Suppose that the channel gain of user on each subcarrier of same sub-band all equates.Here use G _n ^sThe equivalent channel gain of expression user n on sub-band s.

Define r (ε, R simultaneously _n) for the user to n for reaching R _nRate requirement and the lowest signal-to-noise demand of error rate demand when being ε.Then

$\stackrel{\&OverBar;}{{\mathrm{\&Gamma;}}_n}=b_n\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}{P}_n^s{G}_n^s\&GreaterEqual;r(\mathrm{\&epsiv;},R_n)$

The user is to the suboptimum allocation algorithm based on greedy algorithm of chooser frequency band so:

(1) the unappropriated user's collection of definition and initialization is N _Live=1,2 ..., and N}, unappropriated subband sets is S _Live=1,2 ..., S}, allocation matrix are A={a _NsInitialization $a_{\mathrm{ns}}=0,\&ForAll;n,\&ForAll;s.$

(2) suppose that all users can arbitrarily take any sub-band.Equivalent power consumption when user n takies sub-band s is:

$P_n^s=r(\mathrm{\&epsiv;},R_n)/b_n\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ns}}{G}_n^s$

(3) each user all select its in each sub-frequency bands the power consumption minimum be designated as P _Nmin ^s, then at all P _Nmin ^sSelect peaked that user priority to carry out subband allocation, i.e. the user of power consumption maximum elder generation chooser frequency band, thus at utmost reduce total power consumption.

(4) repeating step (3) is brought in constant renewal in A={a _Ns, until

The B minimum power is the resource allocation algorithm of target

The user to the distribution of having carried out sub-band after, carrying out with the minimum power in a subcarrier based on the SER expression formula of deriving previously is the power division of source node and via node of target.Satisfy minimize under the situation that the right error rate of each user requires the right power of each user and, promptly finished the minimized purpose of gross power.Definition P ₁, P ₂, P represents that respectively a user is to the source node power on a subcarrier, via node power, gross power.Optimization problem is defined as:

minP＝P ₁+P ₂

$s.t.\left\{\begin{array}{c}{P}_e\&le;\mathrm{\&epsiv;}\\ {\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HSA00000067345300061.png"\; state="\{\{state\}\}">P</figure-callout>}}_1\&le;{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HSA00000067345300061.png"\; state="\{\{state\}\}">P</figure-callout>}}_{1\max },{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">P</figure-callout>}}_2\&le;{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">P</figure-callout>}}_{2\max }\end{array}\right.$

Simple in order to calculate, at first consider first constraints, and then carry out verification with second constraints.Use method of Lagrange multipliers, the minimum power assignment problem can be converted to:

Making source node power ratio in total power consumption is r, i.e. P ₁=rP, P ₂=(1-r) P,

And bring the upper limit of error sign ratio into following formula and carry out abbreviation and calculate: $P=\frac{2r{\mathrm{\&delta;}}_{\mathrm{sr}}+2(1-r){\mathrm{\&delta;}}_{\mathrm{rd}}}{r{\mathrm{\&delta;}}_{\mathrm{sr}}}$

Simple cubic equation about r:

$r^3(8\mathrm{\&epsiv;}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="rd"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">rd</figure-callout>}}^2{\mathrm{\&delta;}}_{\mathrm{sr}}-4\mathrm{\&epsiv;}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{rd}}{\mathrm{\&delta;}}_{\mathrm{sr}}^2-4\mathrm{\&epsiv;}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{rd}}^3)+r^2(4\mathrm{\&epsiv;}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{rd}}{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="sr"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">sr</figure-callout>}}^2+12\mathrm{\&epsiv;}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{rd}}^3-16\mathrm{\&epsiv;}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="rd"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">rd</figure-callout>}}^2{\mathrm{\&delta;}}_{\mathrm{sr}})$

$+r(8\mathrm{\&epsiv;}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="rd"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">rd</figure-callout>}}^2{\mathrm{\&delta;}}_{\mathrm{sr}}-12\mathrm{\&epsiv;}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{rd}}^3-B{\mathrm{\&delta;}}_{\mathrm{rd}}{\mathrm{\&delta;}}_{\mathrm{sr}}+B{\mathrm{\&delta;}}_{\mathrm{sr}}^2)+4\mathrm{\&epsiv;}{\mathrm{\&delta;}}_{\mathrm{sd}}{\mathrm{\&delta;}}_{\mathrm{rd}}^3-B{\mathrm{\&delta;}}_{\mathrm{rd}}{\mathrm{\&delta;}}_{\mathrm{sr}}=0$

By simple cubic equation ax ³+ bx ²+ cx+d=0 separate for:

$x=-\frac{b}{3a}+\sqrt[3]{-\frac{27a^{2}d-9\mathrm{abc}+2b^3}{54a^3}+\sqrt{{\left(\frac{27a^{2}d-9\mathrm{abc}+2b^3}{54a^3}\right)}^2}+{\left(\frac{3\mathrm{ac}-{\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">b</figure-callout>}}^2}{9a^2}\right)}^3}$

$+\sqrt[3]{-\frac{27a^{2}d-9\mathrm{abc}+2b^3}{54a^3}-\sqrt{{\left(\frac{27a^{2}d-9\mathrm{abc}+2b^3}{54a^3}\right)}^2}+{\left(\frac{3\mathrm{ac}-{\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">b</figure-callout>}}^2}{9a^2}\right)}^3}$

The value of r and P can be tried to achieve by the numerical computation method in the matlab, then P can be obtained respectively ₁, P ₂, P value.Second constraints is carried out verification in using then.The special δ that works as _Sd=δ _Sr=δ _RdThe time, can get: $r=4\mathrm{\&epsiv;}-\frac{B}{{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="sd"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">sd</figure-callout>}}^2},$ $P=\frac{2}{4\mathrm{\&epsiv;}-\frac{B}{{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="sd"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">sd</figure-callout>}}^2}}$

By the numerical computation method of matlab, the numerical value of resource allocation contrast among substitution optimum distribution result when Fig. 4 gets different value for channel parameter.

By the matlab simulation analysis, adopting with the minimum power is the system's total power consumption after the resource optimization of target distributes and the comparison of fixed allocation power.Along with the increase of speed, system's total power consumption also increases, and adopts the situation power consumption of dynamic assignment will lack 3-4dBm.The advantage of the resource allocation methods that the proof patent proposes.

With the maximum coverage is the resource allocation algorithm of target

Maximum coverage refers to the distance between maximization source node and the destination node.How to carry out with the maximum coverage on the basis of total power consumption be the resource allocation of target minimizing in cooperative ultra-wide band system under the non-regenerative relaying.

According to the characteristic of radio ultra wide band system, suppose that the multipath energy of transmission signals and transmission range are proportional relations, that is:

Here get κ=2 under the indoor environment, v=2.During the coverage maximization, via node one fixes on the line of source node and destination node.Here the distance that defines source node and destination node is D _{S, d}, establishing via node proportion in the distance of source and destination node is u, i.e. D _{S, r}=uD _{S, d}, D _{R, d}=(1-u) D _{S, d}, the gross power of source node and destination node is P, is the P that tries to achieve above.

Make via node and destination node power be respectively P ₁=rP, P ₂=(1-r) P.Then the resource optimization problem is:

maxD _s，d

$s.t.\left\{\begin{array}{c}{P}_e\&le;\mathrm{\&epsiv;}\\ \mathrm{rP}\&le;{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HSA00000067345300061.png"\; state="\{\{state\}\}">P</figure-callout>}}_{1\max },(1-r)P\&le;{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">P</figure-callout>}}_{2\max },0<r<1\end{array}\right.$

Use method of Lagrange multipliers, the minimum power assignment problem can be converted to: $\left\{\begin{array}{c}1+\mathrm{\&lambda;}\frac{\&PartialD;P_e}{\&PartialD;D_{\mathrm{sd}}}=0\\ 1+\mathrm{\&lambda;}\frac{\&PartialD;P_e}{\&PartialD;r}=0\\ \frac{\&PartialD;P_e}{\&PartialD;u}=0\\ P_e-\mathrm{\&epsiv;}=0\end{array}\right.$

Just the upper limit of error sign ratio is brought following formula into and is carried out that abbreviation calculates and by separating following simple cubic equation, can obtain u, D _Sd, the r value.

$\left\{\begin{array}{c}x=\frac{-u}{2u}\\ D_{\mathrm{sd}}=\frac{{\mathrm{xu}}^{-2}-{(1-u)}^{-2}{x}^{-2}+u^{-1}{(1-u)}^{-1}}{x}\\ r=\frac{-u+{\mathrm{<figure-callout\; id="2"\; label="u"\; filenames="HSA00000067345300051.png,HSA00000067345300061.png"\; state="\{\{state\}\}">u</figure-callout>}}^2}{1-2u-u^2(1-2u+u^2)D_{\mathrm{sd}}^2}\\ u^3(2\mathrm{\&epsiv;B}+4B-16)+u^2(-8B+14-2\mathrm{\&epsiv;B})+u(5B-6)+(1-B)=0\end{array}\right.$

Fig. 5 is the system's total power consumption after the resource optimization of target distributes and the comparison of fixed allocation power for adopting with the minimum power.As seen along with the increase of speed, system's total power consumption also increases, and adopts the situation power consumption of dynamic assignment will reduce 3-4dBm.

Fig. 6 is that channel gain is got δ respectively after adopting minimum power to be the resource allocation algorithm of target _Sd=1, δ _Sr=1, δ _RdCalculate P in=10 o'clock substitution optimal solutions ₁=0.6194P divides equally power P with adopting source node and via node ₁Though visible two curves are more or less the same during=0.5P, systematic function is better than mean allocation after the employing optimum distribution.

Fig. 7 is under ultra broadband CM4 channel model, among the substitution optimum distribution result, uses the resource allocation result that Matlab calculates when signal to noise ratio is got different value, comprises power division, relaying position and maximum coverage.Because the communication distance of CM4 channel model is 4～10m, can as seen use cooperative diversity technique from table 1 after coverage increase a lot.And under the situation of high s/n ratio, the position of via node is more and more close to mid point, and the power that source node distributes also more and more approaches half of gross power.

The above; only be the preferable embodiment of the present invention; but protection scope of the present invention is not limited thereto; anyly be familiar with those skilled in the art in the technical scope that the present invention discloses; be equal to replacement or change according to technical scheme of the present invention and inventive concept thereof, all should be encompassed within protection scope of the present invention.

Claims (3)

1. the resource optimal distribution method of a non-regenerative cooperative multiband ultra-wideband system is characterized in that

At first, by non-regenerative cooperative multiband ultra-wideband system is carried out performance evaluation, derive the signal to noise ratio of this system and the closed expression of error sign ratio SER;

Secondly, on the basis of the closed expression of error sign ratio SER and the upper limit thereof, sub-band and two variable resources of power division are minimized the resource optimization allocation algorithm that proposes with the maximized optimization aim of coverage and distribute according to sending gross power, according to information theory and network theory above-mentioned optimization aim and variable are carried out modeling respectively;

Then, utilize the suboptimization strategy of proceeding in two phases to be optimized: the first step, adopt the suboptimum distribution method based on greedy algorithm to be optimized at above-mentioned optimization aim to sub-band; Second step, at above-mentioned optimization aim to power division based on protruding optimum theory, the derivative of the error sign ratio SER upper limit is updated in the optimization aim implementation procedure, utilizes the method for finding the solution simple cubic equation to be applied in the mathematical derivation of resource optimization allocation algorithm of the present invention simultaneously;

Finally, the suboptimization strategy of walking by above-mentioned two steps finally obtains the utilization of resources mode of suboptimization.

2. method according to claim 1 is characterized in that the closed expression of described error sign ratio SER is:

A wherein _NsBe the element in the allocation of subbands matrix, and

Wherein N represents the user, and s represents frequency band; b _nThe coefficient factor during for user's different rates, and $b_n=\left\{\begin{array}{cc}2& R_n\&le;80\mathrm{Mbps}\\ 1& 80\mathrm{Mbps}<R_n<200\mathrm{Mbps}\\ 1/2& R_n>200\mathrm{Mbps}\end{array}\right\}$

δ wherein _{S, d} ^s(n), δ _{S, r} ^s(n), δ _{R, d} ^s(n) expression expression nth user centering respectively, source to destination node, source to relaying and via node to destination node average channel gain on s subband; P _S1kAnd P _S2kBe subcarrier k, the transmitting power of the middle double bounce of k '; σ in addition ²Be channel noise power, θ is an integration variable.

3. method according to claim 1 is characterized in that the optimization mathematical expression of described sub-band is:

A wherein _NsBe the element in the allocation of subbands matrix; P _n ^sBe the equivalent power consumption on each subcarrier; G _n ^sThe equivalent channel gain of expression user n on sub-band s; R (ε, R _n) for the user to n for reaching R _nRate requirement and the lowest signal-to-noise demand of error rate demand when being ε; Finally obtain P _Nmin ^sBe that each user selects its power consumption minimum in each sub-frequency bands.

Images