CN102883294A - Segmental time interval billing method relevant to user behaviors - Google Patents

Abstract

The invention relates to a segmental time interval billing method relevant to user behaviors. A relation between a network rate and an operator profit is established by using a user network utility function and a satisfaction function, and a network profit extreme earned by an operator and a condition which is required to be met by the network rate can be calculated according to the relation; and furthermore, an optimal segmenting mode of the operator under different segmental time interval numbers, rates corresponding to the segments and profit loss of the network profit relative to the profit extreme can be obtained by optimizing a segmental time interval billing mechanism.

Description

A kind of at times charging method of associated user behavior

Technical field

The invention belongs to network billing technical field in the communication system, be specially the at times charging method of associated user behavior in a kind of Operation Network.

Background technology

The user used the behavior of network often to have temporal Preference in one day, and it is the basic reason that causes network traffics to fluctuate in time.In recent years, along with take provide any time, the broadband access in any place is as the high speed development of the mobile Internet of target, this temporal fluctuation of network traffics becomes more obvious.On the other hand, the mobile operator billing mechanism adopts usage-based billing more, and adopts nontraffic sensitive.Under the effect of this billing mechanism, the network charges of user payment is directly proportional with the network total flow of its consumption, and with when use these flows to have nothing to do.In the limited situation of Internet resources, time preference's property of user network behavior will cause the network congestion of the idle and busy of the Internet resources of idle.The idle waste that will cause resource of Internet resources; Network congestion then will affect user's network experience, reduce the expense (WTP) that the user is willing to mean the network service payment.No matter be that Internet resources are idle, or network congestion, all will directly affect the network profit of operator.The time become billing mechanism adapts to the user by dynamically changing the network rate time preference's property, this has alleviated the impact of user time Preference on the carrier network income to a great extent.Yet in the situation that the user does not understand fully, the frequent variations of rate will be suffered user's resistance.At times charging becomes billing mechanism as a kind of half and can in network profit and good compromise of user's acceptance intercropping, be expected to be accepted by operator and user.By existing patent and correlation technique retrieval are found that existing at times charging and Network Users'Behaviors Analysis method have:

(1) a kind of segmented charging method (CN03104823.4) is realized at times charging by inquiring about a default volatile data base table.A kind of method of real-time segmentation charging (CN03116111.1) has provided when several accounting regimes are crossed in user's a use procedure, the processing method of at times charging.Described two kinds of methods have provided the specific embodiments of segmented charging, but do not provide the method for determining optimal segmentation mode and each section rate.

(2) a kind of real-time charging method and system (CN200910243211.7) come the predictive user bandwidth demand by user's history with record, and then control the behavior that the user uses network according to the balance amount information of user account.It just formulates corresponding network service strategy based on user's network behavior, does not relate to charging policy.Intelligent pipeline flow control methods (CN201210010674.0) based on user behavior is dynamically adjusted flow and charging control strategy according to the variation of service feature, thereby solves to a certain extent the problem of wireless network resource and assignment of traffic inequality.Two kinds of above-mentioned methods are all less than to improve operator's income as target.Existing block meter rate method, mostly only provided simple implementing method, do not provide one and formulate the at times criterion of billing mechanism, and analyze and utilize the method for user network behavior mostly to only limit to control user behavior, thereby do not utilize user's network behavior to formulate the method for block meter rate strategy maximization carrier network income.

Summary of the invention

The mobile Internet access demand is the trend of rapid growth in recent years, this concerning mobile operator be opportunity also be the challenge.The great demand of mobile broadband access has been brought up a huge industry on the one hand, and it has brought great challenge also for the access network of operator on the other hand.How in the situation that does not increase Internet resources and operation cost, guide user's network behavior with billing mechanism, improve network profit and become operator's outline.The present invention becomes billing mechanism and the network profit that can obtain of billing mechanism at times when utilizing the network rate that the mechanism of action of user behavior is determined that operator adopts, and has provided operator determine optimal segmentation mode and the method for the rate of correspondence with it when adopting block meter rate mechanism.Become when adopting by comparison simultaneously, billing mechanism and at times the income during billing mechanism just can obtain operator because of the loss in revenue of adopting billing mechanism at times to cause; Operator can consider that synthetically loss in revenue and user select optimum timesharing hop count, segmentation method and the rate corresponding with each section to the acceptance of billing mechanism at times by the method.

The technical scheme that the present invention solves described problem is, a kind of at times charging method of associated user behavior is characterized in that, in the mobile phone Operation Network, charging platform adopts the related mobilephone terminal user of following means and network billing data:

(1) obtains relation between customer consumption bandwidth and network rate and the relation between network rate and operator's income by network utility function and user satisfaction function S (x);

(2) thus by the network rate mechanism of action of user behavior being come user's network behavioral data in basic metering period processed obtains the user network behaviors feature parameter;

(3) under specific user's behavioural characteristic, obtain the optimum toll rate in the basic cycle And the network profit limit of operator

(4) obtain when adopting different segments by optimizing, at times the network profit of billing mechanism, segmented mode and the rate corresponding with day part;

(5) under different segments, network profit and income limit R that at times billing mechanism optimization is obtained _LimCompare and obtain corresponding loss in revenue rate η _ML

Compared with prior art, the invention has the beneficial effects as follows:

Rate Network Based is to the mechanism of action of user behavior and the network rate mechanism of action to operator's income, for operator provides one to formulate the at times Optimality Criteria of billing mechanism; This criterion can be under different segments, with the network profit maximization of operator.

Become billing mechanism and the at times loss in revenue during billing mechanism when adopting by comparison operator, provided and estimated that operator is because of the method for the loss in revenue adopting billing mechanism at times and bring.

In order to be illustrated more clearly in embodiments of the invention, the below does to introduce simply to correlation graph and the derivation of equation of required use among the embodiment: pass through extent function S (x) differentiate to the user in the described step (1), and solving equation S ' (x)=0 obtains the bandwidth of user selection consumption and the relation between the relation between the network rate and network rate and operator's income; In the described step (2) by the user at two kinds of different rate p _a, p _bUnder bandwidth selection x (p _a), x (p _b) obtain user's network behavior characteristic parameter; In the described step (3) by to the network profit of operator and rate concern R (p) differentiate, and solving equation R ' (p)=0 obtains optimum toll rate

Maximum network profit

And the income limit

When operator adopts at times billing mechanism in the described step (4), by separate optimization problem max (R '+R "+...); Variable: p ', p " ..., n ₁, n ₂... } and obtain optimum segmented mode, corresponding rate and network profit rate.

Apparently, the accompanying drawing in the following describes only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawings according to these accompanying drawings.

Description of drawings is as follows:

The segmented mode of the at times billing mechanism of two sections systems of Fig. 1

Fig. 2 nontraffic sensitive p _a=0.20 or p _bThe bandwidth of user selection consumption in=0.25 o'clock

Fig. 3 user behavior characteristic parameter sequence Γ { σ _i, η _i}

Become the billing mechanism optimum results during Fig. 4

Fig. 5 is the billing mechanism optimum results at times

Fig. 6 network rate changes the impact on network profit

Two sections accounting regimes processed of Fig. 7 distribute

Fig. 8 network rate is to the influence curve of income

The growth rate curve that Fig. 9 network profit changes along with the network rate.

Embodiment

Below in conjunction with the subordinate list in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described only is the present invention's part embodiment, rather than whole embodiment.Based on embodiments of the invention, the every other embodiment that those of ordinary skills obtain under the prerequisite of not making creative work belongs to the scope of protection of the invention.

1. choose the user network utility function

It has shown the relation between the bandwidth x of network utility u that the user obtains and its consumption.Wherein, η is the user bandwidth demand elasticity, and 1/ η is larger, and bandwidth demand is larger; x ₀Be correction factor, expression needs QoS to ensure the primary bandwidth demand of class business.

2. obtain the extent function S (x) that the user uses network=σ u (x)-px; Wherein, p is the network rate, and σ represents the network utility level, and it is used to regulating networks effectiveness and network rate to the weight of user satisfaction impact.

3. the user who to S (x) differentiate and solving equation S ' (x)=0, obtains behavioural characteristic and be { σ, η } selects the bandwidth of consuming when network rate p $\hat{x}\left(p\right)=\frac{1}{\mathrm{\η}}(\ln \frac{p}{\mathrm{\σ\η}}-\ln (1-\frac{p}{\mathrm{\σ\η}}-\sqrt{1-\frac{2p}{\mathrm{\σ\η}}})).$

4. if be fixed as p in the network rate _aOr p _bThe time user selection consumption bandwidth be x (p _a) and x (p _b), with { p _a, x (p _a) and { p _b, x (p _b) generation respectively $\hat{x}\left(p\right)=\frac{1}{{\mathrm{\η}}_0}(\ln \frac{p}{{\mathrm{\σ}}_0{\mathrm{\η}}_0}-\ln (1-\frac{p}{{\mathrm{\σ}}_0\mathrm{\σ}{\mathrm{\η}}_0}-\sqrt{1-\frac{2p}{{\mathrm{\σ}}_0\mathrm{\σ}{\mathrm{\η}}_0}}))$ And simultaneous just can solve user network behaviors feature parameter { σ ₀, η ₀.If basic metering period (a day) is divided into 24 periods, can obtain the cybernetics control number sequence Γ { σ of user within the basic cycle just repeat above process _i, η _i; I=1,2 ..., 24.

5. at specific user behavior feature { σ ₀, η ₀Under, obtain the network profit of operator and the relation between rate $R\left(p\right)=\frac{p}{{\mathrm{\η}}_0}(\ln \frac{p}{{\mathrm{\σ}}_0{\mathrm{\η}}_0}-\ln (1-\frac{p}{{\mathrm{\σ}}_0{\mathrm{\η}}_0}-\sqrt{1-\frac{2p}{{\mathrm{\σ}}_0{\mathrm{\η}}_0}}));$ Solving equation R ' (p)=0 obtains optimum toll rate

Maximum network profit

And the income limit $R_{\lim }=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}R\left({\hat{p}}_i\right).$

6. because operator carries out charging to a plurality of basic cycles [0, T] that link together.In minutes two sections at times billing mechanism, " the corresponding period should be { n with rate p ' and p respectively ₂+ 1 ..., n, 1,2 ..., n ₁, { n ₁+ 1 ..., n ₂, as shown in Figure 1.Operator can be expressed as in the income of two accounting regimes $R^{\&prime;}=\frac{T}{n}(\underset{i=1}{\overset{{\mathrm{<figure-callout\; id="1"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_1}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{p^{\&prime;}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right)+\underset{i={\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_2+1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{p^{\&prime;}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right));$ $R^{\&prime;\&prime;}=\frac{T}{n}\underset{i={\mathrm{<figure-callout\; id="1"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_1+1}{\overset{{\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_2}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{p^{\&prime;\&prime;}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right).$

Solving-optimizing problem: max (R '+R "), variable: p ', p ", n ₁, n ₂; Just the result can be optimized At this moment, the network profit loss is expressed as: ${\mathrm{\&eta;}}_{2L}=1-\frac{\max (R^{\&prime;}+R^{\&prime;\&prime;})}{R},$

8. with process 3,4 similar, can be in the hope of segments m=3,4 ...,, obtain corresponding optimizing fractional method, corresponding rate and η at 23 o'clock _3L, η _4L... η _23L(distinguishingly, when segments is 24, become billing mechanism when billing mechanism is equal at times, try to achieve the front)

Emulation experiment

1. be respectively p at nontraffic sensitive _a=0.20 and p _b=0.25 o'clock, Fig. 2 provided the bandwidth x (p of customer consumption _b) and x (p _b), utilize these data can obtain user behavior characteristic parameter sequence Γ { σ _i, η _i; I=1,2 ..., 24, as shown in Figure 3.

2. be Γ { σ to the network behavior characteristic parameter _i, η _iThe user, by optimize can draw the time become in the billing mechanism, the optimum toll rate corresponding with this user behavior characteristic parameter is

And the network profit limit $R_{\lim }=0.4477\underset{i=1}{\overset{24}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i,$ As shown in Figure 4.

3. to have provided at the user behavior characteristic parameter be Γ { σ to Fig. 5 _i, η _iThe time, separate the at times billing mechanism optimum results that the multi-parameters optimization problem obtains; Comprise segmented mode, the rate corresponding with each section and corresponding loss in revenue rate.

Appendix: the derivation of equation in the literary composition

1 user network behavior modeling

1.1 user utility surplus

Effectiveness is the concept in the microeconomics, and it is used for describing the commodity of specified quantitative or the satisfaction degree that service brings to the user.In the network economics, utility function u (x) commonly used describes the relation between user network effectiveness u and its consumption network bandwidth x.It is generally acknowledged that utility function has following characteristics at its domain of definition x ∈ [0 ,+∞]:

1) network utility is non-negative, i.e. u (x) 〉=0;

2) u (0)=0, namely x →+during ∞, u (x) → 1;

3) utility function monotonic increase, namely u ' (x) 〉=0.

Constructed following utility function model herein:

$u\left(x\right)=\frac{1-e^{-\mathrm{\&eta;}(x-x_0)}}{1+e^{-\mathrm{\&eta;}(x-x_0)}}---\left(1\right)$

Wherein, x ₀Be correction factor, expression needs QoS to ensure the primary bandwidth demand of class business; η represents the user bandwidth demand elasticity, and 1/ η is larger, and user's network bandwidth requirements is also just larger, and vice versa.

The user is by satisfying its demand with network, the user also needs to pay certain network charges simultaneously.The user uses network of network effectiveness higher, and it uses the satisfaction of network also just higher; And network charges is higher, and satisfaction is lower.For this reason, we represent that with network utility surplus S the user uses the satisfaction of network, and expression formula is as follows:

S＝σu(x)-P (2)

Wherein, P is network price; σ represents the network utility level, and it is used to regulating networks effectiveness and network price to the weight of user satisfaction impact.Simultaneously σ is larger, and when obtaining the network utility of equivalent, the expense (WTP) that the user is willing to mean the network service payment is also just higher.Represent user's bandwidth efficiency with product σ η, the back can prove that it will directly affect the network profit of operator.And if only if S 〉=0 o'clock, the user just can use network service.To sum up, user's network behavior feature Γ can be by parameter group { σ, η, x ₀Unique definite.

1.2 user network behavior

For the ease of analyzing, be without loss of generality, at first user's bandwidth demand made the following assumptions:

1) user's behavior is subject to the impact of three factors, and user network demand, user are ready network charges and the network price paid;

2) the linear billing mechanism of network using, i.e. P=px, wherein p is the network rate;

3) only considering need not QoS and ensure that class is professional, also is x ₀=0, this moment, user's network behavior feature Γ parameter group was reduced to { σ, η };

4) the certain bandwidth of user selection consumption makes its Maximum Satisfaction, namely

$\underset{\arg \left\{x\right\}}{\max }S\left(x\right)=\underset{\arg \left\{x\right\}}{\max }(\mathrm{\&sigma;u}\left(x\right)-\mathrm{px})---\left(3\right)$

According to the hypothesis of front, can prove that S (x) is convex function.So, the bandwidth that has the user selection consumption of particular network behavioural characteristic Γ

Should satisfy

$S^{\&prime;}\left(\hat{x}\right)=0$ Γ{σ，η} (4)

If σ=σ 0 and η=η ₀, solve an equation (4) can obtain:

$\hat{x}\left(a\right)=\frac{1}{{\mathrm{\&eta;}}_0}(\ln a-\ln (1-a-\sqrt{1-2a})),$ $a=\frac{p}{{\mathrm{\&sigma;}}_0{\mathrm{\&eta;}}_0}\&GreaterEqual;1/2---\left(5\right)$

If p=p ₀, solve an equation (4) can get:

$\hat{x}\left(b\right)=\frac{1}{\mathrm{\&eta;}}\ln (b-1+\sqrt{b^2-2b}),$ $b=\frac{\mathrm{\&sigma;\&eta;}}{p_0}\&GreaterEqual;2---\left(6\right)$

Formula (5) will be for the impact of Probe into Network rate on operator's income, and formula (6) then will be for inquiring into the impact of user network behavior on operator's income.User behavior feature Γ and bandwidth demand just can be set up in through type (5), (6) Between relation, as long as the bandwidth demand x of given certain user when the network rate is p _tT ∈ [0, T] just can obtain the user network behaviors feature Γ { σ of this user within the cycle [0, T] _t, η _t; T ∈ [0, T].

Became billing mechanism optimization at 2 o'clock

2.1 carrier network income analysis

Consider that operator adopts linear billing mechanism, namely the network service expense paid of user is directly proportional with its network traffics.For behavioural characteristic Γ { σ ₀, η ₀The user, the network profit of operator can be expressed as

With a=p/ σ ₀η ₀And formula (5) is brought into and can be got

$R\left(a\right)={\mathrm{\&sigma;}}_{0}a(\ln a-\ln (1-a-\sqrt{1-2a})),$ $a=\frac{p}{{\mathrm{\&sigma;}}_0{\mathrm{\&eta;}}_0}\&GreaterEqual;1/2---\left(7\right)$

According to the description of formula (7), Fig. 8 has provided the relation between network rate p and network profit R.As can be seen from the figure along with the increase of network rate p, network profit R increases afterwards first and reduces, and R will obtain maximum around here.Simultaneously, also can draw and have optimum toll rate

So that operator obtains maximum return.

Differentiate can get to formula (7)

$R^{\&prime;}\left(p\right)=\frac{1}{{\mathrm{\&eta;}}_0}(\ln a-\ln (1-a-\sqrt{1-2a})+\frac{1}{\sqrt{1-2a}(1-a+\sqrt{1-2a})})---\left(8\right)$

Solving equation R ' (p)=0 just can get

The time, network is obtained maximum return:

max R(p)＝0.4477σ ₀ (9)

This moment, the bandwidth of customer consumption was

$x\left(\hat{p}\right)=1.5433/{\mathrm{\&eta;}}_0---\left(10\right)$

According to the description of formula (8), Fig. 9 has provided growth rate of earnings R ' (p) and the relation of network rate p.As can be seen from the figure, at optimum toll rate

Near R ' (p) changes slowly, and the variation of network rate p is little on the impact of carrier network income.For example, operator will keep network profit and be lower than 5% loss in revenue, and the excursion of network rate only need to satisfy Get final product.

The growth rate that network profit changes along with the network rate

With carrier network loss in revenue rate R _LBe defined as non-optimal price and optimal price

Lower operator obtains the normalization difference of income.Fig. 6 has provided the rate relativity shift

With R _LBetween relation.Consider the factor of network cost, operator is in the situation of keeping identical loss in revenue, and price can suitably be higher than

Thereby save Internet resources and reduce network cost.

For ease of the statement of back, order

$f\left(a\right)=a(\ln a-\ln (1-a-\sqrt{1-2a}));$ $a=\frac{p}{{\mathrm{\&sigma;}}_0{\mathrm{\&eta;}}_0}\&GreaterEqual;1/2---\left(11\right)$

R this moment (a) can be designated as

R(a)＝σ ₀f(a) $a=\frac{p}{{\mathrm{\&sigma;}}_0{\mathrm{\&eta;}}_0}\&GreaterEqual;1/2---\left(12\right)$

Basic metering period [0, T] is divided into the n section, uses i=1,2 ..., n comes each section is numbered.Suppose that the user has different network behavior feature Γ { σ in each section _i, η _i; I=1,2 ..., n; The optimum toll rate corresponding with day part just is so

I=1,2 ..., n, this moment, the income statement of operator was shown

$R_{\lim }=\frac{T}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{{\hat{p}}_i}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right)=\frac{T}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}0.4477{\mathrm{\&sigma;}}_i---\left(13\right)$

Formula (13) is operator at network behavior feature Γ { σ _i, η _i; I=1,2 ..., the income limit under the n, it will be by setting the rate of a correspondence to each the period i in the basic cycle

Realize.The network rate changes on the impact of network profit as shown in Figure 6.

3.2 the time constant billing mechanism loss in revenue

Be that the nontraffic sensitive of constant billing mechanism is p if operator adopts, this moment, nontraffic sensitive was with respect to optimum toll rate

Relativity shift can be expressed as It satisfies

$({\hat{p}}_i-p)/{\hat{p}}_i\&Element;[1-p/\min \{{\hat{p}}_i\},$ $1-p/\max \left\{{\hat{p}}_i\right\}]---\left(14\right)$

Formula (14) and Fig. 6 are contrasted the network profit loss that causes in the time of just can estimating roughly constant billing mechanism when operator adopts.Be not difficult to find out, with user behavior feature Γ { σ _i, η _iBecome optimum toll rate when corresponding

Variance is less, also is user bandwidth efficient { σ _iη _iVariance is less; The network profit loss of taking nontraffic sensitive to cause is also less.

Verified for specific user behavior parameter σ in 2.1 _i, η _i, total with it optimum toll rate of correspondence that exists

So that the network profit R (p of operator _i) obtain maximum.By protruding optimum theory as can be known, for user behavior characteristic sequence Γ { σ _i, η _i; I=1,2 ..., n also certainly exists optimum nontraffic sensitive

So that the income of operator in basic metering period obtains maximum.At this moment, the income of operator can be expressed as:

$R_{\max }=\frac{T}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{\hat{p}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right)---\left(15\right)$

$\hat{P}\&Element;\left[0.2901\min \right\{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i\},$ 0.2901max{σ _iη _i}]

The below will be when adopting loss in revenue during constant billing mechanism do quantitative calculating.Obviously, operator adopts optimum nontraffic sensitive

The maximum return R that obtains _MaxInevitable less than income limit R _Lim, by comparing R _MaxAnd R _LimJust can calculate the loss in revenue of constant billing mechanism when adopting.Therefore, adopt the loss in revenue rate of nontraffic sensitive to be expressed as

${\mathrm{\&eta;}}_L=1-\frac{R_{\max }}{R_{\lim }}---\left(16\right)$

With formula (13) with (15) bring formula (16) into and abbreviation can get

${\mathrm{\&eta;}}_L=1-\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{\hat{p}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right)}{0.4477\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i}---\left(17\right)$

By formula (17), in the time of just can calculating when adopting constant billing mechanism, the loss in revenue rate η of operator _L

2.3 half becomes the loss in revenue of billing mechanism

Because operator carries out charging to a plurality of basic cycles [0, T] that link together.Half two sections systems becomes under the billing mechanism, and " the corresponding period should be { n with rate p ' and p respectively ₂+ 1 ..., n, 1,2 ..., n ₁, { n ₁+ 1 ..., n ₂, as shown in Figure 7.

This moment, the network profit of operator can be expressed as

$R^{\&prime;}+R^{\&prime;\&prime;}=\frac{T}{n}(\underset{i=1}{\overset{{\mathrm{<figure-callout\; id="1"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_1}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{p^{\&prime;}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right)+\underset{i={\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_2+1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{p^{\&prime;}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right)+\underset{i={\mathrm{<figure-callout\; id="1"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_1+1}{\overset{{\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_2}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{p^{\&prime;\&prime;}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right))---\left(18\right)$

Operator not only will determine optimum toll rate in order to realize the maximization of income

Also to choose optimum subsection point

The network profit maximization problems of this moment can be described as

max(R′+R″) (19)

Variable: p ', p ", n ₁, n ₂, $p^{\&prime;},p^{\&prime;\&prime;}\&Element;\left[\min \right\{{\hat{p}}_i\},\max \{{\hat{p}}_i\left\}\right]$

Simultaneous formula (18), (19) result that just can be optimized

At this moment, operator is because take the loss in revenue rate of two sections nontraffic sensitives processed to be expressed as

${\mathrm{\&eta;}}_{2L}=1-\frac{\max (R^{\&prime;}+R^{\&prime;\&prime;})}{R}---\left(20\right)$

With max (R '+R ") and R _LimCan obtain after bringing formula (20) and simplification into

${\mathrm{\&eta;}}_{2L}=1-\frac{\underset{i=1}{\overset{{\mathrm{<figure-callout\; id="1"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_1}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{{\hat{p}}^{\&prime;}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right)+\underset{i={\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_2+1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{{\hat{p}}^{\&prime;}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right)+\underset{i={\mathrm{<figure-callout\; id="1"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_1+1}{\overset{{\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="DEST\_PATH\_HDA00002338256600012.png,DEST\_PATH\_HDA00002338256600031.png"\; state="\{\{state\}\}">n</figure-callout>}}_2}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_{i}f\left(\frac{{\hat{p}}^{\&prime;\&prime;}}{{\mathrm{\&sigma;}}_i{\mathrm{\&eta;}}_i}\right)}{0.4477\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i}---\left(21\right)$

By formula (21), just can calculate for specific user behavior feature Γ { σ _i, η _i; I=1,2 ..., n, operator is because adopting the loss in revenue of two sections nontraffic sensitives processed.

Loss in revenue rate η greater than two sections nontraffic sensitives _MLComputational methods and the billing mechanism of two sections systems similar, repeat no more here.Can predict, along with billing mechanism segments m levels off to n, the network profit loss late η of operator _MLTo level off to 0.Obtaining { η _ML; M=1,2 ..., behind the n, considering the perception degree that the user changes network price the resistance of network price frequent variations and user, suitable segments can be selected to the requirement of network profit loss late according to own by operator.

Claims (5)

1. the at times charging method of an associated user behavior is characterized in that, in the mobile phone Operation Network, charging platform adopts the related mobilephone terminal user of following means and network billing data:

(1) obtains relation between customer consumption bandwidth and network rate and the relation between network rate and operator's income by network utility function and user satisfaction function S (x);

(2) thus by the network rate mechanism of action of user behavior being come user's network behavioral data in basic metering period processed obtains the user network behaviors feature parameter;

(3) under specific user's behavioural characteristic, obtain the optimum toll rate in the basic cycle And the network profit limit of operator $R_{\lim }=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}R\left({\hat{p}}_i\right);$

(4) obtain when adopting different segments by optimizing, at times the network profit of billing mechanism, segmented mode and the rate corresponding with day part;

(5) under different segments, network profit and income limit R that at times billing mechanism optimization is obtained _LimCompare and obtain corresponding loss in revenue rate η _ML

2. the at times charging method of associated user behavior according to claim 1, it is characterized in that, pass through extent function S (x) differentiate to the user in the described step (1), and solving equation S ' (x)=0 obtains the bandwidth of user selection consumption and the relation between the relation between the network rate and network rate and operator's income.

3. the at times charging method of associated user behavior according to claim 1 is characterized in that, in the described step (2) by the user at two kinds of different rate p _a, p _bUnder bandwidth selection x (p _a), x (p _b) obtain user's network behavior characteristic parameter.

4. the at times charging method of associated user behavior according to claim 1, it is characterized in that, in the described step (3) by to the network profit of operator and rate concern R (p) differentiate, and solving equation R ' (p)=0 obtains optimum toll rate

Maximum network profit

And the income limit $R_{\lim }=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}R\left({\hat{p}}_i\right).$

5. the at times charging method of associated user behavior according to claim 1 is characterized in that, when operator adopts at times billing mechanism in the described step (4), by separate optimization problem max (R '+R "+...); Variable: p ', p " ..., n ₁, n ₂... } and obtain optimum segmented mode, corresponding rate and network profit rate.

Images