CN102355672A - Method for allocating adaptive resources in cognitive OFDM (orthogonal frequency division multiplexing) system - Google Patents

Abstract

The invention provides a method for allocating adaptive resources in a cognitive OFDM (orthogonal frequency division multiplexing) system, and the method is characterized in that a cognitive radio can solve the problem of a shortage of radio frequency spectrum resources through improving the utilization rate of the frequency spectrum resources; and the orthogonal frequency division multiplexing (OFDM) technology is considered as a technology which can be perfectly combined with a cognitive system. The method comprises the following steps: starting from the maximization of the overall transmission bit rate of secondary users, under the limitation of the overall transmission power of the secondary users and the interference thresholds of primary users, and base on a new constructed cost function, carrying out subchannel allocation by using a Hungarian algorithm; then, according to the needs of the secondary users, carrying out adaptive bit and power allocation. By using the method provided by the invention, spectrum holes among frequency bands of different primary users can be utilized effectively, so that the throughput of a secondary user system is maximized, and the overall interference capacity of each primary user is less than the interference threshold of each primary user.

Description

Adaptive resource allocation method in the cognitive ofdm system

Technical field

The present invention relates to a kind of be used in particular in the cognitive radio networks and belong to communication technical field based on the resource allocation implementation of OFDM.

Background technology

Along with the fast development of wireless communication technology, radio spectrum resources has become the indispensable precious resources of information-intensive society.Yet the data that provide according to FCC (FCC); The availability of frequency spectrum of having distributed but has only 15%-85%; This is because existing frequency spectrum licensing scheme adopts fixing spectrum allocation may, makes the availability of frequency spectrum low, thereby causes a large amount of frequency spectrum resource wastes; Therefore in wireless communication technology, frequency spectrum inserts and the reasonability of distributing seems more important than the physics of frequency spectrum is rare.And cognitive radio technology is considered to solve the low best solution of wireless frequency spectrum utilance.Cognitive radio (Cognitive Radio; CR) technology proposes the framework that a kind of dynamic spectrum is shared; Allow cognitive user (Cognitive User; CU) do not influence authorized user (Primary User PU) utilizes a large amount of idle frequency spectrums intelligently under the prerequisite of work, dynamically carry out spectrum allocation may and whenever and wherever possible, the communication of high reliability.Therefore studying dynamic channel allocation and access technology under the cognitive radio environment seems and more is of practical significance.

Hence one can see that, and the self-adapting data transmission technology is that cognitive radio system utilizes idle frequency spectrum to communicate, thus one of whole key technology that improves the availability of frequency spectrum.Because cognitive radio system possibly be operated in the frequency band of non-constant width, and frequency spectrum is generally discontinuous, can this specific character of self adaptation so data transfer mode is necessary.OFDM (Orthogonal Frequency Division Multiplexing) is owing to can dynamically utilize frequency spectrum resource; And load according to subchannel characteristic self adaptation; Can take situation to the authorized user frequency spectrum again and carry out rapid analysis, have availability of frequency spectrum height, the anti-multipath effect capability is strong; Can control and utilize resources such as frequency spectrum, power flexibly, in the system of broadband wireless communication design, obtain extensive use.OFDM is applied in the cognitive radio,, can easily realizes link-adaptive transmission through each subcarrier is dynamically selected modulation and encoding scheme according to its different separately channel condition.

The technological radio communication just of OFDM (OFDM) transmits the key technology of two-forty, wide band multimedia service in the limited wireless frequency range.Along with improving constantly of data rate, the performance of high-speed data communication system not only receives noise limit, and main influence comes from the intersymbol interference that the time delay extended attribute of wireless channel causes.This intersymbol interference mainly is owing to exist many different radio propagation path of time delay to cause between the transmitter and receiver.Generally speaking, as long as time delay expansion is far smaller than the cycle of sending symbol, then the influence that causes of intersymbol interference almost can be ignored.For high-speed data service, send the cycle of symbol and can compare mutually, even less than the time delay expansion, will introduce serious intersymbol interference this moment, cause systematic function sharply to descend with the time delay expansion.

The OFDM basic principle is exactly the rate data streams that data flow is at a high speed resolved into multidiameter delay, on a plurality of carrier waves, transmits.For the parallel subcarrier of low rate, because the symbol period broadening, the time delay expansion that multipath effect causes diminishes relatively, in each OFDM symbol, inserts certain protection after the time, and intersymbol interference almost can be ignored.

Self adaptation assignment problem in the tradition ofdm system; Generally be to given transmitted power and transmission bandwidth; Under the condition of (QoS) required BER that guarantees service quality; Temporal properties according to subchannel are the bit rate on multi-user's allocated sub-channels and the subchannel adaptively, make the power system capacity maximization.But main user often uses adjacent frequency spectrum simultaneously with time user in cognitive system, thereby can produce inevitable phase mutual interference, is an important problem so how to utilize the characteristic of OFDM technology to reduce this phase mutual interference.

Summary of the invention

Technical problem: the objective of the invention is to be to provide the adaptive resource allocation method in a kind of cognitive ofdm system; This method is from maximizing total transmitted bit speed of time user; Based on the new cost function that makes up, use improved Hungary algorithm to carry out the subchannel distribution in restriction of the total transmitted power of inferior user and main user's interference threshold restriction down; According to time user's demand, carry out adaptive bit and power division then.

Technical scheme: the present invention provides the adaptive resource allocation method in a kind of OFDM cognitive radio system, and this method makes up cost function according to channel situation; And use improved Hungary algorithm to carry out subchannel and distribute; Obtain time user's sets of sub-channels, according to time user's demand, carry out adaptive bit and power division then; To guarantee down the inferior custom system capacity of maximization in restriction of the total transmitted power of inferior user and main user's interference threshold restriction.

This method may further comprise the steps:

A. time user base station is upgraded the communication link gain of time user k in real time according to the information of perception gained Main user is to the interfering link gain of inferior user k

Inferior user k is to main user's interfering link gain

And between each subchannel and the shared frequency spectrum of main user apart from d _Nl, d wherein _NlBe n subchannel and l the distance between the shared frequency band of main user;

B. calculate k time user when the n subchannel is distributed 1 bit, the transmitted power increment of generation $\mathrm{\Δ}{P}_{k,n}=\frac{{\mathrm{\σ}}^2+{\mathrm{\Σ}}_{l=1}^L{J}_{k,n}^l}{{\left|h_{k,n}^{\mathrm{Ss}}\right|}^2}{2}^{b_{k,n}}$ And the interference increment that the main user of l is produced $\mathrm{\Δ}{I}_{k,n}^l=\mathrm{\Δ}{P}_{k,n}{\mathrm{IF}}_{k,n}^l,$ Wherein, σ ²Be the variance of additive white Gaussian noise,

Be l main user interference to inferior user k on the n subchannel, b _{K, n}The bit number that on the n subchannel, distributes for inferior user k,

Be the gain on subchannel n between t user base station of time slot and the individual time receiver user of k,

Be k user on the n subchannel to l main user's interference factor, L is main user's a number;

C. for Power Limitation and interference-limited equilibrium are considered; Transmitted power increment and interference increment are got compromise; Be defined as k user distribute 1 bit on the n subchannel cost function for

wherein α be regulatory factor, with it do to L the main user average cost function that all obtains inferior user k of making even ${\stackrel{\‾}{\mathrm{\ΔC}}}_{k,n}=\frac{\underset{l=1}{\overset{L}{\mathrm{\Σ}}}\mathrm{\Δ}{C}_{k,n}^l}{L};$

D. for K time user and N subchannel, adopt improvement Hungary algorithm to carry out subchannel and distribute, make whole custom system cost minimum, obtain the sets of sub-channels A of time user k thus _k

E. choose R _k/ β _kMinimum inferior user carries out subchannel and Bit Allocation in Discrete, wherein R _kBe the desired transmission rate of inferior user k, β _kFor the transmission rate of inferior user k acquisition, from sets of sub-channels A _kThe middle cost of selecting

Minimum subchannel distributes 1 bit, and judges whether time user's transmitted power and the suffered interference of main user surpass thresholding separately, if surpass, distribute and finish, if do not have, continue to distribute.

Beneficial effect: for the OFDM cognitive radio system; The invention provides a kind of new Dynamic Resource Allocation for Multimedia algorithm; This algorithm is when guaranteeing to satisfy total transmitted power restriction of time user and main user's interference threshold restriction, and with subchannel, power and bit reasonably distribute between inferior user; Realize time custom system maximum capacity, and satisfy time user's demand.

Description of drawings

Topological relation between Fig. 1 master user and the inferior user,

Fig. 2 master user and time user insert the distribution situation of frequency spectrum.

Embodiment

Topological relation among the present invention between main user and the inferior user is as shown in Figure 1; In the OFDM cognitive radio networks; Exist L main user and K time user; Link 1 is respectively main user to inferior user with link 3; Inferior user is to main user's interfering link; Link 2 is a time user's communications link, and

and

is respectively three gains on the link.

Main user is as shown in Figure 2 with the distribution situation that time user inserts frequency spectrum, supposes main user 1, and the frequency band that 2...L takies is respectively B ₁, B ₂..., B _LHz, and known by inferior user that inferior user can insert the both sides of spectrum distribution at the shared frequency band of main user.Inferior user adopts the OFDM transmission mechanism, can use spectrum division to become the N number of sub-carrier, and supposes that each subcarrier occupied frequency bandwidth is Δ fHz.

In cognitive system, main user and time user often use adjacent frequency spectrum simultaneously, thereby the interference Mathematical Modeling that produces is following:

A. time user is to main user's interference

The power spectral density of k time CU n number of sub-carrier can be expressed as:

${\mathrm{\φ}}_{k,n}\left(f\right)=P_{k,n}{T}_s{\left(\frac{\sin \mathrm{\πf}{T}_s}{\mathrm{\πf}{T}_s}\right)}^2---\left(1\right)$

Wherein, P _{K, n}Be the total transmitted power on k user's the n number of sub-carrier, T _sIt is an OFDM symbol duration.Thus, k time user's n number of sub-carrier can be expressed as l main user's interference:

$I_{k,n}^l=\left|h_{k,n}^{\mathrm{sp}}\right|P_{k,n}{T}_s{\∫}_{d_{\mathrm{nl}}-B_l/2}^{d_{\mathrm{nl}}+B_l/2}{\left(\frac{\sin \mathrm{\πf}{T}_s}{\mathrm{\πf}{T}_s}\right)}^2\mathrm{df}=P_{k,n}{\mathrm{IF}}_{k,n}^l---\left(2\right)$

Wherein, d _NlBe shared frequency band of n number of sub-carrier and l the distance between the shared frequency band of main user, B _lBe the frequency bandwidth of l main CU, Be the gain on subchannel n between the individual main receiver user of k time user's transmitter and l,

Be interference factor.

B. main user is to inferior user's interference

Power spectral density after main user's the signal process M rank fast Fourier transform (FFT) can be represented as follows:

Wherein, w is digital angular frequency, φ _PU(e ^Jw) be the power spectral density of main subscriber signal, main subscriber signal can produce through elliptic filter through white noise.Thus, the individual main user of l can be expressed as the interference of n number of sub-carrier:

$J_{k,n}^l={\left|h_{k,n}^{\mathrm{ps}}\right|}^2{\∫}_{d_{\mathrm{nl}}-\mathrm{\Δf}}^{d_{\mathrm{nl}}+\mathrm{\Δf}}E\left\{I_N\left(w\right)\right\}\mathrm{dw}---\left(4\right)$

Wherein, the gain on subchannel n between

l main user's transmitter and k time user's the receiver.

Suppose that time user's transmitter can well the channel perception state, k time signal noise interference ratio (SINR) of user on subchannel n can be expressed as:

${\mathrm{\γ}}_{k,n}=\frac{{\left|h_{k,n}^{\mathrm{ss}}\right|}^2}{{\mathrm{\σ}}^2+{\mathrm{\Σ}}_{l=1}^L{J}_{k,n}^l}---\left(5\right)$

Wherein, σ ²Be the variance of additive white Gaussian noise, Be the interference summation of L main user to k time user's n number of sub-carrier,

It is the gain on subchannel n between k time user's transmitter and the receiver.

According to shannon formula, the transmission rate on k time user's the n number of sub-carrier can be expressed as:

R _k，n＝Δflog ₂(1+P _k，nγ _k，n) (6)

Transmitted bit number on k time user's the n number of sub-carrier can be expressed as

In the OFDM cognitive system, in the face of two optimization restrictions, promptly time user sends gross power restriction and the acceptable interference-limited of main user.Wherein, inferior user's transmitted power is relevant with channel situation, interference and distance dependent that inferior user produces main user, and inferior user and main user's spectral distance are near more, disturb big more.When the good channel of inferior subscriber channel situation and main user's spectral distance are very near, just need carry out the equilibrium consideration to Power Limitation and interference-limited, make the throughput maximum of system.Following compromise strategy is proposed: based on the Mathematical Modeling of disturbing among the present invention; Draw respectively when a subchannel distributed a bit; The transmitted power increment that produces and to main user's interference increment; Then, with a regulatory factor with two kinds of increment weighting summations, the cost function when distributing this bit as this subchannel.Be the basis with this cost function, the present invention proposes following allocative decision: at first, carry out subchannel with Hungary's algorithm and distribute; Which then, confirm to time user's allocation bit (if demand is identical, the inferior user good to performance distributes) according to inferior user's demand percentage; Then according to cost function; Select the minimum subchannel of cost to distribute a bit, so repeat, reach threshold values or main user until the total transmitted power of inferior user and disturb and reach threshold value.Thereby satisfying under two kinds of restrictions, making time custom system throughput maximum.

The target of in the OFDM cognitive system, optimizing is to make the maximization of time user capacity, and time user's the total transmitted power of transmission is lower than threshold values simultaneously, and the suffered interference of main user is lower than threshold value separately, specifically can represent as follows:

$\max \underset{k=1}{\overset{K}{\mathrm{\Σ}}}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{k,n}{R}_{k,n}$

S.t.

$\underset{k=1}{\overset{K}{\mathrm{\Σ}}}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{k,n}{P}_{k,n}\≤P_T---\left(9\right)$

$\underset{k=1}{\overset{K}{\mathrm{\Σ}}}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{k,n}{P}_{k,n}{\mathrm{IF}}_{k,n}^l\≤I_{\mathrm{th}}^l,\∀k,n,l---\left(10\right)$

R ₁∶R ₂∶...∶R _K＝β ₁∶β ₂∶...∶β _K (11)

Wherein, ρ _{K, n}Expression channel occupancy situation, inequality (8) explain that each subchannel can only be by one CU, P _TBe total transmitted power restriction of time user, Be l main user's interference threshold, formula (11) is to be the not ratio restriction of homogeneous user's demand setting.

According to (5) and (7), when the n subchannel of k CU was distributed 1 bit, the transmitted power increment of generation can be expressed as:

$\mathrm{\Δ}{P}_{k,n}=\frac{{\mathrm{\σ}}^2+{\mathrm{\Σ}}_{l=1}^L{J}_{k,n}^l}{{\left|h_{k,n}^{\mathrm{ss}}\right|}^2}{2}^{b_{k,n}}---\left(12\right)$

According to (2) and (12), along with the increase of transmitted power, the interference increment that l main user produced can be expressed as:

$\mathrm{\Δ}{I}_{k,n}^l=\mathrm{\Δ}{P}_{k,n}{\mathrm{IF}}_{k,n}^l---\left(13\right)$

For Power Limitation and interference-limited are united consideration, we get compromise with transmitted power increment and interference increment, the cost function when obtaining that the n subchannel of k CU distributed 1 bit, that is:

$\mathrm{\Δ}{C}_{k,n}^l=\mathrm{\α\Δ}{P}_{k,n}+(1-\mathrm{\α})\mathrm{\Δ}{I}_{k,n}^l---\left(14\right)$

Wherein α is a regulatory factor.

This cost function is made even all as the foundation of resource allocation under how main user's scene to L main user:

${\stackrel{\‾}{\mathrm{\ΔC}}}_{k,n}=\frac{\underset{l=1}{\overset{L}{\mathrm{\Σ}}}\mathrm{\Δ}{C}_{k,n}^l}{L}---\left(15\right)$

A. subchannel distributes

In traditional ofdm system; Generally carry out subchannel and distribute, and in the OFDM cognitive radio system, need to consider the disturbing factor of time user main user according to signal to noise ratio (snr); So the cost that each subchannel is distributed 1 bit is as distributing foundation, that is:

$Q_{k,n}=\frac{\underset{l=1}{\overset{L}{\mathrm{\Σ}}}(\mathrm{\α}\frac{1}{{\mathrm{\γ}}_{k,n}}+(1-\mathrm{\α})\frac{1}{{\mathrm{\γ}}_{k,n}}{\mathrm{IF}}_{k,n}^l)}{L}---\left(16\right)$

The same with the method for salary distribution in traditional ofdm system; Each subchannel is distributed to the minimum user of Q value might cause a too many subchannel of CU; And another user is not assigned with the situation of subchannel; So suppose that the ratio of number of subchannels of each CU is the same with the user's request ratio, i.e. N ₁: N ₂: ...: N _K=β ₁: β ₂: ...: β _KTherefore, the number of subchannels that is assigned to of k user is:

Then, remaining subchannel is distributed to current needs maximum (k=argmin _kR _k/ β _k) the user.

Among the present invention; The target that subchannel distributes is to make the cost of whole system minimum, and the classical way that solves this optimization problem is Hungary's algorithm, but traditional Hungary algorithm can only be to the scene of N user and N subchannel; Allocation scenarios of the present invention is K time user and N subchannel, that is:

$Q=\left[\begin{array}{c}{Q}_{\mathrm{1,1}}{\mathrm{<figure-callout\; id="2"\; label="Q"\; filenames="HDA0000083970940000012.png"\; state="\{\{state\}\}">Q</figure-callout>}}_{\mathrm{1,2}}...Q_{1,\mathrm{<figure-callout\; id="2"\; label="N\; Q"\; filenames="HDA0000083970940000012.png"\; state="\{\{state\}\}">N</figure-callout>}}& \\ Q_{}{}_{\mathrm{2,1}}{\mathrm{<figure-callout\; id="2"\; label="Q"\; filenames="HDA0000083970940000012.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="Q"\; filenames="HDA0000083970940000012.png"\; state="\{\{state\}\}">Q</figure-callout>\; </figure-callout>}}_{\mathrm{2,2}}...{\mathrm{<figure-callout\; id="2"\; label="Q"\; filenames="HDA0000083970940000012.png"\; state="\{\{state\}\}">Q</figure-callout>}}_{2,N}\\ .......\\ Q_{K.1}{Q}_{K,2}...Q_{K,N}\end{array}\right]---\left(18\right)$

Therefore, the cost matrix is expanded to the size of N * N:

Use Hungary's algorithm to distribute again, at last with allocation result by reverting to K * N rank: belong to N _kThe subchannel that the row of set is assigned to is all distributed to time user k, and object lesson is described below.

Suppose to have 2 inferior users and 6 subchannel, first time CU 2 subchannel, second CU 4 subchannel, allocation result is:

$\mathrm{\&rho;}=\left[\begin{array}{cccccc}1& 0& 0& 0& 0& 0\\ 0& 1& 0& 0& 0& 0\\ 0& 0& 1& 0& 0& 0\\ 0& 0& 0& 0& 1& 0\\ 0& 0& 0& 1& 0& 0\\ 0& 0& 0& 0& 0& 1\end{array}\right]$

Then actual 2 * 6 required rank allocation matrixs should be:

$\mathrm{\&rho;}=\left[\begin{array}{cccccc}1& 1& 0& 0& 0& 0\\ 0& 0& 1& 1& 1& 1\end{array}\right]$

B. power and Bit Allocation in Discrete

The power that the present invention proposes with the basic thought of bit distribution algorithm is: confirm to which time user's allocation bit (if demand is identical according to inferior user's demand percentage (11); The inferior user good to performance distributes); Then according to (15); Minimum subchannel distributes 1 bit to select

; So repeat, reach threshold values or main user until the total transmitted power of inferior user and disturb and reach threshold value, distribute and finish.Specific algorithm is described below:

1. initiation parameter

A) each main user's interference threshold is set L ∈ 1,2 ..., L}, inferior user's transmitted power restriction P _TAnd inferior user's request ratio beta ₁: β ₂: ...: β _K

B) the total transmitted power P=0 of inferior user, total interference I=0 that main user receives.

C) transmitted bit of user k is counted R _k=0, k ∈ 1,2 ..., K}.

D) the bit number b that distributes on each subchannel _{K, n}=0.

E) inferior user's transmitted power increment P _{K, n}=0, the interference increment on each main user

The systematic cost increment

Wherein k ∈ 1,2 ..., K}, n ∈ 1,2 ..., N}, l ∈ 1,2 ..., L}.

2. subchannel distributes

A) according to (11), (17) formula is calculated the number of subchannels N that k time user is assigned to _k, wherein k ∈ 1,2 ..., K}.

B) calculate

If N-N ^*≠ 0, then selecting label is k=arg min _kR _k/ β _kInferior user, N _k=N _k+ 1, repeating step b), until N-N ^*=0.

C) calculate the Q value according to formula (16), and make up the cost matrix, use improved Hungary algorithm to obtain subchannel allocation matrix and the sets of sub-channels A that distributes to time user k _k, wherein k ∈ 1,2 ..., K}.

3. bit and power division

A) search R _k/ β _kIn minimum value, the selection label is k ^*=arg min _kR _k/ β _kInferior user.According to formula (12); (13) interference increment of calculating respectively on each main user of transmitted power increment

obtains wherein n ∈ { 1 according to formula (15); 2; ...; N}; L ∈ { 1; 2 ..., L}.If k ^*Value unique, directly search

In minimum value, select label to do $n^{*}=\mathrm{Arg}{\mathrm{Min}}_n{\stackrel{\&OverBar;}{\mathrm{\&Delta;\; C}}}_{k^{*},n}$ Subchannel, $n^{*}\&Element;A_{k^{*}}.$

B) if k ^*Value not unique, select label to do

Subchannel,

(convenient for expressing, k in the iterative step of back ^*Also use k ^*Expression)

C) judge

Whether reached the maximum transmit power restriction P that time user allows _T,

Whether reach the maximum interference threshold that l main user allows

If do not reach, then change step d), if reach, distribute and finish.

D) with subchannel n ^*Distribute to time user k ^*,

Give subchannel n ^*Distribute 1 bit

Inferior user k ^*The transmitted bit number

Return the step a) iteration.

Claims (1)

1. the adaptive resource allocation method in the cognitive ofdm system is characterized in that this method may further comprise the steps:

A. time user base station is upgraded the communication link gain of time user k in real time according to the information of perception gained

Main user is to the interfering link gain of inferior user k

Inferior user k is to main user's interfering link gain

And between each subchannel and the shared frequency spectrum of main user apart from d _Nl, d wherein _NlBe n subchannel and l the distance between the shared frequency band of main user;

B. calculate k time user when the n subchannel is distributed 1 bit, the transmitted power increment of generation $\mathrm{\&Delta;}{P}_{k,n}=\frac{{\mathrm{\&sigma;}}^2+{\mathrm{\&Sigma;}}_{l=1}^L{J}_{k,n}^l}{{\left|h_{k,n}^{\mathrm{Ss}}\right|}^2}{2}^{b_{k,n}}$ And the interference increment that the main user of l is produced $\mathrm{\&Delta;}{I}_{k,n}^l=\mathrm{\&Delta;}{P}_{k,n}{\mathrm{IF}}_{k,n}^l,$ Wherein, σ ²Be the variance of additive white Gaussian noise,

Be l main user interference to inferior user k on the n subchannel, b _{K, n}The bit number that on the n subchannel, distributes for inferior user k, Be the gain on subchannel n between t user base station of time slot and the individual time receiver user of k,

Be k user on the n subchannel to l main user's interference factor, L is main user's a number;

C. for Power Limitation and interference-limited equilibrium are considered; Transmitted power increment and interference increment are got compromise; Be defined as k user distribute 1 bit on the n subchannel cost function for

wherein α be regulatory factor, with it do to L the main user average cost function that all obtains inferior user k of making even ${\stackrel{\&OverBar;}{\mathrm{\&Delta;C}}}_{k,n}=\frac{\underset{l=1}{\overset{L}{\mathrm{\&Sigma;}}}\mathrm{\&Delta;}{C}_{k,n}^l}{L};$

D. for K time user and N subchannel, adopt improvement Hungary algorithm to carry out subchannel and distribute, make whole custom system cost minimum, obtain the sets of sub-channels A of time user k thus _k

E. choose R _k/ β _kMinimum inferior user carries out subchannel and Bit Allocation in Discrete, wherein R _kBe the desired transmission rate of inferior user k, β _kFor the transmission rate of inferior user k acquisition, from sets of sub-channels A _kThe middle cost of selecting Minimum subchannel distributes 1 bit, and judges whether time user's transmitted power and the suffered interference of main user surpass thresholding separately, if surpass, distribute and finish, if do not have, continue to distribute.

Images