CN103888204A - Method for modeling corn field wireless sensor network channel multi-scale fading mold - Google Patents

Abstract

The invention relates to the field of communication, and discloses a method for modeling a corn field wireless sensor network channel multi-scale fading mold. The method includes the steps of S1, key factors causing channel time-space difference are extracted according to environment factors which influence wireless signal transmission in a sampling area; S2, under different environment factors and the key factors, feature data of wireless signal transmission are collected in the sampling area; S3, according to the collected data and the key factors, channel multi-scale fading modeling is carried out. By means of the method, formulized modeling of wireless sensor network signal transmission of a corn field in gradual environment can be achieved, and channel theoretic supports are provided for node deployment, topology control, routing selection and the like in wireless sensor network monitoring application.

Description

The modeling method of the multiple dimensioned decline model of maize field radio sensor network channel

Technical field

The present invention relates to the communications field, be specifically related to the modeling method of the multiple dimensioned decline model of maize field radio sensor network channel.

Background technology

Agricultural land for growing field crops mainly refers to that those carry out the soil of large area crop-planting, and the crop that is generally used for field planting has: wheat, corn, potato etc.The collection of the agriculture field planting environmental information based on wireless sensor network, transmission, monitoring have become the important step of high-quality plant development.Radio communication is the basis of wireless sensor network data transmission, wireless signal is in communication process, be subject to the impact of variety classes barrier that the phenomenons such as reflection, scattering and diffraction in various degree can occur, cause the decline of signal energy, and the variation of time delay, phase place, frequency etc.In farmland, crop all experiences different growth phases every year, and the branches and leaves density variation in these stages is very large, signal is propagated to the impact producing in various degree.

In existing radio network information channel decline Modeling Research, some research emphasis are the correlations between the decline of signal large scale and environmental factor, some research needs distinguishable multipath path, the modeling of signal large scale is only studied in some research, and some research is only carried out the modeling of signal small scale from desirable statistical model.

The shortcoming of existing scheme is:

1, research emphasis is that the scheme of the correlation between signal strength signal intensity and environmental factor is not considered the impact of multipath fading on signal strength signal intensity and signal quality, not accurate enough to choosing of environmental characteristic parameter;

2, the scheme that research need to be differentiated multipath path need to effectively be distinguished each reflection, scattering component, and each paths is carried out to Reconstruction, but in actual applications, especially in farmland production environment, because shelter is intensive, channel condition complexity, without obvious distinguishable multipath path, also just cannot be described reconstruction to channel with limited several reflections, scattering path;

3, the scheme of only studying the modeling of signal large scale is not furtherd investigate discussion by the complicated gradual change envirment factor in farmland to the impact of dissemination channel model.Though the method can be described channel circumstance in large scale, because not carrying out associatedly with site environment parameter, also just cannot carry out forecast analysis to the fading factor in similar scene according to measurable environmental parameter, practical value is low;

The scheme of 4, only carrying out the modeling of signal small scale from desirable statistical model is by setting up how much distributions of scattering object wireless channel, adopt the methods such as ray trace to study modeling, but corn growth circumstance complication gradual change, without obvious distinguishable multipath, be difficult to adopt ray tracing method to carry out small scale modeling.

Summary of the invention

Technical problem to be solved by this invention is that prior art is not considered in corn planting complex environment, multiple circulation way exists and without obvious distinguishable multi-path influence simultaneously, and and various factors under impact and corn planting complex environment that not exclusively two kinds of separate yardsticks declines on signal propagation characteristics, cause radio network information channel to decline modeling accuracy is low even can not modeling.

For this purpose, the present invention proposes the modeling method of the multiple dimensioned decline model of maize field radio sensor network channel, and the method comprises:

S1. according to the envirment factor that affects radio signal propagation in sample area, extract the key factor that causes channel space and time difference;

S2. under the varying environment factor and key factor condition, gather the radio signal propagation characteristic in sample area;

S3. according to the data and the described key factor that gather, carry out the multiple dimensioned decline model modeling of channel.

Wherein, in step S1, described envirment factor comprises the height H of crop in sample area _p, crop leaf area A _l, crop stem stalk area A _c, crop fruit surface area A _f, land area A in sample area _g, distance d between transmitting antenna and reception antenna; Described key factor comprises the height H of blocking of crop _b, crop surface area density indices P SAD, described H _b=H _a-H _p, wherein H _afor antenna height, the computing formula of described PSAD is as follows:

$\mathrm{PSAD}=\frac{A_l+A_c+A_f}{A_G\×H_p}.$

Wherein, in step S2, described radio signal propagation feature comprises: signal strength signal intensity, Packet Error Ratio.

Wherein, in step S2, described sample area comprises H _bthe sample area of > 0 and H _bthe sample area of < 0, described radio signal propagation characteristic comprises H _bradio signal propagation characteristic and the H of the sample area of > 0 _bthe radio signal propagation characteristic of the sample area of < 0.

Wherein, between step S2 and step S3, the method further comprises:

S21. at described H _bthe sample area of > 0, carries out channel large scale decline PL _bcalculating;

S22. obtain described H _bthe radio signal propagation characteristic of the sample area of > 0, carries out matching to the data that gather, and obtains channel large scale decline model.

Wherein, in step S21, described channel large scale decline PL _bfor:

${\mathrm{PL}}_b=10.1g\left(\frac{P_s}{P_r}\right)$

Wherein, P _sfor the transmitting power of transmitting antenna, P _rthe power module of the wireless signal receiving for reception antenna;

Described $P_r=\frac{P_s{G}_s{G}_r{\mathrm{\&lambda;}}^2}{{\left(4\mathrm{\&pi;}\right)}^2{d}^n};$

Wherein, G _s, G _rbe respectively transmitting antenna and receiving antenna gain, λ is wavelength, and d is the distance between transmitting antenna and reception antenna, and n is fading factor.

Wherein, described step S22 comprises: obtain described H _bthe radio signal propagation characteristic of the sample area of > 0, by described radio signal propagation data to described P _rcarry out matching, obtain fading factor n and H _b, PSAD functional relation:

n=a·ln(H _b)+b·ln(PSAD)+c

Described channel large scale decline model is:

PL=(a·ln(H _b)+b·ln(PSAD)+c)lgd+A

Wherein, H _b> 0, PL is the decline of the wireless signal that sends of transmitting antenna, and a, b and c are fitting coefficient, and described fitting coefficient is determined according to PSAD corresponding to plant growth stage, H _bfor blocking height;

Wherein,

f is wireless signal frequency, G _s, G _rbe respectively transmitting antenna and receiving antenna gain, c is light velocity constant.

Wherein, described step S3 comprises:

The envelope gross power P of the wireless signal S31. receiving according to reception antenna, described envelope gross power P comprises large scale component power P _band small scale multipath noise power P _m, build described P _mmodel:

$P_m=\frac{P-P_b}{2}$

Wherein, the formula of described P is as follows:

$P=P_s\&CenterDot;\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{l_i}^2$

Wherein, P _sfor the transmitting power of transmitting antenna, N is the number in multipath path; l _iit is the reflection fading coefficients of i paths;

S32. according to described P _m, build signal to noise ratio snr model;

$\mathrm{SNR}=\frac{P_b}{P_m+\mathrm{AWGN}};$

Wherein, AWGN is additive white Gaussian noise;

S33. according to described SNR, build Packet Error Ratio PER model;

$\mathrm{PER}=\frac{(M-1)}{2}\&times;e^{-\frac{\mathrm{SNR}}{2}};$

Wherein, M is the heterogeneous coefficient of modulation;

S34. obtain described H _bthe radio signal propagation characteristic of the sample area of < 0, obtains the Packet Error Ratio PER measured value of reception antenna;

According to H _benvironment key factor H in the sample area of < 0 _b, PSAD, and channel large scale decline model, obtains described large scale component power P _bpredicted value P _b';

S35. according to described PER measured value and described large scale component power P _bpredicted value P _b', to described PER model and P _mmodel carries out curve fitting, and obtains the multiple dimensioned decline model of channel:

$\mathrm{PL}=(a^{\&prime;}\&CenterDot;\ln \left(H_b\right)+b^{\&prime;}\&CenterDot;\ln \left(\mathrm{PSAD}\right)+c^{\&prime;})1\mathrm{gd}+1g(i\&CenterDot;e^{H_b}+j\&CenterDot;e^{\mathrm{PSAD}}+k)+B$

Wherein, H _b<0, PL is the decline of the wireless signal that sends of transmitting antenna, and a ', b ', c ', i, j, k are fitting coefficient, and described fitting coefficient is determined according to PSAD corresponding to plant growth stage;

Wherein, $B=201\mathrm{gf}-101g\left[\frac{G_s{G}_r{c}^2}{{\left(4\mathrm{\&pi;}\right)}^2}\right]+1g\frac{2}{3},$ F is wireless signal frequency, G _sfor transmitter antenna gain (dBi), c is light velocity constant.

Preferably, described method further comprises: S4. is according to coefficient R ²evaluate the model of described acquisition, wherein said R ²computing formula as follows:

$R^2=1-\frac{\underset{i=1}{\overset{S}{\mathrm{\&Sigma;}}}{[{\mathrm{\&xi;}}_i-\widehat{\mathrm{\&xi;}]}}^2}{\underset{i=1}{\overset{S}{\mathrm{\&Sigma;}}}{[{\mathrm{\&xi;}}_i-\stackrel{\&OverBar;}{\mathrm{\&xi;}}]}^2}$

Wherein, ξ _ithe radio signal propagation characteristic gathering,

described ξ _iregressand value,

for described ξ _imean value, the number that S is collecting sample point.

Than prior art, the beneficial effect of method provided by the invention is: according to complex environment factor impact in maize field as, spacing in the rows, corn growth situation parameter are as dense degree of plant height, blade and fruit etc., and antenna height, the relative distance etc. of monitoring node, extract space and time difference key factor, the wireless channel decline under maize field environment is carried out to modeling.Consider large scale and the multipath fading effect of wireless signal under the complex environment of farmland, selective analysis due to the intensive growth of crop cause without the probability distribution of the small scale effects such as distinguishable multipath to signal of communication quality, carrying out power spectral density solves, finally draw the parameterized model of channel fading characteristic by measured data matching, for subsequent communications prediction of quality in the application of corn planting environment radio sensor network monitoring, node location deployment, network topology control, coverage metric, routing optimization etc. provide basic theory basis and foundation.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

Fig. 1 shows the modeling method flow chart of the multiple dimensioned decline model of maize field radio sensor network channel;

Fig. 2 shows the modeling method flow chart of the multiple dimensioned decline model of maize field radio sensor network channel of embodiment 2;

Fig. 3 shows H _bthe sample area of > 0;

Fig. 4 shows H _bthe sample area of < 0;

Fig. 5 shows signal strength signal intensity under space and time difference condition and the variation relation figure of envirment factor;

Fig. 6 shows Packet Error Ratio and communication distance, blocks the graph of a relation of height.

Embodiment

For making object, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly described, obviously, described embodiment is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment in the present invention, those of ordinary skills, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.

Embodiment 1:

The modeling method of the open multiple dimensioned decline model of maize field radio sensor network channel of the present embodiment, as shown in Figure 1, the method comprises:

The present invention proposes the modeling method of the multiple dimensioned decline model of maize field radio sensor network channel, and the method comprises:

S1. according to the envirment factor that affects radio signal propagation in sample area, extract the key factor that causes channel space and time difference;

S2. under the varying environment factor and key factor condition, gather the radio signal propagation characteristic in sample area;

S3. according to the data and the described key factor that gather, carry out the multiple dimensioned decline model modeling of channel.

Wherein, in step S1, described envirment factor comprises the height H of crop in sample area _p, crop leaf area A _l, crop stem stalk area A _c, crop fruit surface area A _f, land area A in sample area _g, distance d between transmitting antenna and reception antenna; Described key factor comprises the height H of blocking of crop _b, crop surface area density indices P SAD, described H _b=H _a-H _p, wherein H _afor antenna height, the computing formula of described PSAD is as follows:

$\mathrm{PSAD}=\frac{A_l+A_c+A_f}{A_G\&times;H_p}.$

Wherein, in step S2, described radio signal propagation feature comprises: signal strength signal intensity, Packet Error Ratio.

Wherein, in step S2, described sample area comprises H _bthe sample area of > 0 and H _bthe sample area of < 0, described radio signal propagation characteristic comprises H _bradio signal propagation characteristic and the H of the sample area of > 0 _bthe radio signal propagation characteristic of the sample area of < 0.

Wherein, between step S2 and step S3, the method further comprises:

S21. at described H _bthe sample area of > 0, carries out channel large scale decline PL _bcalculating;

S22. obtain described H _bthe radio signal propagation characteristic of the sample area of > 0, carries out matching to the data that gather, and obtains channel large scale decline model.

Wherein, in step S21, described channel large scale decline PL _bfor:

${\mathrm{PL}}_b=10.1g\left(\frac{P_s}{P_r}\right)$

Wherein, P _sfor the transmitting power of transmitting antenna, P _rthe power module of the wireless signal receiving for reception antenna;

Described $P_r=\frac{P_s{G}_s{G}_r{\mathrm{\&lambda;}}^2}{{\left(4\mathrm{\&pi;}\right)}^2{d}^n};$

Wherein, G _s, G _rbe respectively transmitting antenna and receiving antenna gain, λ is wavelength, and d is the distance between transmitting antenna and reception antenna, and n is fading factor.

Wherein, described step S22 comprises: obtain described H _bthe radio signal propagation characteristic of the sample area of > 0, by described radio signal propagation data to described P _rcarry out matching, obtain fading factor n and H _b, PSAD functional relation:

n=a·ln(H _b)+b·ln(PSAD)+c

Described channel large scale decline model is:

PL=(a·ln(H _b)+b·ln(PSAD)+c)lgd+A

Wherein, H _b> 0, PL is the decline of the wireless signal that sends of transmitting antenna, and a, b and c are fitting coefficient, and described fitting coefficient is determined according to PSAD corresponding to plant growth stage, H _bfor blocking height; At H _bwhen > 0, the small scale effect of channel is ignored, and only considers large scale effect, so there is PL _b=PL;

Wherein,

f is wireless signal frequency, G _s, G _rbe respectively transmitting antenna and receiving antenna gain, c is light velocity constant.

Wherein, described step S3 comprises:

The envelope gross power P of the wireless signal S31. receiving according to reception antenna, described envelope gross power P comprises large scale component power P _band small scale multipath noise power P _m, build described P _mmodel:

$P_m=\frac{P-P_b}{2}$

Wherein, the formula of described P is as follows:

$P=P_s\&CenterDot;\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{l_i}^2$

Wherein, P _sfor the transmitting power of transmitting antenna, N is the number in multipath path; l _iit is the reflection fading coefficients of i paths;

S32. according to described P _m, build signal to noise ratio snr model;

$\mathrm{SNR}=\frac{P_b}{P_m+\mathrm{AWGN}};$

Wherein, AWGN is additive white Gaussian noise;

S33. according to described SNR, build Packet Error Ratio PER model;

$\mathrm{PER}=\frac{(M-1)}{2}\&times;e^{-\frac{\mathrm{SNR}}{2}};$

Wherein, M is the heterogeneous coefficient of modulation;

S34. obtain described H _bthe radio signal propagation characteristic of the sample area of < 0, obtains the Packet Error Ratio PER measured value of reception antenna;

According to H _benvironment key factor H in the sample area of < 0 _b, PSAD, and channel large scale decline model, obtains described large scale component power P _bpredicted value P _b;

S35. according to described PER measured value and described large scale component power P _bpredicted value P _b,, to described PER model and P _mmodel carries out curve fitting, and obtains the multiple dimensioned decline model of channel:

$\mathrm{PL}=(a^{\&prime;}\&CenterDot;\ln \left(H_b\right)+b^{\&prime;}\&CenterDot;\ln \left(\mathrm{PSAD}\right)+c^{\&prime;})1\mathrm{gd}+1g(i\&CenterDot;e^{H_b}+j\&CenterDot;e^{\mathrm{PSAD}}+k)+B$

Wherein, H _b<0, PL is the decline of the wireless signal that sends of transmitting antenna, and a ', b ', c ', i, j, k are fitting coefficient, and described fitting coefficient is determined according to PSAD corresponding to plant growth stage;

Wherein, $B=201\mathrm{gf}-101g\left[\frac{G_s{G}_r{c}^2}{{\left(4\mathrm{\&pi;}\right)}^2}\right]+1g\frac{2}{3},$ F is wireless signal frequency, G _sfor transmitter antenna gain (dBi), c is light velocity constant.

Preferably, described method further comprises: S4. is according to coefficient R ²evaluate the model of described acquisition, wherein said R ²computing formula as follows:

$R^2=1-\frac{\underset{i=1}{\overset{S}{\mathrm{\&Sigma;}}}{[{\mathrm{\&xi;}}_i-\widehat{\mathrm{\&xi;}]}}^2}{\underset{i=1}{\overset{S}{\mathrm{\&Sigma;}}}{[{\mathrm{\&xi;}}_i-\stackrel{\&OverBar;}{\mathrm{\&xi;}}]}^2}$

Wherein, ξ _ithe radio signal propagation characteristic gathering,

described ξ _iregressand value,

for described ξ _imean value, the number that S is collecting sample point.

Embodiment 2:

The modeling method of the open multiple dimensioned decline model of maize field radio sensor network channel of the present embodiment, three kinds of basic transmission meanss that affect radio propagation in the present embodiment are reflection, diffraction and scattering.From farm environment self-organizing network signal transmission path, the electromagnetic wave sending from transmitting node is mainly propagated to receiving node with three kinds of different modes by three paths:

Straightline propagation, if any blocking of crop, electromagnetic wave is propagated in the mode of scattering;

Part electromagnetic wave received node after ground return receives;

Part electromagnetic wave is to crop top transmitting, and produces diffraction at crop top end, after received by receiving node.

It should be noted that three kinds of modes exist simultaneously, and also not exclusively separate, so in the time considering Channel Modeling, should consider comprehensively.

The modeling method of the multiple dimensioned decline model of the disclosed maize field radio sensor network channel of the present embodiment, as shown in Figure 2, specifically comprises:

1. space and time difference key factor extracts

The envirment factor that affects radio signal propagation is numerous, how from various factors, to extract the key factor that causes channel circumstance space and time difference, becomes one of emphasis of modeling.

First, whether environment is propagated and is caused that to block be the matter of utmost importance of Channel Modeling signal, simple according to antenna height, cannot directly judge whether environment causes and block signal los path, in the present embodiment to block height H _bfor parameter is carried out Channel Modeling, wherein

H _b=H _a-H _p

Wherein H _afor antenna height, H _pfor plant height.

If H _bbeing greater than 0, as shown in Figure 3, illustrating between communication node and have unobstructed single order Ferned Area, is line-of-sight transmission, mainly considers large scale fading effect; If H _bbe less than 0, as shown in Figure 4, between communication node, be obstructed in single order Ferned Area, must consider large scale decline and multipath fading simultaneously.

The coverage extent that its secondary environment is propagated signal is obviously relevant to the dense degree of crop, and electromagnetic wave can produce corresponding reflection and scattering process after inciding crop surface, the present invention introduces the long-pending dnesity index PSAD(Plant Surface Area Density of crop surface thus) in order to characterize plant growth dense degree, be defined as the long-pending interior crop total surface area of unit group falling bodies, with m ²/ m ³represent, computing formula is as follows:

$\mathrm{PSAD}=\frac{A_l+A_c+A_f}{A_G\&times;H_p}$

Wherein A _lfor Crop leaf area in sample area, A _cfor crop stalk area in sample area, A _ffor crop and fruit surface area in sample area, A _gfor land area in sample area, H _pfor plant height.At H _bbe less than at 0 o'clock, i.e. when signal line-of-sight propagation is obstructed, PSAD parameter has merged the parameters such as leaf area, fruit stem area, plant height, spacing in the rows, can better embody the dense degree of plant growth.Further, because milpa the middle and late growth stage is also inhomogeneous up and down, thus not identical in the value of differing heights PSAD, generally can be divided into bottom, leaf layer, canopy three parts and PSAD is measured and analyze.

The propagation model of wireless channel can be divided into two kinds of large scale propagation model and small scale propagation models.Large-scale model is mainly used in describing the change in signal strength in long distance (hundreds of or a few km) between transmitter and receiver, in general the distance between large scale decline and transmitting antenna and reception antenna is inversely proportional to, and (as seashore and hinterland, city and rural area) has different fading factors in different areas.Small-scale model is used for describing the quick variation of short distance (several wavelength) or short time (second level) interior received signal strength, but these two kinds of models are not separate, in same wireless channel, both there is large scale decline, also there is multipath fading.

2. large scale decline modeling

Work as H _bbe greater than at 0 o'clock, between transmitting receiving node, single order Ferned Area is unobstructed, mainly considers large scale effect when channel fading modeling.Large scale decline model is comparatively fixing, and basic model is index decline model.The power of the wireless signal receiving for reception antenna is

$P_r=\frac{P_s{G}_s{G}_r{\mathrm{\&lambda;}}^2}{{\left(4\mathrm{\&pi;}\right)}^2{d}^n}$

In formula, P _sfor the transmitting power of transmitting node; G _s, G _rbe respectively transmitting antenna and receiving antenna gain; λ is wavelength; D is the distance between transmitting antenna and reception antenna; N is and the fading factor of environmental correclation, n=2 in free space time, n>2 under all the other conditions.

Under large scale fade condition, with the channel fading PL in logarithmic form definition signal transmitting procedure _bhave:

${\mathrm{PL}}_b=10\&CenterDot;\lg \left(\frac{P_s}{P_r}\right)=10\&CenterDot;\mathrm{nlgd}+20\mathrm{lgf}-10\lg \left[\frac{G_s{G}_r{c}^2}{{\left(4\mathrm{\&pi;}\right)}^2}\right]$

At H _bwhen > 0, the small scale effect of channel is ignored, and only considers large scale effect, so there is PL _b=PL, PL is the decline of the wireless signal that sends of transmitting antenna;

For extensive farm environment self-organizing application network, f, Gs, Gr are determined value, add that c and π are constant, and variable only has apart from d, and with the fading factor n of environmental correclation.For the key of large scale decline modeling, be that environment fading factor n is carried out to formulism to be described.

Propagate measured data according to the signal in corn planting environment, the signal strength signal intensity under acquisition space and time difference condition and the variation relation of envirment factor, as shown in Figure 5.To H _bbe greater than 0 partial data and carry out matching, the approximating methods such as available least square method carry out Multiple Factor Fitting to environment fading factor, draw fading factor n and block height H _band crop surface is amassed the functional relation between dnesity index PSAD.

PL=(a·ln(H _b)+b·ln(PSAD)+c)lgd+A (H _b>0)

Wherein a, b, c is fitting coefficient, and along with plant growth constantly changes, the value difference of PSAD in different growth phases, thereby can draw different fitting coefficients, generally to corn growth process, can divide emerge, jointing, heading three phases carry out modeling analysis, wherein

$A=20\mathrm{lgf}-10\lg \left[\frac{G_s{G}_r{c}^2}{{\left(4\mathrm{\&pi;}\right)}^2}\right]$

It under established condition, is a constant.Distinguishingly, the ln that in formula, matching is used, the functions such as lg are not unique solution, according to embodiment of the present invention measured data, use the fitting degree of this function higher, and have certain representativeness.

3. multiple dimensioned associating modeling

Work as H _bbe less than or equal at 0 o'clock, the small scale effect that signal is propagated is remarkable gradually, and most important two key elements that affect multipath fading are exactly multipath effect and Doppler effect.Under agricultural planting condition, not temporal evolution of the position that monitoring node is, is static network, so without considering Doppler effect.Along with environment is propagated increasing the weight of of coverage extent to signal, the line-of-sight propagation path of signal is blocked, and can only pass through crop surface, the reflections such as ground, scattering, or the mode such as the diffraction of canopy is propagated, thereby the multiple different transmission path forming, it causes each path arriving signal to have different amplitudes, phase place and time delay, therefore can produce time dispersive effect and the frequency selective fading of signal, below are all factors that analyses of Multipath Effects modeling need to be considered, but due to corn planting circumstance complication, signal may form countless differentiated transmission paths after by intensive plant, existing method cannot be carried out effective modeling analysis, the invention provides multipath fading modeling method under a kind of power spectrumanalysis corn environment.

The complex envelope that sends bandpass signal is:

$\tilde{s}\left(t\right)=\mathrm{Re}\left[s\left(t\right)e^{j2\mathrm{\&pi;}{f}_{c}t}\right]$

Wherein, f _cfor signal carrier frequency, R _erepresent the real part of complex signal, if co-exist in N bar multipath transmisstion path, make the _ithe path of paths is d _i, reflection fading coefficients is l _i, the light velocity is c, and the equal position of all nodes is fixed, and is static network, does not have Doppler effect, and receiving signal is each paths signal sum,

$\tilde{r}\left(t\right)=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{l}_i\tilde{s}(t-\frac{d_i}{c})$

By transmitted signal substitution, obtain

$\tilde{r}\left(t\right)=\mathrm{Re}\left[\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{l}_i{e}^{j2\mathrm{\&pi;}{f}_c(t-{\mathrm{\&tau;}}_i)}s(t-{\mathrm{\&tau;}}_i)\right]$

Wherein

it is the time delay on i paths.Make in formula

be normalized calculating, will receive signal indication is orthogonal form,

$\tilde{r}\left(t\right)=u_1\left(t\right)\cos \left(t\right)\cos 2\mathrm{\&pi;}{f}_{c}t+u_2\left(t\right)j\sin 2\mathrm{\&pi;}{f}_{c}t$

Wherein

$u_1\left(t\right)=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{l}_i\cos 2\mathrm{\&pi;}{f}_c(t-{\mathrm{\&tau;}}_i)$

$u_2\left(t\right)=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{l}_i\sin 2\mathrm{\&pi;}{f}_c(t-{\mathrm{\&tau;}}_i)$

In the time that N is very large, can be by u ₁and u (t) ₂(t) be considered as relatively independent Gaussian random process, and because multipath delay is random, can think phase angle 2 π f _c(t-τ _i) [π, π) above obey and be uniformly distributed, according to auto-correlation function, can receive signal envelope gross power

$P=\left(E\right[{u_1}^2\left(t\right)]+E[{u_2}^2\left(t\right)\left]\right)\&CenterDot;P_s=P_s\&CenterDot;\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{l_i}^2$

But because of multipath effect, cause phase difference between Different Diameter, and cause signal amplitude to offset, the multipath noise power P of the generation of multipath effect part _mfor

$P_m=\frac{P-P_b}{2}$

Wherein, P _bthe large scale component power of the wireless signal receiving for reception antenna;

Considering that the actual signal to noise ratio of system under multipath effect is

$\mathrm{SNR}=\frac{P_b}{P_m+\mathrm{AWGN}}$

Wherein, AWGN is additive white Gaussian noise;

In communication system, the pass of Packet Error Ratio and signal to noise ratio is:

$\mathrm{PER}=\frac{(M-1)}{2}\&times;e^{-\frac{\mathrm{SNR}}{2}}$

Wherein, M is the heterogeneous coefficient of modulation, and distinguishingly, under QPSK condition, M gets 4.

The PER drawing according to actual measurement, as shown in Figure 6, and adopt that large-scale model draws obtain described large scale component P _bpredicted value P _b', to PER and P _mcarry out curve fitting, progressively counter pushing away, finally draws the multiple dimensioned channel fading model of reality of considering under small scale effect.

$\mathrm{PL}=(a^{\&prime;}\&CenterDot;\ln \left(H_b\right)+b^{\&prime;}\&CenterDot;\ln \left(\mathrm{PSAD}\right)+c^{\&prime;})1\mathrm{gd}+1g(i\&CenterDot;e^{H_b}+j\&CenterDot;e^{\mathrm{PSAD}}+k)+B$

Wherein a ', b ', c ', i, j, k is fitting coefficient, the value difference of PSAD in different growth phases, thereby can draw different fitting coefficients, generally to corn growth process, can divide emerge, jointing, heading three phases carry out modeling analysis

$B=20\mathrm{lgf}-10\lg \left[\frac{G_s{G}_r{c}^2}{{\left(4\mathrm{\&pi;}\right)}^2}\right]+\lg \frac{2}{3}$

It under established condition, is a constant.Distinguishingly, the ln that in formula, matching is used, lg, e ^xbe not unique solution Deng function, according to embodiment of the present invention measured data, use the fitting degree of this function higher, and there is certain representativeness.

4. coefficient of determination fitting effect is evaluated

Coefficient R ²∈ [0,1] is mainly used to weigh the relation between model path loss measured value and predicted value, R ²more approach 1, show that path loss model estimated value and the higher degree of fitting of measured value correlation are high, path loss fitting effect is good, the reaction actual communication situation that now model more can be definite, R ²computing formula is as follows.

$R^2=1-\frac{\underset{i=1}{\overset{S}{\mathrm{\&Sigma;}}}{[{\mathrm{\&xi;}}_i-\widehat{\mathrm{\&xi;}]}}^2}{\underset{i=1}{\overset{S}{\mathrm{\&Sigma;}}}{[{\mathrm{\&xi;}}_i-\stackrel{\&OverBar;}{\mathrm{\&xi;}}]}^2}$

Wherein, ξ _ithe radio signal propagation characteristic gathering,

described ξ _iregressand value,

for described ξ _imean value, the number that S is collecting sample point.The method proposing according to the present invention, take 2.4G wireless signal as objective for implementation, carries out the modeling of channel fading characteristic, the R of matched curve ²be up to 0.997, minimum is 0.908, illustrates that independent variable is high to the explanation degree of dependent variable, the radio sensor network channel feature of having rebuild preferably maize field environment.

Although described by reference to the accompanying drawings embodiments of the present invention, but those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, such modification and modification all fall into by within claims limited range.

Claims (9)

1. the modeling method of the multiple dimensioned decline model of maize field radio sensor network channel, is characterized in that, the method comprises:

S1. according to the envirment factor that affects radio signal propagation in sample area, extract the key factor that causes channel space and time difference;

S2. under the varying environment factor and key factor condition, gather the radio signal propagation characteristic in sample area;

S3. according to the data and the described key factor that gather, carry out the multiple dimensioned decline model modeling of channel.

2. method according to claim 1, is characterized in that, in step S1, described envirment factor comprises the height H of crop in sample area _p, crop leaf area A _l, crop stem stalk area A _c, crop fruit surface area A _f, land area A in sample area _g, distance d between transmitting antenna and reception antenna; Described key factor comprises the height H of blocking of crop _b, crop surface area density indices P SAD, described H _b=H _a-H _p, wherein H _afor antenna height, the computing formula of described PSAD is as follows:

$\mathrm{PSAD}=\frac{A_l+A_c+A_f}{A_G\&times;H_p}.$

3. method according to claim 1, is characterized in that, in step S2, described radio signal propagation feature comprises: signal strength signal intensity, Packet Error Ratio.

4. method according to claim 2, is characterized in that, in step S2, described sample area comprises H _bthe sample area of > 0 and H _bthe sample area of < 0, described radio signal propagation characteristic comprises H _bradio signal propagation characteristic and the H of the sample area of > 0 _bthe radio signal propagation characteristic of the sample area of < 0.

5. according to the method described in claim 2 or 4, be further characterized in that, between step S2 and step S3, the method further comprises:

S21. at described H _bthe sample area of > 0, carries out channel large scale decline PL _bcalculating;

S22. obtain described H _bthe radio signal propagation characteristic of the sample area of > 0, carries out matching to the data that gather, and obtains channel large scale decline model.

6. method according to claim 5, is characterized in that, in step S21, and described channel large scale decline PL _bfor:

${\mathrm{PL}}_b=10.1g\left(\frac{P_s}{P_r}\right)$

Wherein, P _sfor the transmitting power of transmitting antenna, P _rthe power module of the wireless signal receiving for reception antenna;

Described $P_r=\frac{P_s{G}_s{G}_r{\mathrm{\&lambda;}}^2}{{\left(4\mathrm{\&pi;}\right)}^2{d}^n};$

Wherein, G _s, G _rbe respectively transmitting antenna and receiving antenna gain, λ is wavelength, and d is the distance between transmitting antenna and reception antenna, and n is fading factor.

7. method according to claim 6, is characterized in that, described step S22 comprises: obtain described H _bthe radio signal propagation characteristic of the sample area of > 0, by described radio signal propagation data to described P _rcarry out matching, obtain fading factor n and H _b, PSAD functional relation:

n=a·ln(H _b)+b·ln(PSAD)+c

Described channel large scale decline model is:

PL=(a·ln(H _b)+b·ln(PSAD)+c)lgd+A

Wherein, H _b> 0, PL is the decline of the wireless signal that sends of transmitting antenna, and a, b and c are fitting coefficient, and described fitting coefficient is determined according to PSAD corresponding to plant growth stage, H _bfor blocking height;

Wherein,

f is wireless signal frequency, G _s, G _rbe respectively transmitting antenna and receiving antenna gain, c is light velocity constant.

8. according to the method described in claim 2,3,6 or 7, it is characterized in that, described step S3 comprises:

The envelope gross power P of the wireless signal S31. receiving according to reception antenna, described envelope gross power P comprises large scale component power P _band small scale multipath noise power P _m, build described P _mmodel:

$P_m=\frac{P-P_b}{2}$

Wherein, the formula of described P is as follows:

$P=P_s\&CenterDot;\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{l_i}^2$

Wherein, P _sfor the transmitting power of transmitting antenna, N is the number in multipath path; l _iit is the reflection fading coefficients of i paths;

S32. according to described P _m, build signal to noise ratio snr model;

$\mathrm{SNR}=\frac{P_b}{P_m+\mathrm{AWGN}};$

Wherein, AWGN is additive white Gaussian noise;

S33. according to described SNR, build Packet Error Ratio PER model;

$\mathrm{PER}=\frac{(M-1)}{2}\&times;e^{-\frac{\mathrm{SNR}}{2}};$

Wherein, M is the heterogeneous coefficient of modulation;

S34. obtain described H _bthe radio signal propagation characteristic of the sample area of < 0, obtains the Packet Error Ratio PER measured value of reception antenna;

According to H _benvironment key factor H in the sample area of < 0 _b, PSAD, and channel large scale decline model, obtains described large scale component power P _bpredicted value P _b;

S35. according to described PER measured value and described large scale component power P _bpredicted value P _b', to described PER model and P _mmodel carries out curve fitting, and obtains the multiple dimensioned decline model of channel:

$\mathrm{PL}=(a^{\&prime;}\&CenterDot;\ln \left(H_b\right)+b^{\&prime;}\&CenterDot;\ln \left(\mathrm{PSAD}\right)+c^{\&prime;})1\mathrm{gd}+1g(i\&CenterDot;e^{H_b}+j\&CenterDot;e^{\mathrm{PSAD}}+k)+B$

Wherein, H _b<0, PL is the decline of the wireless signal that sends of transmitting antenna, and a ', b ', c ', i, j, k are fitting coefficient, and described fitting coefficient is determined according to PSAD corresponding to plant growth stage;

Wherein, $B=201\mathrm{gf}-101g\left[\frac{G_s{G}_r{c}^2}{{\left(4\mathrm{\&pi;}\right)}^2}\right]+1g\frac{2}{3},$ F is wireless signal frequency, G _sfor transmitter antenna gain (dBi), c is light velocity constant.

9. according to the method in claim 2 or 3, be further characterized in that, described method further comprises: S4. is according to coefficient R ²evaluate the model of described acquisition, wherein said R ²computing formula as follows:

$R^2=1-\frac{\underset{i=1}{\overset{S}{\mathrm{\&Sigma;}}}{[{\mathrm{\&xi;}}_i-\widehat{\mathrm{\&xi;}]}}^2}{\underset{i=1}{\overset{S}{\mathrm{\&Sigma;}}}{[{\mathrm{\&xi;}}_i-\stackrel{\&OverBar;}{\mathrm{\&xi;}}]}^2}$

Wherein, ξ _ithe radio signal propagation characteristic gathering,

described ξ _iregressand value, for described ξ _imean value, the number that S is collecting sample point.

Images