CN104158703A - Computing method of packet error rate based on probability theory and packet collision model - Google Patents

Abstract

The invention relates to a computing method of a packet error rate based on a probability theory and a packet collision model. The computing method aims at solving the problems that when a distance between an interference node and a receiver is unknown or unvalued, the existing computing method is difficult to give a specific value of the packet error rate owing to uncertainty of a power, which reaches the receiver. The computing method comprises the following steps of step 1, determining a probability distribution function FPr(x) of a power Pr of an interference signal, which reaches an expected SUN receiver, according to distance distribution between an interference WLAN (Wireless Local Area Network) transmitter and the expected SUN receiver; step 2, determining a probability distribution function of a signal to noise ratio of the expected SUN receiver; step 3, determining a probability distribution function of an upper limit of the packet error rate; step 4, determining a probability distribution function of the packet error rate; and step 5, determining an average value of the packet error rate. The computing method is applied to the field of communication.

Description

A kind of based on probability touch upon bag collision model Packet Error Ratio computational methods

Technical field

The present invention relates to touch upon and wrap the Packet Error Ratio computational methods of collision model based on probability.

Background technology

Intelligent grid is a kind of complicated and meticulous system that has merged electric power, control and the communication technology, and it can monitor and optimize the whole electric power transfer service from power station to terminal use.A main feature of intelligent grid is to carry out real-time transmitted in both directions to electric energy and various data message, thereby makes electric power supply and user's demand reach balance.With electric power transfer supporting, ubiquitous communication network is to realize the intelligentized basis of electrical network.Just current development communication technologies situation, the application electrical network intellectuality of radio communication and network technology is inevitable choice.For promoting the application of wireless communication technology in intelligent grid information access and transmission, ieee standardization tissue is set up special working group---IEEE 802.15.4g intelligent common cause net (Smart Utility Network, SUN) standardization effort group.The new wireless network standards IEEE 802.15.4g that this working group formulates is a global network standard for the application of Based Intelligent Control on a large scale, and its major function is to realize detection and the control of whole system by managing significantly the real-time Transmission of data message in scope.This standard definition a kind of low-power consumption, low message transmission rate, multi-hop wireless network cheaply, its main application scenarios is adjacent region data access and the transmission network in intelligent grid.

The SUN working frequency range of IEEE 802.15.4g prescribed by standard is the common frequency band of exempting from license, and most typical frequency range is ISM 2.4GHz frequency range.In intelligent grid neighborhood networking application, SUN is mainly used in connecting information access and the transmission between intelligent electric meter, by the intelligent electric meter composition wireless multi-hop network in certain geographic range.For a kind of new wireless network standards of exempting from licensed band that is operated in, study the problem that must consider when it is networking with the mutual interference characteristic of other co-existin networks on common frequency band, WLAN (the Wireless Local Area Network) network that is for example applied to equally neighborhood net is also operated in the frequency range of 2.4GHz, thereby wlan network is inevitable to the interference of SUN network.Because SUN is a kind of new network standard, the correlative study of SUN network service behaviour while also interference for the wlan network of random distribution specially at present.

Generally carry out the interference characteristic of evaluating network by Packet Error Ratio (PER), average received time, mean transit delay and the network throughput of data packet transmission, and the PER performance of network other performances such as throughput, time delay are also directly determined.Therefore,, by analyzing the PER characteristic of SUN network node, can draw theoretically the interference characteristic of SUN network internal.Conventionally in the time calculating PER, if at least there is an error bit in a packet, be called an EDP.Packet Error Ratio size not only with the error rate of data packet transmission about also relevant with the length of packet.Ask the conventional method of system Packet Error Ratio to represent by formula below

P _p＝1-(1-P _b) ^N

In formula, P _prepresent Packet Error Ratio, P _bthe error rate of expression system, total bit number that N is packet.This Packet Error Ratio computation model is a very coarse model, has only considered error rate of system characteristic in computational process, does not consider the data packet transmission pattern of two systems, thereby can not reflect exactly the Packet Error Ratio performance of system.

Traditional bag collision model is having significant limitation analyzing aspect Packet Error Ratio, is only applicable to the system works performance of Analysis interference node while fixing to receiver distance.And in practice, as the deployment personnel of SUN network, we are difficult to provide clearly the specifying information of its heterogeneous network (being wlan network) node location, and the position of interfering nodes may be also variable, thereby the traditional bag collision model of very difficult application calculates the Packet Error Ratio of SUN system.For example, coexisting in situation of the SUN shown in Fig. 2 and WLAN, as the deployment personnel of SUN system, expect SUN transmitter T _dwith the position V that expects SUN receiver _rknown, thus between distance r _suncan obtain, and in actual conditions, we are difficult to obtain WLAN interference source T _ipositional information, thereby itself and V _rbetween distance r be unknown, but in most of situation, can analyze the approximate region (i.e. grey circle ring area in figure) that interference source distributes, and then obtain the excursion [R of interference source and receiver distance _min, R _max], in the situation that cannot obtaining interference source actual position, can suppose that interference source is random existence in this region, be interval [R apart from r _min, R _max] upper equally distributed stochastic variable.Coexist in scene this, original bag collision model due to its cannot Adaptive change the error rate lost problem-solving ability.

Summary of the invention

The object of the invention is to solve in the time that interfering nodes arrives receiver apart from the unknown or is underrange, due to the uncertainty of its arrival receiver power, be difficult to provide the problem of the concrete numerical value of Packet Error Ratio by existing computational methods, and a kind of average Packet Error Ratio of setting up by join probability opinion and existing bag collision model and the computational methods of Packet Error Ratio distribution function are provided, its concrete grammar carries out according to following steps:

Above-mentioned goal of the invention is achieved through the following technical solutions:

Step 1, according to disturbing WLAN transmitter to arrive the power P of expectation SUN receiver with expecting the range distribution between SUN receiver to determine interference signal _rprobability-distribution function

Step 2, arrive the power P of expecting SUN receiver according to interference signal _rprobability-distribution function to determine the probability-distribution function of the Signal to Interference plus Noise Ratio of expecting SUN receiver;

Step 3, determine the probability-distribution function of upper-bound on BER according to the probability-distribution function of Signal to Interference plus Noise Ratio of expecting SUN receiver;

The probability-distribution function of collision model and upper-bound on BER is wrapped in step 4, combination, determines the probability-distribution function of Packet Error Ratio;

Step 5, obtain the mean value of Packet Error Ratio.

Invention effect

The Packet Error Ratio algorithm that the present invention proposes is compared conventional data packet collision model Packet Error Ratio computational methods, on the basis of considering data packet collision, considered interference source impact on packet Packet Error Ratio to the unfixed situation of receiver distance simultaneously, go out to exist the heterogeneous network of random distribution interference source from the angle analysis of probability theory, interference signal arrives the power distribution situation of receiver, and then obtain the distribution situation of the error rate, the distribution of the collision time obtaining in conjunction with original bag collision model, obtains more accurate system Packet Error Ratio numerical value.Packet Error Ratio model based on data packet collision, has not only considered the error ratio characteristic of system, and considers the timing relationship of the data packet transmission of two phase mutual interference, the i.e. mean collision time of packet.For example, in the time having and only have a WLAN interfering nodes job, within a SUN packet duration, only do not have and exist with the collision of WLAN interfering data bag and with WLAN interfering data bag collision both of these case.If use N ₀, N ₁be illustrated respectively in the bit number that does not bump with WLAN in SUN packet and bump with WLAN, the computing formula of Packet Error Ratio can further be improved and be

$P_p=1-{(1-P_{b,1})}^{N_1}{(1-P_{b,0})}^{N_0}$

Wherein P _{b, 0}, P _{b, 1}the expression that distributes is noiseless and go alone the error rate of disturbing while existence.The present invention can not only analyze the service behaviour that has master network while disturbing network preferably, and is having directive significance aspect master network layout.As shown in Figure 3-4, consider that SUN system patient maximum Packet Error Ratio in the situation that of normal work is 0.1, for ensureing the normal work of SUN system, disturb WLAN transmitter at least will be distributed in outside the region apart from SUN receiver 47m～77m, thereby provide guidance for the deployment of SUN system, WLAN interference source is distributed in apart from the annular region of SUN receiver 50m～100m time, the system Packet Error Ratio average that application the present invention tries to achieve, simulation result under Packet Error Ratio probability-distribution function and this condition, can find out from the probability-distribution function curve of Packet Error Ratio, the likelihood ratio that Packet Error Ratio is less is larger, and very difficult generation of situation that Packet Error Ratio is greater than 0.6, that is to say, the interference that is subject to WLAN in the most of the time of SUN system works is all little, can find out simultaneously, gap between system Packet Error Ratio average and simulation result that application the present invention tries to achieve is less, correctness and the validity of algorithm of the present invention are proved.

Brief description of the drawings

Fig. 1 is flow chart of the present invention;

Fig. 2 is SUN and the WLAN topology diagram while coexisting, V _r, T _dand T _isUN receiver is expected in representative respectively, expects SUN transmitter, disturbs WLAN transmitter;

Fig. 3 is the comparison of algorithm of the present invention and simulation result and traditional algorithm;

Fig. 4 is probability-distribution function and the average Packet Error Ratio of Packet Error Ratio.

Embodiment

Embodiment one: a kind of of present embodiment touches upon and wrap the Packet Error Ratio computational methods of collision model based on probability, specifically prepares according to following steps:

Step 1, according to disturbing WLAN transmitter to arrive the power P of expectation SUN receiver with expecting the range distribution between SUN receiver to determine interference signal _rprobability-distribution function

Step 2, arrive the power P of expecting SUN receiver according to interference signal _rprobability-distribution function to determine the probability-distribution function of the Signal to Interference plus Noise Ratio of expecting SUN receiver;

Step 3, determine the probability-distribution function of upper-bound on BER according to the probability-distribution function of Signal to Interference plus Noise Ratio of expecting SUN receiver;

The probability-distribution function of collision model and upper-bound on BER is wrapped in step 4, combination, determines the probability-distribution function of Packet Error Ratio;

Step 5, obtain the mean value of Packet Error Ratio.

Embodiment two: present embodiment is different from embodiment one: the distance r in described step 1 between interference source and expectation SUN receiver is uniformly distributed random variable, the probability-distribution function F of r in annulus _r(x) be

$F_r\left(x\right)=\frac{x-R_{\min }}{R_{\max }-R_{\min }};$

In conjunction with interference signal transmitting power, according to the path loss model of signal transmission, obtain interference signal and arrive the power P of expecting SUN receiver _rdistribution:

Some signal transmission power P that disturb _trepresent, interference signal arrives the power P of expecting SUN receiver _rfor wherein b is path loss coefficient;

Be [R apart from the span of r _min, R _max], [R _min, R _max] be interference source and the excursion of expecting SUN receiver distance, P _r[P between closed area for corresponding span _{r, min}, P _{r, max}] represent, wherein

Arrive according to obtain interference signal apart from the distribution of r the power P of expecting SUN receiver _rprobability-distribution function for

$\begin{array}{c}{F}_{P_r}\left(x\right)=P(\frac{P_t}{r^b}\≤x)\\ =P(r\≥{\left(\frac{P_t}{x}\right)}^{\left(\frac{1}{b}\right)})\\ =\frac{R_{\max }-{\left(\frac{P_t}{x}\right)}^{\left(\frac{1}{b}\right)}}{R_{\max }-R_{\min }}\end{array}.$

Embodiment three: present embodiment is different from embodiment one or two: the value of described path loss coefficient b is 4.

Embodiment four: present embodiment is different from embodiment one to three: the expression formula of expecting the Signal to Interference plus Noise Ratio of SUN receiver in described step 2 is wherein SINR is Signal to Interference plus Noise Ratio, N _rthe power of the white Gaussian noise receiving for receiving terminal, P _defor the power of desired signal arrival receiver;

In like manner, if [the SINR for span of Signal to Interference plus Noise Ratio _min, SINR _max] represent, arrive the power P of expecting SUN receiver according to interference signal _rprobability-distribution function, the probability-distribution function that obtains the Signal to Interference plus Noise Ratio of expecting SUN receiver is

$\begin{array}{c}{F}_{\mathrm{SINR}}\left(x\right)=P(\frac{P_{\mathrm{de}}}{P_r+N_r}\≤x)\\ =P(P_r\≥\frac{P_{\mathrm{de}}}{x}-N_r)\\ =\frac{1}{R_{\max }-R_{\min }}\·[{P_t}^{\left(\frac{1}{b}\right)}{(\frac{P_{\mathrm{de}}}{x}-N_r)}^{\left(\frac{1}{b}\right)}-R_{\min }]\end{array}.$

Embodiment five: present embodiment is different from embodiment one to four: described P _devalue is tried to achieve according to selected path loss model and the transmitting power setting.

Embodiment six: present embodiment is different from embodiment one to five: try to achieve after the Signal to Interference plus Noise Ratio that arrives expectation SUN receiver in described step 3, can be according to the modulation demodulation system of SUN receiver, draw the calculation relational expression between Signal to Interference plus Noise Ratio and the error rate, the physical layer model of SUN node is MR-FSK (the Multi-Regional Frequency Shift Keying) physical layer model specifying in its normative document;

Error rate P when single interference source exists _{b, 1}for:

$P_{b,1}=Q\left(\sqrt{\mathrm{SINR}}\right)$

Wherein, the Q function that Q (x) is Gaussian Profile; The Q function of Gaussian Profile has relation as follows

$Q\left(x\right)\≤\frac{1}{2}\exp (-\frac{x^2}{2})$

Thereby can obtain the upper limit of the error rate, i.e. P _{b, up}

$P_{b,\mathrm{up}}=\frac{1}{2}\exp (-\frac{\mathrm{SINR}}{2})$

In like manner, if P _{b, up}[P for span _{b, up, min}, P _{b, up, max}] represent, can obtain the probability-distribution function of upper-bound on BER according to the distribution of the Signal to Interference plus Noise Ratio obtaining for

$\begin{array}{c}{F}_{P_{b,\mathrm{up}}}\left(x\right)=P(\frac{1}{2}\exp (-\frac{\mathrm{SINR}}{2})\≤x)\\ =P(\mathrm{SINR}\≥-2\ln \left(2x\right))\\ =1-F_{\mathrm{SINR}}(-2\ln 2x)\\ =\frac{1}{R_{\max }-R_{\min }}\·[R_{\max }-{P_t}^{\left(\frac{1}{b}\right)}{(\frac{P_{\mathrm{de}}}{-2\ln 2x}-N_r)}^{(-\frac{1}{b})}]\end{array}.$

Embodiment seven: present embodiment is different from embodiment one to six: wrap collision model in described step 4 and be specially:

Go alone in the situation of disturbing existence when considering, within the expected data bag duration, only have collision and do not collide both of these case to exist;

If expected data bag total time length is L _s, every bit duration is T _b, the total bit number of expected data bag if use N ₀represent in expected data bag the bit number not bumping with interfering data bag, N ₁represent the bit number bumping with interfering data bag in expected data bag, obviously have equation to be related to N=N ₁+ N ₀set up;

In conjunction with the bit number of expected data bag and the error rate of corresponding bit, can obtain the Packet Error Ratio P of expected data bag _pfor

$P_p=1-{(1-P_{b,1})}^{N_1}{(1-P_{b,0})}^{N_0}$

Wherein P _{b, 1}, P _{b, 0}distribute to represent to go alone and disturb and error rate when noiseless existence.

Embodiment eight: present embodiment is different from embodiment one to seven: P in described step 4 _{b, 0}calculating and step 3 in P _{b, 1}computational methods identical, interference signal is wherein arrived to the power P of receiver _rnumerical value bring 0 into; In the situation that Signal to Interference plus Noise Ratio is greater than 1, go alone absolute error between upper-bound on BER value and the actual value of disturbing while existence little, and along with the increase error of signal to noise ratio reduces rapidly;

In real work situation, Signal to Interference plus Noise Ratio is always greater than 1, thereby the upper limit of the available error rate replaces the error rate in above formula, thereby obtains the expression formula of Packet Error Ratio,

$P_{p,\mathrm{up}}=1-{(1-P_{b,\mathrm{up}})}^{N_1}{(1-P_{b,0})}^{N_0}.$

Embodiment nine: present embodiment is different from embodiment one to eight: the described probability-distribution function that utilizes the mode of substitution of variable to obtain Packet Error Ratio:

(1) first order p _{p, up}=1-P; Then make Z=lnP, have

Z＝(N-N ₁)ln(1-P _b,0)+N ₁ln(1-P _b,up)

＝Nln(1-P _b,0)-N ₁[ln(1-P _b,0)-ln(1-P _b,up)]

Make again T=ln (1-P _{b, 0})-ln (1-P _{b, up}), consider and level off to 0 time as x have ln (1-x) ≈-x; And error rate P _{b, up}value is all very little, so there is T ≈ ln (1-P _{b, 0})+P _{b, up}; The Packet Error Ratio P of variable T _{p, up}probability-distribution function F _t(x) be

$\begin{array}{c}{F}_T\left(x\right)=P(P_{b,\mathrm{up}}+\ln (1-P_{b,0})\≤x)\\ =F_{P_{b,\mathrm{up}}}(x-\ln (1-P_{b,0}))\\ =\frac{1}{R_{\max }-R_{\min }}\·[R_{\max }-{P_t}^{\left(\frac{1}{b}\right)}{(\frac{P_{\mathrm{de}}}{-2\ln 2(x-\ln (1-P_{b,0}))}-N_r)}^{(-\frac{1}{b})}]\end{array}$

If [the T for span of T _min, T _max] represent T _min=P _{b, up, min}+ ln (1-P _{b, 0}), T _max=P _{b, up, max}+ ln (1-P _{b, 0});

(2) then define three variables D, D ₁, D ₂, wherein D=lnN ₁t, D ₁=lnN ₁, D ₂=lnT; Obviously there is equation D=D ₁+ D ₂set up; According to the Packet Error Ratio probability-distribution function F of the span of the variable T obtaining and variable T _t(x), can obtain variables D ₂span and variables D ₂packet Error Ratio probability distribution equation; If D ₂[D for span _{2, min}, D _{2, max}] represent D _{2, min}=lnT _min, D _{2, max}=lnT _max, variables D ₂packet Error Ratio probability distribution equation for

$\begin{array}{c}{F}_{D_2}\left(x\right)=P(\ln T\≤x)\\ =F_T\left(e^x\right)\\ =\frac{1}{R_{\max }-R_{\min }}\·[R_{\max }-{P_t}^{\left(\frac{1}{b}\right)}{(\frac{P_{\mathrm{de}}}{-2\ln 2(e^x-\ln (1-P_{b,0}))}-N_r)}^{(-\frac{1}{b})}]\end{array}$

The bit number that expected data bag is interior and interfering data bag is overlapping is uniformly distributed in its span, in other words, and N ₁for equally distributed discrete random variable in interval [1, N]; Therefore variables D ₁span be [0, lnN], variables D ₁the Packet Error Ratio probability distribution equation of getting arbitrary value in this interval is

$P(D_1=\ln k)=\frac{1}{N},k\∈[1,N]$

Then can obtain the span of D and the Packet Error Ratio probability distribution Equation f of variables D _d(x); If [the D for span of D _min, D _max] represent D _min=D _{2, min}, D _max=D _{2, max}+ lnN; The Packet Error Ratio probability distribution Equation f of variables D _d(x) be

$\begin{array}{c}{f}_D\left(x\right)=\underset{k=1}{\overset{N}{\mathrm{\&Sigma;}}}P(D_1=\ln k)f_{D_2}(x-\ln k)\\ =\left\{\begin{array}{cc}\frac{1}{N}{f}_{D_2}\left(x\right),& D_{2,\min }\&le;x<D_{2,\min }+\ln 2\\ \frac{1}{N}(f_{D_2}\left(x\right)+f_{D_2}(x-\ln 2)),& D_{2,\min }+\ln 2\&le;x<D_{2,\min }+\ln 3\\ .& .\\ .& .\\ .& .\\ \frac{1}{N}\underset{i=1}{\overset{N-1}{\mathrm{\&Sigma;}}}{f}_{D_2}(x-\ln i),& D_{2,\min }+\ln (N-1)\&le;x<D_{2,\min }+\ln N\\ \frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{f}_{D_2}(x-\ln i),& D_{2,\min }+\ln N\&le;x<D_{2,M}\\ \frac{1}{N}\underset{i=2}{\overset{N}{\mathrm{\&Sigma;}}}{f}_{D_2}(x-\ln i),& D_{2,\min }\&le;x<D_{2,\max }+\ln 2\\ .& .\\ .& .\\ .& .\\ \frac{1}{N}{f}_{D_2}(x-\ln N),& D_{2,\max }+\ln (N-1)\&le;x\&le;D_{2,\max }+\ln N\end{array}\right.\end{array}$

The probability-distribution function F of variables D _d(x) be

$\begin{array}{c}{F}_D\left(x\right)={\&Integral;}_{D_{\min }}^x{f}_D\left(t\right)\mathrm{dt}\\ =\left\{\begin{array}{cc}\frac{1}{N}{F}_{D_2}\left(x\right),& D_{2,\min }\&le;x<D_{2,\min }+\ln 2\\ \frac{1}{N}(F_{D_2}\left(x\right)+F_{D_2}(x-\ln 2)),& D_{2,\min }+\ln 2\&le;x<D_{2,\min }+\ln 3\\ .& .\\ .& .\\ .& .\\ \frac{1}{N}\underset{i=1}{\overset{N-1}{\mathrm{\&Sigma;}}}{F}_{D_2}(x-\ln i),& D_{2,\min }+\ln (N-1)\&le;x<D_{2,\min }+\ln N\\ \frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{F}_{D_2}(x-\ln i),& D_{2,\min }+\ln N\&le;x<D_{2,M}\\ \frac{1}{N}+\frac{1}{N}\underset{i=2}{\overset{N}{\mathrm{\&Sigma;}}}{F}_{D_2}(x-\ln i),& D_{2,\min }\&le;x<D_{2,\max }+\ln 2\\ .& .\\ .& .\\ .& .\\ \frac{N-1}{N}+\frac{1}{N}{F}_{D_2}(x-\ln N),& D_{2,\max }+\ln (N-1)\&le;x\&le;D_{2,\max }+\ln N\end{array}\right.\end{array}$

(3) relational expression of variable Z and D is Z=Nln (1-P _{b, 0})-e ^d, therefore can determine the span [Z of variable Z _min, Z _max] and the Packet Error Ratio probability-distribution function of variable Z, wherein the Packet Error Ratio probability-distribution function F of variable Z _z(x) be

F _Z(x)＝1-F _D(ln(Nln(1-P _b,0)-x))

And having equation, variable Z and Packet Error Ratio be related to P _{p, up}=1-e ^z, in like manner can determine P _{p, up}span [P _{p, up, min}, P _{p, up, max}] and Packet Error Ratio probability-distribution function, wherein $P_{p,\mathrm{up},\min }=1-e^{Z_{\max }},P_{p,\mathrm{up},\max }=1-e^{Z_{\min }},$ The Packet Error Ratio probability-distribution function of variable Z for

$F_{P_{p,\mathrm{up}}}\left(x\right)=1-F_Z\left(\ln (1-x)\right).$

Embodiment ten: present embodiment is different from embodiment one to nine: the interference strength and the service behaviour that reflect system under present case in described step 5 with the mean value of Packet Error Ratio; In conjunction with the probability-distribution function of variables D and with the relation of Packet Error Ratio probability-distribution function, can draw the average value P of Packet Error Ratio _p,vfor

$\begin{array}{c}{P}_{p,v}={\&Integral;}_{D_{\min }}^{D_{\max }}{f}_D\left(x\right)[1-e^{(N\ln (1-P_{b,0})-e^x)}]\mathrm{dx}\\ =\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\&Sigma;}}}{\&Integral;}_{D_{2,\min +\ln k}}^{D_{2,\max }+\ln k}{f}_{D_2}(x-\ln k)[1-e^{(N\ln (1-P_{b,0})-e^x)}]\mathrm{dx}\\ 1-\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\&Sigma;}}}{\&Integral;}_{D,2\min +\ln k}^{D_{2,\max }+\ln k}{f}_{D_2}(x-\ln k)e^{(N\ln (1-P_{b,0})-e^x)}\mathrm{dx}\end{array}$

Adopt following examples to verify beneficial effect of the present invention:

Embodiment mono-:

The present embodiment is a kind of based on the touch upon Packet Error Ratio computational methods of data packet collision model of probability, specifically prepares according to following steps:

L-G simulation test compliance test result

Table 1

Adopt the simulated conditions of table 1, selected SUN physical layer mode of operation is MR-FSK pattern 1, and what the frequent section of work was 2.4GHz exempts to permit common frequency band, tries to achieve noise power N by following formula _rexpect that with SUN transmitter arrives the power of receiver:

N _r＝BkT

Wherein, k is Boltzmann constant;

$P_{\mathrm{de}}=\frac{P_{\mathrm{sun}}}{r_{\mathrm{sun}}^b}$

Wherein, r _sunfor SUN expects the distance between transmitter and receiver.

The Packet Error Ratio characteristic curve of bringing above parameter into obtain in algorithm of the present invention disturbed SUN receiving node as shown in Figure 3.

Fig. 3 has provided theoretical Packet Error Ratio value that the Packet Error Ratio theoretical calculation model invented obtains and emulation and has tried to achieve the contrast situation of the Packet Error Ratio value that Packet Error Ratio value and conventional method obtain, and wherein the Packet Error Ratio value of simulation curve is the mean value of 1000 simulation results.As can be seen from Figure 3, the Packet Error Ratio value difference that the theoretical Packet Error Ratio value that the inventive method is tried to achieve and emulation obtain is apart from less, while considering calculating, application is the upper limit of the error rate, the theoretical value that the present invention tries to achieve is greater than the result that emulation obtains really, thereby has proved the correctness of the Packet Error Ratio computation model based on probability theory and bag collision model of the present invention.The Packet Error Ratio value that the Packet Error Ratio value that the present invention tries to achieve is tried to achieve than conventional method is closer to actual value, because traditional Packet Error Ratio is only considered data packet collision, and specifically expected data bag is not carried out to segment processing according to collision situation, and only consider that interference position fixes this special circumstances while calculating the error rate.In the situation that expecting that transmission range is fixing, along with disturbing the increase of the distance between transmitting node distributed areas and disturbed node, the Packet Error Ratio value of disturbed receiving node decreases.Simultaneously, consider that SUN system patient maximum Packet Error Ratio in the situation that of normal work is 0.1, for ensureing the normal work of SUN system, disturb WLAN transmitter at least will be distributed in outside the region apart from SUN receiver 47m～77m, thereby provide guidance for the deployment of SUN system.

Fig. 4 has provided WLAN interference source and has been distributed in apart from the annular region of SUN receiver 50m～100m time, the simulation result under system Packet Error Ratio average, Packet Error Ratio probability-distribution function and this condition that application the present invention tries to achieve.Can find out from the probability-distribution function curve of Packet Error Ratio, the likelihood ratio that Packet Error Ratio is less is larger, and very difficult generation of situation that Packet Error Ratio is greater than 0.6 that is to say, the interference that is subject to WLAN in the most of the time of SUN system works is all little.Can find out, the gap between system Packet Error Ratio average and simulation result that application the present invention tries to achieve is less simultaneously, has proved correctness and the validity of algorithm of the present invention.

Claims (10)

1. based on the touch upon Packet Error Ratio computational methods of bag collision model of probability, it is characterized in that: a kind ofly specifically carry out according to following steps based on the touch upon Packet Error Ratio computational methods of bag collision model of probability:

Step 1, according to disturbing WLAN transmitter to arrive the power P of expectation SUN receiver with expecting the range distribution between SUN receiver to determine interference signal _rprobability-distribution function

Step 2, arrive the power P of expecting SUN receiver according to interference signal _rprobability-distribution function to determine the probability-distribution function of the Signal to Interference plus Noise Ratio of expecting SUN receiver;

Step 3, determine the probability-distribution function of upper-bound on BER according to the probability-distribution function of Signal to Interference plus Noise Ratio of expecting SUN receiver;

The probability-distribution function of collision model and upper-bound on BER is wrapped in step 4, combination, determines the probability-distribution function of Packet Error Ratio;

Step 5, obtain the mean value of Packet Error Ratio.

According to claim 1 a kind of based on probability touch upon bag collision model Packet Error Ratio computational methods, it is characterized in that: the distance r in described step 1 between interference source and expectation SUN receiver is uniformly distributed random variable, the probability-distribution function F of r in annulus _r(x) be

$F_r\left(x\right)=\frac{x-R_{\min }}{R_{\max }-R_{\min }};$

In conjunction with interference signal transmitting power, according to the path loss model of signal transmission, obtain interference signal and arrive the power P of expecting SUN receiver _rdistribution:

Some signal transmission power P that disturb _trepresent, interference signal arrives the power P of expecting SUN receiver _rfor wherein b is path loss coefficient;

Be [R apart from the span of r _min, R _max], [R _min, R _max] be interference source and the excursion of expecting SUN receiver distance, P _r[P between closed area for corresponding span _{r, min}, P _{r, max}] represent, wherein

Arrive according to obtain interference signal apart from the distribution of r the power P of expecting SUN receiver _rprobability-distribution function for

$\begin{array}{c}{F}_{P_r}\left(x\right)=P(\frac{P_t}{r^b}\&le;x)\\ =P(r\&GreaterEqual;{\left(\frac{P_t}{x}\right)}^{\left(\frac{1}{b}\right)})\\ =\frac{R_{\max }-{\left(\frac{P_t}{x}\right)}^{\left(\frac{1}{b}\right)}}{R_{\max }-R_{\min }}\end{array}.$

According to claim 2 a kind of based on probability touch upon bag collision model Packet Error Ratio computational methods, it is characterized in that: the value of described path loss coefficient b is 4.

According to claim 3 a kind of based on probability touch upon bag collision model Packet Error Ratio computational methods, it is characterized in that: the expression formula of expecting the Signal to Interference plus Noise Ratio of SUN receiver in described step 2 is wherein SINR is Signal to Interference plus Noise Ratio, N _rthe power of the white Gaussian noise receiving for receiving terminal, P _defor the power of desired signal arrival receiver;

In like manner, if [the SINR for span of Signal to Interference plus Noise Ratio _min, SINR _max] represent, arrive the power P of expecting SUN receiver according to interference signal _rprobability-distribution function, the probability-distribution function that obtains the Signal to Interference plus Noise Ratio of expecting SUN receiver is

$\begin{array}{c}{F}_{\mathrm{SINR}}\left(x\right)=P(\frac{P_{\mathrm{de}}}{P_r+N_r}\&le;x)\\ =P(P_r\&GreaterEqual;\frac{P_{\mathrm{de}}}{x}-N_r)\\ =\frac{1}{R_{\max }-R_{\min }}\&CenterDot;[{P_t}^{\left(\frac{1}{b}\right)}{(\frac{P_{\mathrm{de}}}{x}-N_r)}^{\left(\frac{1}{b}\right)}-R_{\min }]\end{array}.$

According to claim 4 a kind of based on probability touch upon bag collision model Packet Error Ratio computational methods, it is characterized in that: described P _devalue is tried to achieve according to path loss model selected in step 1 and the transmitting power setting.

According to claim 5 a kind of based on probability touch upon bag collision model Packet Error Ratio computational methods, it is characterized in that: in described step 3, try to achieve after the Signal to Interference plus Noise Ratio that arrives expectation SUN receiver, can, according to the modulation demodulation system of SUN receiver, draw the calculation relational expression between Signal to Interference plus Noise Ratio and the error rate;

Error rate P when single interference source exists _{b, 1}for:

$P_{b,1}=Q\left(\sqrt{\mathrm{SINR}}\right)$

Wherein, the Q function that Q (x) is Gaussian Profile; The Q function of Gaussian Profile has relation as follows

$Q\left(x\right)\&le;\frac{1}{2}\exp (-\frac{x^2}{2})$

Thereby can obtain the upper limit of the error rate, i.e. P _{b, up}

$P_{b,\mathrm{up}}=\frac{1}{2}\exp (-\frac{\mathrm{SINR}}{2})$

In like manner, if P _{b, up}[P for span _{b, up, min}, P _{b, up, max}] represent, can obtain the probability-distribution function of upper-bound on BER according to the distribution of the Signal to Interference plus Noise Ratio obtaining for

$\begin{array}{c}{F}_{P_{b,\mathrm{up}}}\left(x\right)=P(\frac{1}{2}\exp (-\frac{\mathrm{SINR}}{2})\&le;x)\\ =P(\mathrm{SINR}\&GreaterEqual;-2\ln \left(2x\right))\\ =1-F_{\mathrm{SINR}}(-2\ln 2x)\\ =\frac{1}{R_{\max }-R_{\min }}\&CenterDot;[R_{\max }-{P_t}^{\left(\frac{1}{b}\right)}{(\frac{P_{\mathrm{de}}}{-2\ln 2x}-N_r)}^{(-\frac{1}{b})}]\end{array}.$

According to claim 6 a kind of based on probability touch upon bag collision model Packet Error Ratio computational methods, it is characterized in that: in described step 4, wrap collision model and be specially:

Go alone in the situation of disturbing existence when considering, within the expected data bag duration, only have collision and do not collide both of these case to exist;

If expected data bag total time length is L _s, every bit duration is T _b, the total bit number of expected data bag if use N ₀represent in expected data bag the bit number not bumping with interfering data bag, N ₁represent the bit number bumping with interfering data bag in expected data bag, obviously have equation to be related to N=N ₁+ N ₀set up;

In conjunction with the bit number of expected data bag and the error rate of corresponding bit, can obtain the Packet Error Ratio P of expected data bag _pfor

$P_p=1-{(1-P_{b,1})}^{N_1}{(1-P_{b,0})}^{N_0}$

Wherein P _{b, 1}, P _{b, 0}distribute to represent to go alone and disturb and error rate when noiseless existence.

According to claim 7 a kind of based on probability touch upon bag collision model Packet Error Ratio computational methods, it is characterized in that: P in described step 4 _{b, 0}calculating and step 3 in P _{b, 1}computational methods identical, interference signal is wherein arrived to the power P of receiver _rnumerical value bring 0 into; In the situation that Signal to Interference plus Noise Ratio is greater than 1, go alone absolute error between upper-bound on BER value and the actual value of disturbing while existence little, and along with the increase error of signal to noise ratio reduces rapidly;

In real work situation, Signal to Interference plus Noise Ratio is always greater than 1, thereby the upper limit of the available error rate replaces the error rate in above formula, thereby obtains the expression formula of Packet Error Ratio,

$P_{p,\mathrm{up}}=1-{(1-P_{b,\mathrm{up}})}^{N_1}{(1-P_{b,0})}^{N_0}.$

According to claim 8 a kind of based on probability touch upon bag collision model Packet Error Ratio computational methods, it is characterized in that: the described probability-distribution function that utilizes the mode of substitution of variable to obtain Packet Error Ratio:

(1) first order p _{p, up}=1-P; Then make Z=lnP, have

Z＝(N-N ₁)ln(1-P _b,0)+N ₁ln(1-P _b,up)

＝Nln(1-P _b,0)-N ₁[ln(1-P _b,0)-ln(1-P _b,up)]

Make again T=ln (1-P _{b, 0})-ln (1-P _{b, up}), consider and level off to 0 time as x have ln (1-x) ≈-x; And error rate P _{b, up}value is all very little, so there is T ≈ ln (1-P _{b, 0})+P _{b, up}; The Packet Error Ratio P of variable T _{p, up}probability-distribution function F _t(x) be

$\begin{array}{c}{F}_T\left(x\right)=P(P_{b,\mathrm{up}}+\ln (1-P_{b,0})\&le;x)\\ =F_{P_{b,\mathrm{up}}}(x-\ln (1-P_{b,0}))\\ =\frac{1}{R_{\max }-R_{\min }}\&CenterDot;[R_{\max }-{P_t}^{\left(\frac{1}{b}\right)}{(\frac{P_{\mathrm{de}}}{-2\ln 2(x-\ln (1-P_{b,0}))}-N_r)}^{(-\frac{1}{b})}]\end{array}$

If [the T for span of T _min, T _max] represent T _min=P _{b, up, min}+ ln (1-P _{b, 0}), T _max=P _{b, up, max}+ ln (1-P _{b, 0});

(2) then define three variables D, D ₁, D ₂, wherein D=lnN ₁t, D ₁=lnN ₁, D ₂=lnT; Obviously there is equation D=D ₁+ D ₂set up; According to the Packet Error Ratio probability-distribution function F of the span of the variable T obtaining and variable T _t(x), can obtain variables D ₂span and variables D ₂packet Error Ratio probability distribution equation; If D ₂[D for span _{2, min}, D _{2, max}] represent D _{2, min}=lnT _min, D _{2, max}=lnT _max, variables D ₂packet Error Ratio probability distribution equation for

$\begin{array}{c}{F}_{D_2}\left(x\right)=P(\ln T\&le;x)\\ =F_T\left(e^x\right)\\ =\frac{1}{R_{\max }-R_{\min }}\&CenterDot;[R_{\max }-{P_t}^{\left(\frac{1}{b}\right)}{(\frac{P_{\mathrm{de}}}{-2\ln 2(e^x-\ln (1-P_{b,0}))}-N_r)}^{(-\frac{1}{b})}]\end{array}$

The bit number that expected data bag is interior and interfering data bag is overlapping is uniformly distributed in its span, in other words, and N ₁for equally distributed discrete random variable in interval [1, N]; Therefore variables D ₁span be [0, lnN], variables D ₁the Packet Error Ratio probability distribution equation of getting arbitrary value in this interval is

$P(D_1=\ln k)=\frac{1}{N},k\&Element;[1,N]$

Then can obtain the span of D and the Packet Error Ratio probability distribution Equation f of variables D _d(x); If [the D for span of D _min, D _max] represent D _min=D _{2, min}, D _max=D _{2, max}+ lnN; The Packet Error Ratio probability distribution Equation f of variables D _d(x) be

$\begin{array}{c}{f}_D\left(x\right)=\underset{k=1}{\overset{N}{\mathrm{\&Sigma;}}}P(D_1=\ln k)f_{D_2}(x-\ln k)\\ =\left\{\begin{array}{cc}\frac{1}{N}{f}_{D_2}\left(x\right),& D_{2,\min }\&le;x<D_{2,\min }+\ln 2\\ \frac{1}{N}(f_{D_2}\left(x\right)+f_{D_2}(x-\ln 2)),& D_{2,\min }+\ln 2\&le;x<D_{2,\min }+\ln 3\\ .& .\\ .& .\\ .& .\\ \frac{1}{N}\underset{i=1}{\overset{N-1}{\mathrm{\&Sigma;}}}{f}_{D_2}(x-\ln i),& D_{2,\min }+\ln (N-1)\&le;x<D_{2,\min }+\ln N\\ \frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{f}_{D_2}(x-\ln i),& D_{2,\min }+\ln N\&le;x<D_{2,M}\\ \frac{1}{N}\underset{i=2}{\overset{N}{\mathrm{\&Sigma;}}}{f}_{D_2}(x-\ln i),& D_{2,\min }\&le;x<D_{2,\max }+\ln 2\\ .& .\\ .& .\\ .& .\\ \frac{1}{N}{f}_{D_2}(x-\ln N),& D_{2,\max }+\ln (N-1)\&le;x\&le;D_{2,\max }+\ln N\end{array}\right.\end{array}$

The probability-distribution function F of variables D _d(x) be

$\begin{array}{c}{F}_D\left(x\right)={\&Integral;}_{D_{\min }}^x{f}_D\left(t\right)\mathrm{dt}\\ =\left\{\begin{array}{cc}\frac{1}{N}{F}_{D_2}\left(x\right),& D_{2,\min }\&le;x<D_{2,\min }+\ln 2\\ \frac{1}{N}(F_{D_2}\left(x\right)+F_{D_2}(x-\ln 2)),& D_{2,\min }+\ln 2\&le;x<D_{2,\min }+\ln 3\\ .& .\\ .& .\\ .& .\\ \frac{1}{N}\underset{i=1}{\overset{N-1}{\mathrm{\&Sigma;}}}{F}_{D_2}(x-\ln i),& D_{2,\min }+\ln (N-1)\&le;x<D_{2,\min }+\ln N\\ \frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{F}_{D_2}(x-\ln i),& D_{2,\min }+\ln N\&le;x<D_{2,M}\\ \frac{1}{N}+\frac{1}{N}\underset{i=2}{\overset{N}{\mathrm{\&Sigma;}}}{F}_{D_2}(x-\ln i),& D_{2,\min }\&le;x<D_{2,\max }+\ln 2\\ .& .\\ .& .\\ .& .\\ \frac{N-1}{N}+\frac{1}{N}{F}_{D_2}(x-\ln N),& D_{2,\max }+\ln (N-1)\&le;x\&le;D_{2,\max }+\ln N\end{array}\right.\end{array}$

(3) relational expression of variable Z and D is Z=Nln (1-P _{b, 0})-e ^d, therefore can determine the span [Z of variable Z _min, Z _max] and the Packet Error Ratio probability-distribution function of variable Z, wherein the Packet Error Ratio probability-distribution function F of variable Z _z(x) be

F _Z(x)＝1-F _D(ln(Nln(1-P _b,0)-x))

And having equation, variable Z and Packet Error Ratio be related to P _{p, up}=1-e ^z, in like manner can determine P _{p, up}span [P _{p, up, min}, P _{p, up, max}] and Packet Error Ratio probability-distribution function, wherein $P_{p,\mathrm{up},\min }=1-e^{Z_{\max }},P_{p,\mathrm{up},\max }=1-e^{Z_{\min }},$ The Packet Error Ratio probability-distribution function of variable Z for

$F_{P_{p,\mathrm{up}}}\left(x\right)=1-F_Z\left(\ln (1-x)\right).$

According to claim 9 a kind of based on probability touch upon bag collision model Packet Error Ratio computational methods, it is characterized in that: the interference strength and the service behaviour that in described step 5, reflect system under present case with the mean value of Packet Error Ratio; In conjunction with the probability-distribution function of variables D and with the relation of Packet Error Ratio probability-distribution function, can draw the average value P of Packet Error Ratio _p,vfor

$\begin{array}{c}{P}_{p,v}={\&Integral;}_{D_{\min }}^{D_{\max }}{f}_D\left(x\right)[1-e^{(N\ln (1-P_{b,0})-e^x)}]\mathrm{dx}\\ =\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\&Sigma;}}}{\&Integral;}_{D_{2,\min +\ln k}}^{D_{2,\max }+\ln k}{f}_{D_2}(x-\ln k)[1-e^{(N\ln (1-P_{b,0})-e^x)}]\mathrm{dx}\\ 1-\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\&Sigma;}}}{\&Integral;}_{D,2\min +\ln k}^{D_{2,\max }+\ln k}{f}_{D_2}(x-\ln k)e^{(N\ln (1-P_{b,0})-e^x)}\mathrm{dx}\end{array}$