CN104936185A - Video communication power and rate combined control method in cognitive radio - Google Patents

Abstract

The invention aims at providing a video communication power and rate combined control method in cognitive radio. The method is provided by a utility function of a main user, interference of a slavery user in the main user and income of the main user are taken into consideration, and a utility function aimed at the characteristics of video service of the slavery user is provided in construction of the slavery user, and fully considers that different slavery users have different requirements for the bandwidth and power when videos of different contents are transmitted. The fairness and validity of resource distribution in cognitive radio are embodied, and the unit power price of the slavery users is calculated based on the principle that the effectiveness of the main user is maximal, and thus, power interference of the slavery users in the main user can be reduced as possible.

Description

A kind of video communication integrated power and rate control method in cognitive radio

Technical field

The invention belongs to communication technical field, relate to a kind of video communication integrated power and rate control method in cognitive radio.

Background technology

Cognitive radio has been acknowledged as a kind of wireless network saving bandwidth resources in the communications field.It can allow user more efficient and rational utilize Radio Resource.In cognitive radio networks, primary user is authorized to certain bandwidth, and one or more users are when primary user has frequency spectrum idle, and its idle bandwidth of autonomous perception also shares frequency spectrum with it under the condition of not interfere with primary users own service.

Nowadays the problem transmitting data in more and more research cognitive radio has been had.In cognitive radio, controlling power is improve Internet resources to recycle a very effective method.Reduce the transmitting power of time user, not only can reduce primary user and secondary user power is disturbed, the battery of time customer mobile terminal can also be extended.In more existing resource allocation methods, there is various deficiency, although be all such as power and speed co-allocation mechanism under cognitive radio scene, all do not consider specific business.Nowadays video traffic gets more and more, and as Video chat, Web conference etc., these videos have special requirement in transmitting procedure, as very sensitive to time delay, need to consider user's visual experience etc.Therefore, when transmission of video business, traditional joint Power method of rate control will be no longer suitable.Certainly, many scholars are also had to carry out the research of Resourse Distribute to the video traffic of feature.But blemish in an otherwise perfect thing, they only considered the control of video transmitting procedure in speed, but do not have power control mechanism.In wireless network transmission of video, it is vital that rational power controls, and it can reduce the interference between each user, improves the transmission quality of video.Therefore, under needing a kind of cognitive radio scene, for video traffic, can the method for fair allocation of transmission resources efficiently.

The present invention on existing methods basis, under proposing a kind of scene of cognitive radio transmission of video, based on the joint Power rate control mechanism of non-cooperative game.This mechanism, for transmission of video feature, proposes a kind of fair time user's betting model efficiently.When this model has taken into full account different user transmission different video content, to the requirement of speed also this feature different, the video quality received using user as the income of secondary user, instead of only using throughput as secondary user's income.This embodies the fairness between each user.Meanwhile, the power unit price of secondary user is that primary user calculates according to self effectiveness is maximum, ensure that the income of primary user.

Summary of the invention

Technical problem: be of the present inventionly a kind of method that video communication integrated power and rate control in cognitive radio is provided.The method under the condition of primary user's optimal utility, can calculate penalty factor, and the value of penalty factor is told time user, secondary user is according to this information, carry out game each other, make self optimal utility, finally determine transmission rate and the transmitting power of each user.

Technical scheme: under the invention reside in the scene proposing a kind of cognitive radio transmission of video, based on the joint Power rate control mechanism of non-cooperative game.This mechanism, for transmission of video feature, proposes a kind of fair time user's betting model efficiently.When this model can take into full account different user transmission different video content, to the requirement of speed also this feature different, the video quality received using user as the income of secondary user, instead of only using throughput as secondary user's income.This embodies the fairness between each user.Meanwhile, the power unit price of secondary user is that primary user calculates according to self effectiveness is maximum, ensure that the income of primary user.

The network structure of cognitive radio system involved in the present invention as shown in Figure 1.In figure, primary user is PU, and authorized bandwidth is the frequency spectrum of W (MHz), primary user and primary user's base station BS _pcommunicate.Time user is SU simultaneously _i, i=1,2 ..., N, in the wireless network that this N number of user uses primary user to authorize, with primary user's frequency spectrum share simultaneously, and adopts code division multiple access (CDMA) mode and time user base station BS _scommunicate.

Primary user self has business, needs certain bandwidth, also has certain remaining bandwidth amount simultaneously.The bandwidth information of oneself upgrades by primary user in time, the message that secondary user upgrades according to primary user, utilizes primary user's remaining bandwidth to carry out bandwidth sharing with it.Meanwhile, secondary user is when transmission of video, and its transmitting power can have a certain impact to primary user's communication, so primary user needs to charge to secondary user emission power, to limit the interference of time user.Primary user wishes that income is more high better, and therefore this punishment is that the increase with secondary user emission power increases under guarantee primary user effectiveness maximum case.

Between secondary user in the process of shared bandwidth, all wish can obtain more resource under less cost, therefore need to build the effectiveness that time user utility function represents time user.The utility function of secondary user according to self and the information of penalty factor, carry out game each other, makes the effectiveness that final acquisition is maximum, now then can obtain transmission bandwidth and the transmitting power of each user.

Beneficial effect: the object of the invention is under cognitive radio scene, provides a kind of based on joint Power rate control strategy in game theoretic video transmitting procedure.First present invention contemplates the utility function of primary user, primary user's utility function can be made to reach maximum.Secondly, use penalty factor to effectively reduce the transmitting power of time user, thus reduce time user to the interference of primary user.The structure of secondary user utility function has taken into full account and the feature of different user video content has made Resourse Distribute more fair.

Accompanying drawing explanation

Fig. 1 is the illustraton of model of system.

Embodiment

1, utility function is defined

Definition primary user utility function is as follows:

$U_p=c\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i{g}_i-m{(\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i{g}_i-I^{\mathrm{tar}})}^2---\left(\text{1}\right)$

In (1) formula, Section 1 represents and is revenue of primary user by the total penalty that primary user disturbs secondary user.I ^tarrepresent the jamming target value of primary user, secondary user is more greater than this desired value to the interference of primary user, and that interference is more serious.The implication of Section 2 is time impact of user's interference on primary user's performance.

When asking penalty factor c, ensure that the effectiveness of primary user is maximum, therefore by the utility function U of primary user _pto p _iask local derviation, and derivative is set to 0, then obtain penalty factor c.

$\frac{{\∂U}_p}{{\∂p}_i}={\mathrm{cg}}_i-2m(\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i{g}_i-I^{\mathrm{tar}})=0---\left(3\right)$

Therefore:

$c=2m(\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i{g}_i-I^{\mathrm{tar}})---\left(4\right)$

Can be found out by (4) formula, the penalty factor c in cost function (2), with the interference summation of secondary user to primary user increase and increase.

Because the amount of information of video traffic is comparatively large, therefore to compress before being transmitted.Generally be just used to describe the relation between compression speed and video distortion with rate-distortion model.The present invention is by the rate-distortion model [9] of employing two parameters, and this model formation is as follows:

D(R)＝αe ^-βR(5)

In formula, D is video distortion, and R is the compression speed of this video, and α and the β specific parameter that is video, different video contents, these two parameter values are also different.In a practical situation, more often use signal to noise ratio (PSNR) to describe the quality of video, the relation between it and rate distortion is as follows:

$\mathrm{PSNR}={10\log }_{10}\frac{{255}^2}{D}---\left(6\right)$

Definition time user utility function is as follows:

$u_{\mathrm{si}}=\frac{{\mathrm{PSNR}}_i}{p_i}-{\mathrm{cp}}_i{g}_i---\left(7\right)$

$=\frac{z_i+{\mathrm{\β}}_i{R}_i{\log }_e\left({\mathrm{k\γ}}_i\right)}{p_i}-{\mathrm{cp}}_i{g}_i---\left(8\right)$

Therefore in (8), Section 1 is the income of time user i transmission of video, and Section 2 is the punishment that he causes interference to be subject to primary user.By analyzing above and can knowing

2, the foundation of betting model

In actual transmissions process, in order to ensure that video can roughly be seen clearly, each user will have a minimum transmission rate and minimum emissive power meanwhile, in order to save Internet resources, each user also has a maximum transmission power and maximum transmission power the total interference of secondary user to primary user also has a boundary value, and total interference can not exceed this boundary I _th, and each time user also has a minimum signal interference ratio requirement therefore, secondary user's betting model is:

$\begin{array}{cc}\underset{R_i,p_i}{\max }& u_{\mathrm{si}}=\frac{z_i+{\mathrm{\β}}_i{R}_i{\log }_e\left({\mathrm{k\γ}}_i\right)}{p_i}-{\mathrm{cp}}_i{g}_i\end{array}$

$\begin{array}{cc}s.t.& \begin{array}{c}{p}_i^{\min }\≤p_i\≤p_i^{\max }\\ R_i^{\min }\≤R_i\≤R_i^{\max }\end{array}\end{array}$

$\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i{g}_i\≤I_{\mathrm{th}}$

${\mathrm{\γ}}_i\≥{\mathrm{\γ}}_i^{\mathrm{tar}}---\left(10\right)$

3, Nash Equilibrium uniqueness and existence proof

In a, the bandwidth process that participates in the present invention N number of user, each time user i has maximum, a minimum-rate restriction, namely limit with minimax transmitting power this obvious non-NULL and be convex set, and at theorem in Euclid space R ⁿin, the closed set of a bounded must be compact, and therefore can obtain this game theoretical model and meet above condition 1.

B, by formula (8) to R _iand p _isingle order local derviation is asked to obtain respectively:

$\frac{{\∂u}_{\mathrm{si}}}{{\∂R}_i}=\frac{{\mathrm{\β}}_i[{\log }_e\left({\mathrm{k\γ}}_i\right)-1]}{p_i}---\left(13\right)$

$\frac{{\∂u}_{\mathrm{si}}}{{\∂p}_i}=\frac{{\mathrm{\β}}_i{R}_i-z_i-{\mathrm{\β}}_i{R}_i{\log }_e\left({\mathrm{k\γ}}_i\right)}{p_i^2}-2{\mathrm{cg}}_i---\left(14\right)$

By (11) formula again to R _idifferentiate can obtain:

$\frac{{\&PartialD;}^2{u}_{\mathrm{si}}}{{\&PartialD;R}_i^2}=-\frac{{\mathrm{\&beta;}}_i}{p_i{R}_i}<0---\left(15\right)$

By (12) formula again to p _idifferentiate can obtain:

$\frac{{\&PartialD;}^2{u}_{\mathrm{si}}}{{\&PartialD;p}_i^2}=-\frac{{\mathrm{\&beta;}}_i{R}_i[-1-{\log }_e\left({\mathrm{k\&gamma;}}_i\right)]-{2z}_i}{{p_i}^3}<0---\left(\text{16}\right)$

Found out by (15) and (16) class, the utility function u of secondary user _sito R _iand p _ifor concave function, therefore satisfy condition 2.

In sum, there is Nash Equilibrium Solution in the present invention's time user's betting model.

If in the game that a N people participates in, its best reflection function U (s)={ U ₁(s), U ₂(s) ..., U _n(s) } meet the definition of canonical function: 1, U (s) perseverance is positive number; 2, for arbitrary s, if s>=s', so U (s)>=U (s'); 3, for arbitrary real number m > 1, mU (s) > U (ms) is had.So unique by the Nash Equilibrium that method is demonstrate,proved above.

Proved out that this betting model exists Nash Equilibrium above, clearly betting model of the present invention meets the definition of canonical function, can this Nash Equilibrium be therefore unique.

Frequency spectrum update method adopts following alternative manner:

Step 1: maximum by primary user's effectiveness, determines penalty factor c,

Step 2: arrange time user i (i=1,2 ..., N) initial transmission rate and transmission power level p _i(t), R _i(t) (t=0).

Step 3:t=t+1, $R_i(t+1)=\max \{R_i^{\min },\min ({\arg \max }_{R_i}{u}_{\mathrm{si}}(p_i\left(t\right),R_i\left(t\right)),R_i^{\text{max}})\},$ $p_i(t+1)=\max \{p_i^{\min },\min ({\arg \max }_{p_i}{u}_{\mathrm{si}}(p_i\left(t\right),R_i\left(t\right)),p_i^{\text{max}})\}.$

Step 4: judge and | p _i(t+1)-p _i(t) |≤ε (ε is iteration precision).

Step 5: if meet above-mentioned requirements, then iteration terminates, otherwise returns step 3.

Claims (1)

1. a video communication integrated power and rate control method in cognitive radio, is characterized in that concrete grammar is:

A, determine primary user's utility function: definition primary user utility function is wherein i refers to i-th time user, c be primary user to the power penalty factor of time user, N is the secondary total number of users in system, p _irepresent the transmitting power of i-th user, g _ibe the channel gain of time user i to primary user base station, m is weighted factor, I ^tarit is primary user's target jamming value;

B, solve power penalty factor: by the utility function U of primary user _pto p _idifferentiate also sets to 0, and can try to achieve penalty factor $c=2m(\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{p}_i{g}_i-I^{\mathrm{tar}});$

C, determine time user utility function: definition time user utility function is $u_{\mathrm{si}}=\frac{{\mathrm{PSNR}}_i}{p_i}{-\mathrm{cp}}_i{g}_i=\frac{z_i+{\mathrm{\&beta;}}_i{R}_i{\log }_e\left({\mathrm{k\&gamma;}}_i\right)}{p_i}-{\mathrm{cp}}_i{g}_i,$ Wherein PSNR is the Y-PSNR of video image, is used for describing the quality of video, z _i=2log _e255-log _ea _i, a _iand β _ibe the parameter relevant with i-th user institute transmitted video content, k is constant, γ _itime user's statistic SINR, R _ibe the transmission rate of i-th user, w is the total bandwidth that time user and primary user share, h _ithe channel gain of time user i to time user base station, σ ²it is the Background Noise Power of time user's receiving terminal;

D, the transmission rate determining time user and transmitting power:

Step 1: determine penalty factor $c=2m(\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{p}_i{g}_i-I^{\mathrm{tar}});$

Step 2:t=0, arrange time user i (i=1,2 ..., N) initial transmission rate and transmission power level p _i(t), R _i(t), p _it () is the transmitting power of i-th user's the t time iteration, R _it () is the transmission rate of i-th user's the t time iteration;

Step 3:t=t+1, $R_i(t+1)=\max \{R_i^{\min },\min ({\arg \max }_{R_i}{u}_{\mathrm{si}}(p_i\left(t\right),R_i\left(t\right)),R_i^{\max })\},$ $p_i(t+1)=\max \{p_i^{\min },\min ({\arg \max }_{p_i}{u}_{\mathrm{si}}(p_i\left(t\right),R_i\left(t\right)),p_i^{\max })\};$

Step 4: judge whether to satisfy condition and | p _i(t+1)-p _i(t) |≤ε, ε are iteration precision;

Step 5: if meet above-mentioned requirements, then iteration terminates, otherwise returns step 3.