CN103326984B - A kind of channel allocation method based on conflict threshold restriction - Google Patents

Abstract

The present invention proposes a kind of channel allocation method based on conflict threshold restriction.Described method takies the regularity of distribution of authorizing frequency spectrum and authorizing subchannel idle condition for primary user, introduce the concept of conflict threshold, analyze time feature of user according to its data bag and the idle condition of channel, by selecting as far as possible few access channel when meeting conflict threshold, reduce the probability that time user conflicts with primary user.Conflict threshold combines with effective throughput by the inventive method, according to transmission data package size and its idle probability distribution parameters of each sub-channels, select the minimal number subchannel meeting its data requirement, thus allow more time user send data, reduce with the collision probability of primary user, increase network effective throughput, keep the stable of cognitive OFDM network.

Description

A kind of channel allocation method based on conflict threshold restriction

Technical field

The invention belongs to wireless communication technology field, particularly a kind of channel allocation method based on conflict threshold restriction.

Background technology

Radio communication is just towards high speed, broadband future development, also more urgent to the demand of frequency spectrum resource.Traditional fixed frequency spectrum method of salary distribution also exists the not high problem of the availability of frequency spectrum.Cognitive radio allows the mandate frequency spectrum of cognitive user (i.e. time user) dynamic access authorized user (i.e. primary user), greatly can improve the availability of frequency spectrum, effectively alleviate the problem of frequency spectrum resource anxiety.According to the mode of secondary user access network, spectrum-sharing techniques can be divided into shared (Underlay) and filled type (Overlay) two kinds.Sharedly refer to that time user and primary user share same section of frequency spectrum simultaneously, prerequisite is that secondary user can not cause interference to the communication of primary user, and measurement index is interference temperature; Filled type then refers to that time user can only utilize the frequency spectrum cavity-pocket do not used by primary user to carry out transfer of data, and when primary user has transmission demand, secondary user need keep out of the way.

The task of cognitive radio networks has sudden, may occur following two kinds of situations under filled type frequency spectrum share mode: when primary user has a bursty data, will reactivate by the frequency spectrum of secondary CU, but primary user need not advance notice time user; Another, be then because secondary user is limited to the impact of own hardware and environmental interference factor, judge by accident the state of primary user.Above-mentioned two situations all may cause time user to keep out of the way not in time, and then cause primary user and time data service conflict of user in same frequency range.This conflict not only can produce interference to the transfer of data of primary user, the Transmission caused when also can increase the power overhead of time user and cut other mandate frequency spectrums, and then causes the hydraulic performance decline of whole cognitive system.

OFDM as a kind of key technology for PHY, by numerous communication system and standard adopt.In view of it can chooser carrier wave neatly, and integrating the advantages such as irregular and discontinuous frequency spectrum resource, flexible utilization frequency spectrum and allocation of transmit power according to the orthogonality between each subcarrier and actual transmissions situation, OFDM is considered to the physical layer modulation technology being best suited for cognitive radio system.By the dynamic channel allocating technology in order in cognitive OFDM, optimum allocation is carried out to the good subchannel of channel condition, effectively can avoid the data service conflict of primary and secondary user in same frequency range, while guarantee primary user proper communication, improve the effective throughput of whole network.Therefore, the channel allocation algorithm of research novel practical, having great importance to realizing new generation of wireless wide-band communication system, having a extensive future.

In cognitive OFDM network under filled type frequency spectrum share mode, existing channel allocation method is generally the idle condition based on channel, or the selection of authorizing the rule of frequency spectrum to carry out accessing channel is occupied according to primary user, specifically comprise: 1, record primary user's once used frequency spectrum state information analyze channel idle probability, in conjunction with real-time frequency spectrum state aware result, make time user better can judge that primary user resumes the rule of frequency spectrum, Pre-handoff access channel, evades conflict; 2, secondary user is by calculating, and obtains primary user in residue channel and again returns the minimum channel of acquistion probability, and uses this channel to carry out transfer of data reduction collision probability, have ignored the free time length of channel to the conflict impact of primary user with time user; 3, by the knowledge of SVMs, real-time perception frequency spectrum also combines the free time length probability density of authorizing frequency spectrum, calculate the channel idle probability under given duration, the subchannel that secondary user selects idle probability larger selects access, and whether accurate the probability density of frequency spectrum free time length is larger on algorithm impact; 4, by the access way of physical layer, make time user that unnecessary channel can be used to carry out transfer of data, need more idle frequency spectrum can again access when primary user returns use, this has little significance when frequency spectrum resource anxiety.

Said method is all the angles from improving time user access authorization subchannel probability, considers that primary user and time user occupy the data collision problem of same mandate frequency spectrum.But the utilization of subchannel does not reach best, and network environment is all limited in the cognitive radio networks using single user, the available capacity of system does not reach optimum.

Summary of the invention

The object of the invention is to overcome the deficiencies in the prior art, for single primary user and repeatedly user coexists time collision problem, propose a kind of based on conflict threshold restriction channel allocation method.Described method considers that primary user takies the regularity of distribution of authorizing frequency spectrum and authorizing subchannel idle condition, introduce the concept of conflict threshold, analyze the feature of each user according to its data bag and the idle condition of channel, by selecting as far as possible few access channel when meeting conflict threshold, reduce the probability that time user conflicts with primary user.

In order to solve the problems of the technologies described above, the technical solution used in the present invention is: a kind of channel allocation method based on conflict threshold restriction, comprises step as follows:

Steps A, primary user sets each subchannel conflict threshold σ _i; Determining to authorize the subchannel free time length of frequency spectrum to obey parameter is λ _iexponential distribution; The data package size of secondary user m is wherein, i ∈ { 1,2 ... N}, λ _i>0, N are number of subchannels, and m is time user's sequence number, and m is the natural number being less than or equal to M, and M is time number of users;

Step B, determines that time user m maximum channel occupies duration set l _m,

The maximum channel that secondary user m allows on subchannel i occupies duration for:

$l_{i,m}^{\max }=-\frac{1}{{\mathrm{\λ}}_i}\ln (1-{\mathrm{\σ}}_i)$

Then secondary user m maximum channel occupies duration set l _mfor:

$l_m=\{l_{1,m}^{\max },l_{2,m}^{\max },...,l_{N,m}^{\max }\}$

Step C, secondary user's chooser channel, its concrete steps are as follows:

Step C-1, determines time user's subchannel number,

If secondary user m not yet allocated channel, chooser channel number is n, n is the minimum value meeting conflict threshold constraint; 1≤n≤N;

Step C-2, secondary user's subchannel distributes,

Duration set l is occupied from secondary user m maximum channel _min pick out maximum channel and occupy duration and be greater than subchannel, then occupy duration from maximum channel and be greater than subchannel in pick out n sub-channels and distribute to time user m; Wherein, C is the peak transfer rate of subchannel;

Step C-3, often distributes a sub-channels, upgrades and once gathers l _m, the subchannel maximum channel distributed is occupied duration information and rejects from set;

Step C-4, when all users be all assigned with subchannel or time, subchannel distributes end.

Beneficial effect of the present invention: the present invention proposes a kind of channel allocation method based on conflict threshold restriction.Described method takies the regularity of distribution of authorizing frequency spectrum and authorizing subchannel idle condition for primary user, introduce the concept of conflict threshold, analyze time feature of user according to its data bag and the idle condition of channel, by selecting as far as possible few access channel when meeting conflict threshold, reduce the probability that time user conflicts with primary user.Conflict threshold combines with effective throughput by the inventive method, according to transmission data package size and its idle probability distribution parameters of each sub-channels, select the minimal number subchannel meeting its data requirement, thus allow more time user send data, reduce with the collision probability of primary user, increase network effective throughput, keep the stable of cognitive OFDM network.

Embodiment

Below describe a kind of channel allocation method based on conflict threshold restriction of the present invention in detail.

Network adopts filled type to share frequency spectrum mode, and whole system is made up of a primary user, a cognitive base station and M time user, and M is natural number.If authorize the subchannel that frequency spectrum is B by N number of bandwidth to form, N is natural number, and the peak transfer rate of subchannel is C.For mandate frequency spectrum, primary user enjoys preferential right.Secondary user is then in an ofdm fashion in physical layer access idle sub-channels.

If it is λ that the free time length of subchannel i obeys parameter _iexponential distribution, i ∈ { 1,2 ... N}, λ _i>0, then subchannel i is in the probability density function f of idle condition _i(t) be

$f_i\left(t\right)=\left\{\begin{array}{cc}{\mathrm{\&lambda;}}_i{e}^{-{\mathrm{\&lambda;}}_{i}t},& t\&GreaterEqual;0\\ 0,& t<0\end{array}\right.---\left(1\right)$

Certain moment, secondary user m idle channel i need send size be packet, m is time user's sequence number, and m is the natural number being less than or equal to M; Ideally send the consuming time of this segment data bag, namely take and authorize frequency spectrum duration l _i,mfor:

$l_{i,m}=\frac{R_{\mathrm{req}}^{i,m}}{C}---\left(2\right)$

As subchannel i free time length t _i≤ l _i,mtime, will there is service conflict at subchannel i in primary and secondary user, collision probability for

$p_c^i=p_i(t_i\&le;l_{i,m})={\&Integral;}_{-\&infin;}^{l_{i,m}}{f}_i\left(t\right)\mathrm{dt}=1-e^{-{\mathrm{\&lambda;}}_i{l}_{i,m}}---\left(3\right)$

The set of secondary user m busy channel is w _m.The proper communication of secondary user, means on all busy channels, and the conflict of primary user and time user does not occur, and data could correctly be transmitted, then the probability of secondary user m Successful transmissions for

$p_s^m=\underset{i\&Element;w_m}{\mathrm{\&Pi;}}(1-p_c^i)---\left(4\right)$

As can be seen from formula (4): the sets of sub-channels w of secondary CU _min the number of sub-channels and set, the state of each subchannel will the probability size of impact time user data package Successful transmissions, and the Successful transmissions probability of secondary user can along with channel set w _mthe increase of sub-channels number and reducing.In conjunction with formula (2), along with channel set w _mthe increase of sub-channels number, (namely the packet that secondary user m is assigned on its every sub-channels occupied diminishes diminish), thus on each sub-channels, occupy duration l _i,mto diminish, thus reduce the service conflict probability of primary user and time user.

Definition effective throughput R _e, represent each user data package worst case transmit time T _totalthe data packet number that interior destination node successfully receives, is designated as:

$R_E=\frac{1}{T_{\mathrm{total}}}\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}\underset{i\&Element;w_m}{\mathrm{\&Pi;}}(1-p_c^i)\&CenterDot;R_{\mathrm{req}}^{i,m}---\left(5\right)$

When service conflict occurs for primary user and time user, secondary user must keep out of the way in time, namely interrupts its data taken in frequency range and sends, again find idle mandate frequency spectrum, and switch.This will certainly affect the stability of whole cognition wireless network, and the effective throughput performance of influential system.In conjunction with the characteristics and rules that data in cognitive ofdm system send, if carry out certain constraint to collision probability, the collision probability of primary and secondary user is limited in the threshold range of primary user's setting, then can greatly reduces this adverse influence.On the one hand, ensure that primary user's proper communication, conflict can be made again on the other hand to drop to minimum to the interference that secondary user causes, thus maintain the stability of system.

Therefore, using the service conflict probability of primary user and time user as constraints, under the proper communication prerequisite of protection primary user, by reasonably distributing the channel of multiple users, the effective throughput of whole cognitive system can be made to maximize.Constraints is:

$\underset{A}{\max }{R}_E=\underset{A}{\max }\frac{1}{T_{\mathrm{total}}}\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}\underset{i\&Element;w_m}{\mathrm{\&Pi;}}(1-p_c^i)\&CenterDot;R_{\mathrm{req}}^{i,m}$

$s.t.0\&le;p_c^i\&le;{\mathrm{\&sigma;}}_i$

$\underset{i=1}{\overset{M}{\mathrm{\&Sigma;}}}{a}_{i,m}\&GreaterEqual;1$

In formula, σ _ifor the subchannel i conflict threshold of primary user's setting.

By the conflict threshold σ set _iobtain with formula (3) maximum channel that time user m allows on subchannel i and occupy duration

$l_{i,m}^{\max }=-\frac{1}{{\mathrm{\&lambda;}}_i}\ln (1-{\mathrm{\&sigma;}}_i)---\left(7\right)$

Secondary user when taking n sub-channels, as during using identical threshold value σ as each subchannel constraints, data success sending probability for:

$p_s^m={(1-\mathrm{\&sigma;})}^n---\left(8\right)$

Usually 0< σ <1 is established.Therefore, n is less, then the probability of success is larger.Can obtain thus: when the collision probability threshold value of secondary CU channel and primary user is σ, select subchannel can improve the probability of success of time delivery of user traffic as few as possible, the proper communication of protection primary user.

Under conflict threshold constraint, in cognitive OFDM environment, repeatedly the concrete steps of user's multichannel suboptimum distribution method are:

The first step: initialization.Primary user sets each subchannel conflict threshold σ _i, for convenience of calculating, all σ can be set _iall identical, i.e. σ _i=σ.Determine subchannel free time length according to existing frequency spectrum perception result, wherein authorizing the free time length of frequency spectrum to obey parameter is λ _iexponential distribution.The data package size set of each user is $R_{\mathrm{req}}=\{R_{\mathrm{req}}^{i,1},R_{\mathrm{req}}^{i,2},...R_{\mathrm{req}}^{i,M}\}.$

Second step: according to threshold value σ, convolution (7) estimates that each subchannel allows time user m to occupy the maximum transmitted duration set of use $l_m=\{l_{1,m}^{\max },l_{2,m}^{\max },...,l_{N,m}^{\max }\}.$

3rd step: meeting under user sends demand prerequisite, select the minimum channel number taken, to allow more time user participate in channel allocation.

1) to the secondary user of each unallocated channel, when meeting conflict threshold constraint, select minimum channel number n(1≤n≤N) distribute to this user, to ensure that as much as possible user can be assigned to channel when meeting conflict threshold constraint, promote effective throughput.Specific practice: to the secondary user of each unallocated channel, n increases progressively successively from 1 to N, selects the minimum n meeting conflict threshold constraint and distributes to this user;

2) in the assignment procedure, if secondary user m not yet allocated channel, then from the duration set l of second step _min pick out Occupation time and be greater than subchannel (namely time user occupies the subchannel time and is less than the maximum Occupation time that this channel allows, and reduces the collision probability of primary and secondary user, the proper communication of available protecting primary user with this), select n sub-channels and distribute to time user m;

3) often distribute a sub-channels, upgrade and once gather l _m, the subchannel maximum transmitted duration information distributed is rejected from set;

4) all users be all assigned subchannel or time, subchannel distributes end.

Claims (1)

1., based on a channel allocation method for conflict threshold restriction, it is characterized in that, comprise step as follows:

Steps A, primary user sets each subchannel conflict threshold σ _i; Determining to authorize the subchannel free time length of frequency spectrum to obey parameter is λ _iexponential distribution; The data package size of secondary user m is wherein, i ∈ { 1,2 ... N}, λ _i>0, N are number of subchannels, and m is time user's sequence number, and m is the natural number being less than or equal to M, and M is time number of users;

Step B, determines that time user m maximum channel occupies duration set l _m,

The maximum channel that secondary user m allows on subchannel i occupies duration for:

$l_{i,m}^{\max }=-\frac{1}{{\mathrm{\&lambda;}}_i}ln(1-{\mathrm{\&sigma;}}_i)$

Wherein, σ _ifor subchannel i conflict threshold, and 0< σ _i<1; Then secondary user m maximum channel occupies duration set l _mfor:

$l_m=\{l_{1,m}^{\max },l_{2,m}^{\max },...,l_{N,m}^{\max }\}$

Step C, secondary user's chooser channel, its concrete steps are as follows:

Step C-1, determines time user's subchannel number,

If secondary user m not yet allocated channel, chooser channel number is n, n is the minimum value meeting conflict threshold constraint; 1≤n≤N;

Step C-2, secondary user's subchannel distributes,

Duration set l is occupied from secondary user m maximum channel _min pick out maximum channel and occupy duration and be greater than subchannel, then occupy duration from maximum channel and be greater than subchannel in pick out n sub-channels and distribute to time user m; Wherein, C is the peak transfer rate of subchannel;

Step C-3, often distributes a sub-channels, upgrades and once gathers l _m, the subchannel maximum channel distributed is occupied duration information and rejects from set;

Step C-4, when all users be all assigned with subchannel or time, subchannel distributes end.