CN101854216B - Channel parity researching method based on ionosphere correction layer channel model - Google Patents

Abstract

The invention relates to a channel parity researching method based on an ionosphere correction layer channel model, which comprises the following steps that: step 1: an international reference ionosphere model (IRI 2007) is built; step 2: an ionosphere parameter correcting module is established by integrating the actually measured data so that the ionosphere parameters have time variation; and step 3: an ITS channel model is built, and the impulse response function of an ionosphere channel in a multi-path structure is simulated and calculated by integrating the ionosphere parameters and set signal characteristic parameters. The method not only provides convenience for inputting the parameters and reduces the parameter error in simulation, but also improves the time variation characteristic of the channel model so as to solve the problem that the ITS model is not applicable to the research on channel parity and find out solutions for the research on the channel parity. Consequently, the channel parity researching method based on the ionosphere correction layer channel model has wide practical value and application prospect in the information technology field.

Description

A kind of based on revising the method that the ionospheric channel model carries out the research of channel equity

(1) technical field

The present invention relates to a kind of method of carrying out the research of channel equity based on correction ionospheric channel model, belong to areas of information technology.

(2) background technology

2.1 the concept of channel equity

Though people have experienced nearly half a century to the research of channel, the research of the channel equity that the present invention is proposed but seldom related to.

Ionospheric equity refers to, during short wave communication that the communicating pair that is in the strange land sends out receipts at one time mutually mutual, the information that both sides receive should be the same in theory, so we just claim both sides' signal the ionospheric channel of process be reciprocity.

From above-mentioned concept as can be known, if intercommunication carries out can be absolute the time, the ionospheric channel that passed through of both sides' signal is real-time equity certainly so.This conclusion is readily appreciated that by the reciprocity principle, and almost need prove.But in practical project, sending out mutually mutually of being difficult to accomplish to carry out simultaneously received, even can reach certain precision, that also must pay suitable financial resources cost.Therefore, the channel equity that the present invention discusses refers to: communicating pair in certain time interval (such as: one minute) by turns transmitting-receiving, whether their received information equity; In order to make both sides obtain the same information, what kind of scope the time interval should be limited in.This be research method of the present invention center on key problem.

2.2 the technology that the research of channel equity relates to

Research channel equity must at first be set up the channel model that is fit to present short wave communication requirement.The technology of setting up channel model has developed decades, yet the channel model of not only suitable arrowband but also suitable broadband connections is considerably less.Wherein, the wide band model that is proposed by U.S. Institute for Telecommunication Science (ITS) is the ITS model, and being United States Naval Research Laboratory (NRL) and ITS surveys and the abstract mathematics model that draws through the channel of long term studies and ionosphere quiet period.It is expanded three main aspects from power delay profile, Doppler frequency shift and Doppler and has finished statistical modeling to channel, and from actual measurement, this model has well embodied the original feature of channel, and it is more accurate to compare with other models.Therefore, the ITS model is to develop the most perfect, the most authoritative broad-band channel model at present.

But directly the computer simulation model of setting up according to the ITS model can't be applicable to the research requirement of channel equity fully.The problem of its existence is as follows:

1.ITS when model emulation has the channel of multipath transmission structure, need a large amount of parameter of input.Wherein, the input parameter demand of every reflection path is as follows:

D: the great-circle distance between the transmitter and receiver

f _p: the pentration frequency of ionospheric reflection layer

h ₀: the height at ionospheric reflection layer maximum electron density place

σ: the half thickness of ionospheric reflection layer

f _c: the centre frequency of emission system

A: the power peak that receives signal

σ _τ: the time delay expansion

σ _c: part time delay expansion (time delay at peak value place and minimal time delay poor)

σ _D: receive signal at time-delay τ=τ _cThe time 0.5 times of Doppler expansion.

f _s: τ=τ _cThe place is to deserved Doppler frequency displacement

f _SL: τ=τ _LThe Doppler frequency displacement that the place is corresponding

As seen if there is N bar multipath in the channel of emulation, need to import 11 * N parameter (parameter that also has some to calculate the needs setting is not included) when using the ITS model emulation so at least.

Though, time delay expansion, Doppler's expansion and ionosphere parameter (f _p, h ₀, σ) can obtain or be obtained by the experience of practice summary by actual measurement.But, in the process that studies a question, can not whenever do an emulation and namely once survey, so not only do not meet actual conditions, and can lose the meaning of simulation calculation.

2. in the ITS model, multidiameter (τ _c=σ _τ-σ _c) be to be obtained indirectly by input parameter.It is a constant rather than variable in channel calculation procedure, and is irrelevant with the variation of ionospheric channel.Therefore, multidiameter can't embody the time variation of channel.

In sum, for can being applicable to, the channel simulator model subject study among the present invention also need improve at above-mentioned two large problems in the ITS model based.

(3) summary of the invention

1, purpose: the purpose of this invention is to provide a kind of based on revising the method that the ionospheric channel model carries out the research of channel equity, this method has overcome the deficiencies in the prior art, it has proposed the research of the equity of power transformation absciss layer channel at random the time first, and based on channel model, by the equity of simulation calculation prediction ionospheric channel.

When electromagnetic signal when the ionospheric propagation, can have ionospheric natural information in the signal, this at random, present from parameter form with the electromagnetic signature parameter in receiving signal of natural stochastic source.When ionospheric channel had equity, the both sides of communication can obtain equity and electromagnetic parameter at random mutually.These parameters with randomness and equity can be brought into play enormous function in the confidential corespondence field.In case successfully obtain, will become the major progress in this field.

2, technical scheme: to achieve these goals, the technical solution used in the present invention is as follows:

The present invention is a kind of based on revising the method that the ionospheric channel model carries out the research of channel equity, and the step that this method is concrete is as follows:

Step 1: set up the international reference ionosphere model, namely set up " IRI 2007 " model;

The international reference ionosphere model is disclosed model, and the source code of its realization also is open, so its implementation procedure is omitted herein.

We only need input communication both sides' longitude and latitude and call duration time in the process of simulation analysis, can obtain the critical frequency f of reflection path upper ionized layer by IRI2007 _p, the electron concentration maximum height h ₀And the half thickness σ in reflector.

Step 2: in conjunction with measured data, set up the correcting module of ionosphere parameter; Make the ionosphere parameter have time variation.

Find according to for many years ionospheric probing, the different anniversaries have similar ionosphere Changing Pattern an identical month.Therefore, we have introduced the modifying factor of ionosphere parameter at correcting module, and this factor is reflector (reflect_layer), month (M), the function of (T) and time (t) constantly.That is to say the critical frequency f in different reflector _pHeight h with the electron concentration maximum ₀To change in time in different months and the moment.The detailed process structure that parameter correction unit is divided is seen Fig. 2.

According to the reflector difference, in program, call different correcting modules, in module, call corresponding modifying factor according to month (M) and the moment (T) again, its formula is expressed as follows:

f _p+＝change_f(M，T)·Δt

h ₀+＝change_h(M，T)·Δt (1)

σ＝σ

Step 3: set up the ITS channel model; In conjunction with the signal characteristic parameter of ionosphere parameter and setting, simulation calculation goes out to have the impulse Response Function of the ionospheric channel of multipath structure.

As shown in Figure 1, the ITS model is divided into two parts: deterministic models part and stochastic model part.

1, deterministic models part:

Utilize the corrected ionosphere of step 2 parameter: f _p, h ₀, σ calculates signal by the time delay τ behind the ionospheric propagation _c, its calculation procedure is as follows:

(1) utilizes corrected ionosphere parameter f _p, h ₀, σ is in conjunction with incoming carrier frequency parameter f _c, calculate electromagnetic signal at ionospheric usable reflection height

Computing formula is as follows:

$f_c=f_p{\left\{\right[1+{(D/2\stackrel{\‾}{h})}^2]/[1+\exp ((h_0-\stackrel{\‾}{h})/\mathrm{\σ})\left]\right\}}^{1/2}---\left(2\right)$

(2) the usable reflection height that calculates according to previous step

Spherical distance D and light velocity c in conjunction with the transmitting-receiving two places calculate electromagnetic signal arrives receiving terminal through ionospheric reflection time delay τ _c, it is as follows to calculate the principle formula:

$\stackrel{\‾}{h}={\{{({\mathrm{c\τ}}_c/2)}^2-{(D/2)}^2\}}^{1/2}---\left(3\right)$

2, stochastic model part:

The ITS channel modeling method is namely set up the impulse Response Function of channel, and the impulse response of whole channel is defined as the impulse response sum of every propagation path, and formula (4) is the mathematic(al) representation of model.

$h(t,\mathrm{\τ})=\underset{n}{\mathrm{\Σ}}\sqrt{p_n\left(\mathrm{\τ}\right)}{D}_n(t,\mathrm{\τ}){\mathrm{\ψ}}_n(t,\mathrm{\τ})---\left(4\right)$

N in the formula (4) represents different transmission modes.The overall structure of model is seen Fig. 3, and the structure of single transport pattern is seen Fig. 4.

As seen from Figure 4, the impulse response function of each transmission mode is made up of three parts: all square function P of power time-delay section _n(τ), certainty phase function D _n(τ), Stochastic Modulation function ψ _n(τ).Wherein, the time delay expansion in power time-delay section function table reference road; The certainty phase function characterizes the Doppler frequency shift of channel; The Stochastic Modulation function then characterizes Doppler's expansion of channel.The realization of the randomness of channel part will be divided into following three steps and finish so:

(1) realization of all square functions of power time delay section

Power delay profile (DPP) has determined the shape that retarding power distributes, by τ _c, σ _c, σ _τDetermine with four parameters such as peak power A.(see figure 5)

With P (τ) expression time-delay Power Spectrum Distribution function, broadband short wave channel mode time-delay Power Spectrum Distribution is that Gamma distributes, and its mathematical form is as follows:

$p\left(\mathrm{\τ}\right)=A\frac{{\mathrm{\α}}^{\mathrm{\α}+1}}{\mathrm{\Δ\Γ}(\mathrm{\α}+1)}{z}^{\mathrm{\α}}{e}^{-\mathrm{\αz}}---\left(5\right)$

$z=\frac{(\mathrm{\τ}-{\mathrm{\τ}}_c)}{\mathrm{\Δ}}+1---\left(6\right)$

Δ=τ wherein _c-τ _l, be used for controlling width, then z can be expressed as:

z＝(τ-τ _l)/(τ _c-τ _l)＞0 (7)

Parameter alpha and τ _lWidth and symmetry that control distributes, they depend on time delay expansion and threshold level, in order to obtain their occurrence, make two parameters satisfy formula (8).

lns _v＝α(lnz _L+1-z _L)＝α(lnz _U+1-z _U) (8)

z _L＝(τ _L-τ _l)/(τ _c-τ _l)＞0 (9)

z _U＝(τ _U-τ _l)/(τ _c-τ _l)＞0 (10)

At first

lnz _L-z _L＝lnz _U-z _U

(1-z _L)/(z _U-1)＝(τ _c-τ _L)/(τ _U-τ _c)＜1 (11)

Can obtain z according to following formula associating equation group with Newton iteration method _L, and then obtain parameter alpha and τ _lOccurrence.

α＝(lnz _L+1-z _L) ^-1lnS _v

S _v＝A _fl/A

τ _l＝τ _c-(τ _c-τ _L)/(1-z _L)＜τ _L

Under the condition of knowing A, α, z, can obtain time-delay Power Spectrum Distribution function so.

(2) realization of certainty phase function

Certainty phase function (DPF) is used for characterizing the size that doppler frequency moves, usefulness function D (it is defined as follows for t, τ) expression:

D(t，τ)＝exp2π[f _s+m(τ-τ _c)]t

m＝(f _s-f _sL)/(τ _c-τ _L) (12)

Have according to Euler's formula:

D(t，τ)＝COS2π[f _s+m(τ-τ _c)]t+isin2π[f _s+m(τ-τ _c)]t (13)

Wherein: m is that Doppler frequency-shift is with respect to the rate of change of time-delay τ.f _sBe that time-delay is at τ=τ _cThe time the Doppler frequency shift value; f _SLBe time-delay τ=τ _LThe time the Doppler frequency shift value.

(3) realization of Stochastic Modulation function

For the decline of analog channel impulse response, (t τ) is made of a large amount of time serieses of answering at random Stochastic Modulation function ψ.On each postpones τ, makes up two independently in vain gaussian random sequence represent multiple seasonal effect in time series real part and imaginary part respectively.Therefore each multiple seasonal effect in time series whose amplitude obeys Rayleigh distributes.For the power spectrum width that limits random sequence to reach the Doppler spread spectrum that emulation needs, the width of filter is the Doppler extension width.The implementation procedure of Stochastic Modulation function as shown in Figure 6.

In implementation procedure, adopt discrete functional form with the Stochastic Modulation function representation be ψ (n, τ).

ψ(n，τ)＝x(n，τ)+iy(n，τ) (14)

For each fixing delay τ, can be expressed as:

ψ _n＝x _n+iy _n(n＝0，1，2.....) (15)

x _nAnd y _nBe independent random variables, constituted by different random sequences respectively, the following (ρ of its production process _n, ρ ' _nBe respectively two kinds of different randomizers and produce the gaussian random sequence of independent zero-mean):

x _n＝ρ _n+(x _n-1-ρ _n)λ (16)

y _n＝ρ′ _n+(y _n-1-ρ′ _n)λ (17)

X wherein ₀=(1-λ) ρ ₀, y ₀=(1-λ) ρ ' ₀(18)

λ＝exp[-(Δt)σ _f] (19)

Doppler's expansion is divided into two kinds of situations: Gaussian shape and Lorentzian shape.So σ in the formula (19) _fGet different values:

Gaussian σ _f＝σ _D{2π/(-lnS _v)} ^1/2

Lorentizian σ _f＝2πσ _D{S _v/(1-S _v)} ^1/2

σ _DBe the half-power bandwidth of Doppler expansion, S _v=A _Fl/ A, A _FLBe the signal threshold level.

The present invention can do following analysis to the channel equity: set different simulated conditions, the simulation calculation channel impulse response after implementing to finish according to above-mentioned three big steps; Analyze the equity of the channel under the different condition by the relative multidiameter delay parameter between the multipath.(different simulated conditions comprise: ionosphere status condition and different commitment defini interval conditions.)

3, advantage and effect: the present invention is a kind of based on revising the method that the ionospheric channel model carries out the research of channel equity, and its advantage is as follows:

(1) at the many problem of the input parameter of ITS model

In research approach, we have introduced the ionospheric forecast model on the upper strata of ITS model, and have chosen present the most authoritative, up-to-date international reference ionosphere 2007 versions (IRI 2007) in the world.

As Fig. 1, only need the input transmitter and receiver longitude and latitude, date and local time (obviously being known conditions), then can obtain the ionosphere parameter f by IRI2007 _p, h ₀, σ, binding signal parameter (given in conjunction with practical experience) is as the input of ITS model then, the impulse Response Function of the ionospheric channel that becomes in the time of can obtaining.

Simultaneously and since in a lot of researchs all empirical tests the validity of IRI2007, it is pervasive in the whole world with certain precision.

Therefore in sum, the remarkable advantage of this research method is as follows:

1) significantly reduced the input parameter of model;

2) reduced difficulty getparms;

3) improved the accuracy of simulation result.

(2) be the problem of constant at multidiameter delay in the ITS model

In the step 2 of research approach, we have introduced the correcting module of ionosphere parameter.This module has comprised the modifying factor of each ionosphere parameter, and this factor is reflector (reflect_layer), month (month), the function of (T) and time (t) constantly.That is to say the critical frequency f in different reflector _pHeight h with the electron concentration maximum ₀To change in time in different months and the moment.(Fig. 2 has provided the detailed process structure that parameter correction unit is divided.)

Obviously, behind the correcting module of introducing ionosphere parameter, the ionosphere parameter all can change at each time point, accordingly the multidiameter delay τ that draws according to the ionosphere calculation of parameter _cIt also will be the function of time.Therefore, the second class advantage of this programme is:

1) τ in channel calculation procedure _cIt is time dependent variable;

2) time-varying characteristics of channel have been strengthened.

(3) brief summary

2 points of comprehensive front, the improvement that we implement on the ITS model based, not only provide from the emulation aspect input parameter convenience, reduced the error of parameter, and increased the time-varying characteristics of channel model.Thereby, solved the ITS model and can not be applicable to the problem of studying the channel equity, for solution has been found in the research of channel equity.

(4). description of drawings

The overall structure schematic flow sheet of Fig. 1 the method for the invention

The partial detailed schematic flow sheet of Fig. 2 channel model structure (parameter correction unit branch)

The overall structure schematic diagram of Fig. 3 ITS channel model

The formation schematic diagram of the transfer function of Fig. 4 channel monotype

Fig. 5 power time delay generalized section

The realization flow schematic diagram of Fig. 6 Stochastic Modulation function

The impulse response emulation schematic diagram of ionospheric channel under the short commitment defini interval situation of Fig. 7

The impulse response actual measurement schematic diagram of ionospheric channel under the short commitment defini interval situation of Fig. 8

Label and symbol description among the figure are as follows:

Fig. 2: f wherein _p, h ₀, σ represents the critical frequency in reflector, the height of electron concentration maximum, the half thickness in reflector respectively; Ref_layer represents the reflector: 1 is the E layer, and 2 is the F1 layer, and 3 is the F2 layer; T represents the simulation time point; The total time of t ' for setting; N represents negates; Y represents certainly; M represents month, the communication moment that the T representative is current.

Fig. 5: A is illustrated in τ=τ _cThe time peak power; σ _cThe expansion of expression part time delay; σ _τThe expansion of expression time delay; τ _UThe maximum of expression time-delay; τ _LThe minimum value of expression time-delay; τ wherein _L=τ _c-σ _c, τ _U=σ _τ-τ _c

Fig. 6: ρ _n, ρ ' _nThe gaussian random sequence of independent, the zero-mean that is respectively that two kinds of different randomizers produce; x _nAnd y _nBe ρ _n, ρ ' _nThe independent random variables of Gou Chenging respectively; ψ (t, τ) expression Stochastic Modulation function.

(a) among Fig. 7, (b), (c), (d), (e), the impulse response of channel when (f) 6 width of cloth figure represent the 1st, the 2nd, the 5th, the 10th, the 20th and the 30th cycle of emulation respectively, the time interval between each figure is respectively 10ms, 30ms, 50ms, 100ms, 100ms.The longitudinal axis represents relative amplitude; Transverse axis represents time delay, and unit is: us.

(a) among Fig. 8, (b), (c), (d), the actual measurement impulse response of channel when (e) 5 width of cloth figure represented for the 1st, 5,10,20,30 cycles respectively, each impulse response represents a multipath.Have four multipaths as we can see from the figure.Attention: measured drawing obtains by circular correlation, and the longitudinal axis is the correlation magnitude value, and abscissa represents sampling number, and the data in relative time delay in the table 3 obtain through conversion: the corresponding 1ms of 640 sampled points.)

(5) embodiment

The present invention is a kind of based on revising the method that the ionospheric channel model carries out the research of channel equity, and the present invention is described in further detail that its step is as follows below in conjunction with drawings and Examples:

Referring to shown in Figure 1, study the equity of ionospheric channel, at first to set up can the impulse response of emulation ionospheric channel the channel simulator system, whole system comprises three parts: be used for forecast ionosphere parameter the international reference ionosphere model, revise correcting module and the ITS channel model of ionosphere parameter.

Step 1: set up international reference ionosphere model (IRI2007)

The international reference ionosphere model is disclosed model, and the source code of its realization also is open, so its implementation procedure is omitted herein.

We only need input communication both sides' longitude and latitude and call duration time in the process of simulation analysis, can obtain the critical frequency f of reflection path upper ionized layer by IRI2007 _p, the electron concentration maximum height h ₀And the half thickness σ in reflector.

Step 2: in conjunction with measured data, set up the correcting module of ionosphere parameter; Make the ionosphere parameter have time variation.

Find according to for many years ionospheric probing, the different anniversaries have similar ionosphere Changing Pattern an identical month.Therefore, we have introduced the modifying factor of ionosphere parameter at correcting module, and this factor is reflector (reflect_layer), month (M), the function of (T) and time (t) constantly.That is to say the critical frequency f in different reflector _pHeight h with the electron concentration maximum ₀To change in time in different months and the moment.The detailed process structure that parameter correction unit is divided is seen Fig. 2.

According to the reflector difference, in program, call different correcting modules, in module, call corresponding modifying factor according to month (M) and the moment (T) again, its principle is as follows:

f _p+＝change_f(M，T)·Δt

h ₀+＝change_h(M，T)·Δt (1)

σ＝σ

Step 3: set up the ITS channel model; In conjunction with the signal characteristic parameter of ionosphere parameter and setting, simulation calculation goes out to have the impulse Response Function of the ionospheric channel of multipath structure.

As shown in Figure 1, the ITS model is divided into two parts: deterministic models part and stochastic model part.

1, deterministic models part:

Utilize the corrected ionosphere of step 2 parameter: f _p, h ₀, σ calculates signal by the time delay τ behind the ionospheric propagation _c, its calculation procedure is as follows:

(1) utilizes corrected ionosphere parameter f _p, h ₀, σ is in conjunction with input parameter f _c(expression carrier frequency) calculates electromagnetic signal at ionospheric usable reflection height

It is as follows to calculate the principle formula:

$f_c=f_p{\left\{\right[1+{(D/2\stackrel{\‾}{h})}^2]/[1+\exp ((h_0-\stackrel{\‾}{h})/\mathrm{\σ})\left]\right\}}^{1/2}---\left(2\right)$

(2) the usable reflection height that calculates according to previous step

Spherical distance D and light velocity c in conjunction with the transmitting-receiving two places calculate electromagnetic signal arrives receiving terminal through ionospheric reflection time delay τ _c, it is as follows to calculate the principle formula:

$\stackrel{\‾}{h}={\{{({\mathrm{c\τ}}_c/2)}^2-{(D/2)}^2\}}^{1/2}---\left(3\right)$

2, stochastic model part:

The ITS channel modeling method is namely set up the impulse Response Function of channel, and the impulse response of whole channel is defined as the impulse response sum of every propagation path, and formula (4) is the mathematic(al) representation of model.

$h(t,\mathrm{\τ})=\underset{n}{\mathrm{\Σ}}\sqrt{p_n\left(\mathrm{\τ}\right)}{D}_n(t,\mathrm{\τ}){\mathrm{\ψ}}_n(t,\mathrm{\τ})---\left(4\right)$

N in the formula (4) represents different transmission modes.The overall structure of model is seen Fig. 3, and the structure of single transport pattern is seen Fig. 4.

As seen from Figure 4, the impulse response function of each transmission mode is made up of three parts: all square function P of power time-delay section _n(τ), certainty phase function D _n(τ), Stochastic Modulation function ψ _n(τ).Wherein, the time delay expansion in power time-delay section function table reference road; The certainty phase function characterizes the Doppler frequency shift of channel; The Stochastic Modulation function then characterizes Doppler's expansion of channel.The realization of the randomness of channel part will be divided into following three steps and finish so:

(1) realization of power time delay section function

Power delay profile (DPP) has determined the shape that retarding power distributes, by τ _c, σ _c, σ _τDetermine with four parameters such as peak power A.(see figure 5)

With P (τ) expression time-delay Power Spectrum Distribution function, broadband short wave channel mode time-delay Power Spectrum Distribution is that Gamma distributes, and its mathematical form is as follows:

$p\left(\mathrm{\τ}\right)=A\frac{{\mathrm{\α}}^{\mathrm{\α}+1}}{\mathrm{\Δ\Γ}(\mathrm{\α}+1)}{z}^{\mathrm{\α}}{e}^{-\mathrm{\αz}}---\left(5\right)$

$z=\frac{(\mathrm{\τ}-{\mathrm{\τ}}_c)}{\mathrm{\Δ}}+1---\left(6\right)$

Δ=τ wherein _c-τ _l, be used for controlling width, then z can be expressed as:

z＝(τ-τ _l)/(τ _c-τ _l)＞0 (7)

Parameter alpha and τ _lWidth and symmetry that control distributes, they depend on time delay expansion and threshold level, in order to obtain their occurrence, make two parameters satisfy formula (8).

lns _v＝α(lnz _L+1-z _L)＝α(lnz _U+1-z _U) (8)

z _L＝(τ _L-τ _l)/(τ _c-τ _l)＞0 (9)

z _U＝(τ _U-τ _l)/(τ _c-τ _l)＞0 (10)

At first

lnz _L-z _L＝lnz _U-z _U

(1-z _L)/(z _U-1)＝(τ _c-τ _L)/(τ _U-τ _c)＜1 (11)

Can obtain z according to following formula associating equation group with Newton iteration method _L, and then obtain parameter alpha and τ _lOccurrence.

α＝(lnz _L+1-z _L) ^-1lnS _v

S _v＝A _fl/A

τ _l＝τ _c-(τ _c-τ _L)/(1-z _L)＜τ _L

Under the condition of knowing A, α, z, can obtain time-delay Power Spectrum Distribution function so.

(2) realization of certainty phase function

Certainty phase function (DPF) is used for characterizing the size that doppler frequency moves, usefulness function D (it is defined as follows for t, τ) expression:

D(t，τ)＝exp2π[f _s+m(τ-τ _c)]t

m＝(f _s-f _sL)/(τ _c-τ _L) (12)

Have according to Euler's formula:

D(t，τ)＝COS2π[f _s+m(τ-τ _c)]t+isin 2π[f _s+m(τ-τ _c)]t (13)

Wherein m is that Doppler frequency-shift is with respect to the rate of change of time-delay τ.f _sBe that time-delay is at τ=τ _cThe time the Doppler frequency shift value; f _SLBe time-delay τ=τ _LThe time the Doppler frequency shift value.

(3) realization of Stochastic Modulation function

For the decline of analog channel impulse response, (t τ) is made of a large amount of time serieses of answering at random Stochastic Modulation function ψ.On each postpones τ, makes up two independently in vain gaussian random sequence represent multiple seasonal effect in time series real part and imaginary part respectively.Therefore each multiple seasonal effect in time series whose amplitude obeys Rayleigh distributes.For the power spectrum width that limits random sequence to reach the Doppler spread spectrum that emulation needs, the width of filter is the Doppler extension width.The implementation procedure of Stochastic Modulation function as shown in Figure 6.

In implementation procedure, adopt discrete functional form with the Stochastic Modulation function representation be ψ (n, τ).

ψ(n，τ)＝x(n，τ)+iy(n，τ) (14)

For each fixing delay τ, can be expressed as:

ψ _n＝x _n+iy _n(n＝0，1，2.....) (15)

x _nAnd y _nBe independent random variables, constituted by different random sequences respectively, the following (ρ of its production process _n, ρ ' _nBe respectively two kinds of different randomizers and produce the gaussian random sequence of independent zero-mean):

x _n＝ρ _n+(x _n-1-ρ _n)λ (16)

y _n＝ρ′ _n+(y _n-1-ρ′ _n)λ (17)

X wherein ₀=(1-λ) ρ ₀, y ₀=(1-λ) ρ ' ₀(18)

λ＝exp[-(Δt)σ _f] (19)

Doppler's expansion is divided into two kinds of situations: Gaussian shape and Lorentzian shape.So σ in the formula (19) _fGet different values:

Gaussian σ _f＝σ _D{2π/(-lnS _v)} ^1/2

Lorentizian σ _f＝2πσ _D{S _v/(1-S _v)} ^1/2

σ _DBe the half-power bandwidth of Doppler expansion, S _v=A _Fl/ A, A _FLBe the signal threshold level.

The present invention can do following analysis to the channel equity after implementing to finish according to above-mentioned three big steps:

Set different simulated conditions, the simulation calculation channel impulse response; Analyze the equity of the channel under the different condition by the relative multidiameter delay parameter between the multipath.

Embodiment:

The method that we introduce in conjunction with the present invention, the equity situation of ionospheric channel when analyzing communication is spaced apart 10ms.

The present invention is a kind of based on revising the method that the ionospheric channel model carries out the research of channel equity, and concrete implementation step is as follows:

Step 1: show as Fig. 1: input communication both sides' longitude and latitude position in ionospheric forecast model IRI2007: (X1, Y1) (X2, Y2), local time (point in the mornings 10 on October 25th, 2008) of date and communication, thereby obtain the parameter in each reflector, ionosphere: critical frequency f _p, reflector height h ₀, reflector half thickness σ, its concrete numerical value is as follows:

F1 layer: multipath 1:f _p: 12.02MHz; h ₀: 190km; σ: 20km;

F2 layer: multipath 2:f _p: 15.55MHz; h ₀: 308km; σ: 148km;

Multipath 3:f _p: 15.75MHz; h ₀: 315km; σ: 150km;

Multipath 4:f _p: 17.25MHz; h ₀: 338km; σ: 150km;

Step 2: corrected parameter

As shown in Figure 1, after the ionosphere parameter by each reflector of IRI2007 prediction acquisition, each parameter will be imported into the parameter correction unit branch among Fig. 1.As shown in Figure 2, program is understood the correcting module different according to the different choice in reflector at the parameter correction unit branch, and then according to the corresponding modifying factor of local time selection of communicating by letter, then to f _pAnd h ₀Carry out real-time correction, its mathematical principle is as follows:

f _p+＝change_f(M，T)·Δt

h ₀+＝change_h(M，T)·Δt

σ＝σ

M in the following formula and T represent month and the moment in local time respectively.

In this example, the morning, multipath 1 reflector was the F1 layer, and call duration time is points in the mornings 10 in October, and the modifying factor of routine call is: change_f=0.000028MHz; Change_h=-0.00294km; Multipath 2 is the F2 layer to the reflector of multipath 4, and communication local time be points in the mornings 10 in October, the modifying factor of routine call is: change_f=0.00044MHz; Change_h=-0.0014km.Δ t among Fig. 2 gets 10ms (being 0.01s).

Step 3: set up the ITS channel model, calculate the impulse response of channel.

Fig. 3 is the overall structure schematic diagram of ITS channel model, as shown in Figure 3, whole channel is made up of many single footpaths, therefore in the process that realizes, needs to realize earlier monotype and then with the impulse response function of the whole channel of impulse response stack formation of each monotype.Fig. 4 is the formation schematic diagram of the transfer function of channel monotype, each the multipath parameter that need utilize preceding two steps to obtain, and in conjunction with each single directly signal parameter of pattern of input, thus the impulse response of calculating each single footpath pattern.The input parameter setting of each single footpath pattern and as shown in table 1 by the indirect parameter that calculates:

The setting of table 1 parameter and intermediate parameters

Annotate: the indirect parameter in the table 1 is the numerical value that calculates at first time point according to input parameter.

1. deterministic models part

In the deterministic models part, the ionosphere parameter f in each footpath _p, h ₀Input parameter f with each Dan Jing in the σ associative list 1 _cBring in the following formula, calculate the usable reflection height of each Dan Jing

$f_c=f_p{\left\{\right[1+{(D/2\stackrel{\‾}{h})}^2]/[1+\exp ((h_0-\stackrel{\‾}{h})/\mathrm{\σ})\left]\right\}}^{1/2}$

The usable reflection height that calculates according to previous step

In conjunction with the spherical distance D of transmitting-receiving two places and the time delay τ of the electromagnetic signal process ionospheric reflection arrival receiving terminal that light velocity c calculates each Dan Jing _c, it is as follows to calculate the principle formula:

$\stackrel{\‾}{h}={\{{({\mathrm{c\τ}}_c/2)}^2-{(D/2)}^2\}}^{1/2}$

As shown in table 1, through this step, can obtain the time delay τ of each Dan Jing _cBe respectively: 2462.7us, 3137.1us, 2958.05us and 2936.2us.

2. stochastic model part

(1) realization of power time delay section function

With P (τ) expression time-delay Power Spectrum Distribution function, by τ _c, σ _c, σ _τDetermine with four parameters such as peak power A.Its mathematical form of (see figure 5) is as follows:

$p\left(\mathrm{\τ}\right)=A\frac{{\mathrm{\α}}^{\mathrm{\α}+1}}{\mathrm{\Δ\Γ}(\mathrm{\α}+1)}{z}^{\mathrm{\α}}{e}^{-\mathrm{\αz}}$

z＝(τ-τ _l)/(τ _c-τ _l)＞0

Parameter alpha and τ _lWidth and symmetry that control distributes, they depend on time delay expansion and threshold level, in order to obtain their occurrence, make two parameters satisfy following formula.

lns _v＝α(lnz _L+1-z _L)＝α(lnz _U+1-z _U)

z _L＝(τ _L-τ _l)/(τ _c-τ _l)＞0

z _U＝(τ _U-τ _l)/(τ _c-τ _l)＞0

Can obtain z according to following formula associating equation group with Newton iteration method _L, and then obtain parameter alpha and τ _lOccurrence.

α＝(lnz _L+1-z _L) ^-1lnS _v

S _v＝A _fl/A

τ _l＝τ _c-(τ _c-τ _L)/(1-z _L)＜τ _L

This example is according to the input parameter A of each Dan Jing in the table 1, S _v, σ _c, σ _τ, and the time delay τ that calculates _cCan calculate α and the τ of each Dan Jing _l, the numerical value of each Dan Jing is listed in the table 1.

Knowing A, α, τ so _lCondition under can obtain time-delay Power Spectrum Distribution function.

(2) realization of certainty phase function

Certainty phase function (DPF) is used for characterizing the size that doppler frequency moves, usefulness function D (it is defined as follows for t, τ) expression:

D(t，τ)＝exp 2π[f _s+m(τ-τ _c)]t

m＝(f _s-f _sL)/(τ _c-τ _L)

Have according to Euler's formula:

D(t，τ)＝COS2π[f _s+m(τ-τ _c)]t+isin 2π[f _s+m(τ-τ _c)]t

Wherein m is that Doppler frequency-shift is with respect to the rate of change of time-delay τ.f _sBe that time-delay is at τ=τ _cThe time the Doppler frequency shift value; f _SLBe time-delay τ=τ _LThe time the Doppler frequency shift value.

Through calculating the f of each Dan Jing in this example _s, m, and τ _cValue list in the table 1, then can calculate each corresponding certainty phase function constantly accordingly.

(3) realization of Stochastic Modulation function

For the decline of analog channel impulse response, (t τ) is made of a large amount of time serieses of answering at random Stochastic Modulation function ψ.On each postpones τ, makes up two independently in vain gaussian random sequence represent multiple seasonal effect in time series real part and imaginary part respectively.Therefore each multiple seasonal effect in time series whose amplitude obeys Rayleigh distributes.For the power spectrum width that limits random sequence to reach the Doppler spread spectrum that emulation needs, the width of filter is the Doppler extension width.The implementation procedure of Stochastic Modulation function as shown in Figure 6.

In implementation procedure, adopt discrete functional form with the Stochastic Modulation function representation be ψ (n, τ).

ψ(n，τ)＝x(n，τ)+iy(n，τ)

For each fixing delay τ, can be expressed as:

ψ _n＝x _n+iy _n(n＝0，1，2.....)

x _nAnd y _nBe independent random variables, constituted by different random sequences respectively, the following (ρ of its production process _n, ρ ' _nBe respectively two kinds of different randomizers and produce the gaussian random sequence of independent zero-mean):

x _n＝ρ _n+(x _n-1-ρ _n)λ

y _n＝ρ′ _n+(y _n-1-ρ′ _n)λ

X wherein ₀=(1-λ) ρ ₀, y ₀=(1-λ) ρ ' ₀

λ＝exp[-(Δt)σ _f]

The σ in Gaussian shape so the following formula is got in Doppler's expansion _fValue be:

Gaussian σ _f＝σ _D{2π/(-lnS _v)} ^1/2

σ _DBe the half-power bandwidth of Doppler expansion, S _v=A _Fl/ A, A _FLBe the signal threshold level.

The σ that this example calculates _fList in table 1 with the value of λ.

Also superpose in the impulse Response Function that each time point calculates each Dan Jing according to above-mentioned steps, can obtain the impulse response of whole channel on each time point.

The channel impulse response figure that Fig. 7 calculates for this example, each pulse among the figure represents a multipath; Can analyze the equity of channel between each moment by the relative time delay between each multipath (delay inequality between each footpath).As shown in Figure 7, from left to right used the relative time delay between each multipath T1, T2 and T3 to represent, list in table 2 relative time delay at the different time interval that calculates.By with table 3 in the data that obtain of actual measurement relatively, consistent with measured result from this simulation result of variation in each cycle of data.

The conclusion of the channel equity analysis that this is routine is:

1. lack variation in time hardly in (adjacent 10ms) relative time delay under the commitment defini interval condition;

2.300ms in, survey the variation in relative time delay less than 2us, the variation in the relative time delay that emulation obtains is less than 5us;

3. in this short commitment defini interval channel equity analysis experiment, can think: have equity at 300ms with interior channel under the tranquility of ionosphere.

The emulated data result of the short commitment defini interval situation of table 2

Cycle	T1	T2	T3
					1	477us	19us	180us
2	480us	18us	185us
					5	484us	20us	178us
10	478us	21us	181us
				20	476us	21us	186us
30	478us	25us	180us

Measured data result under the short commitment defini interval situation of table 3

Cycle	T1	T2	T3
					1	475us	7.8us	181.2us
2	475us	7.8us	181.3us
				5	475us	7.8us	179.7us
10	475us	7.8us	179.7us
				20	475us	6.2us	179.7us
30	474us	6.2us	179.7us

Claims (1)

1. one kind based on revising the method that the ionospheric channel model carries out the research of channel equity, and it is characterized in that: these method concrete steps are as follows:

Step 1: set up the international reference ionosphere model, namely set up " IRI 2007 " model;

The international reference ionosphere model is disclosed model, and the source code of its realization also is open, and we only need input communication both sides' longitude and latitude and call duration time in the process of simulation analysis, namely obtains the critical frequency f of reflection path upper ionized layer by IRI2007 _p, the electron concentration maximum height h ₀And the half thickness σ in reflector;

Step 2: in conjunction with measured data, set up the correcting module of ionosphere parameter, make the ionosphere parameter have time variation;

The similar ionosphere Changing Pattern that has in identical month according to the different anniversaries; We have introduced the modifying factor of ionosphere parameter at correcting module, and this factor is reflector reflect_layer, month M, the function of T and time t constantly; That is to say the critical frequency f in different reflector _pHeight h with the electron concentration maximum ₀To change in time in different months and the moment; According to the reflector difference, in program, call different correcting modules, in module, call corresponding modifying factor according to month M and moment T again, its principle is as follows:

f _p+＝change_f(M，T)·Δt

h ₀+＝change_h(M，T)·Δt (1)

σ＝σ

Step 3: set up the ITS channel model; In conjunction with the signal characteristic parameter of ionosphere parameter and setting, simulation calculation goes out to have the impulse Response Function of the ionospheric channel of multipath structure;

The ITS model is divided into two parts, i.e. deterministic models part and stochastic model part;

1, deterministic models part:

Utilize the corrected ionosphere of step 2 parameter: f _p, h ₀, σ calculates signal by the time delay τ behind the ionospheric propagation _c, its calculation procedure is as follows:

(1) utilizes corrected ionosphere parameter f _p, h ₀, σ is in conjunction with incoming carrier frequency parameter f _c, calculate electromagnetic signal at ionospheric usable reflection height

Computing formula is as follows:

$f_c=f_p{\left\{\right[1+{(D/2\stackrel{\‾}{h})}^2]/[1+\exp ((h_0-\stackrel{\‾}{h})/\mathrm{\σ})\left]\right\}}^{1/2}---\left(2\right)$

(2) the usable reflection height that calculates according to previous step Spherical distance D and light velocity c in conjunction with the transmitting-receiving two places calculate electromagnetic signal arrives receiving terminal through ionospheric reflection time delay τ _c, computing formula is as follows:

$\stackrel{\‾}{h}={\{{({\mathrm{c\τ}}_c/2)}^2-{(D/2)}^2\}}^{1/2}---\left(3\right)$

2, stochastic model part:

The ITS channel modeling method is namely set up the impulse Response Function of channel, and the impulse response of whole channel is defined as the impulse response sum of every propagation path, and formula (4) is the mathematic(al) representation of model;

$h(t,\mathrm{\τ})=\underset{n}{\mathrm{\Σ}}\sqrt{p_n\left(\mathrm{\τ}\right)}{D}_n(t,\mathrm{\τ}){\mathrm{\ψ}}_n(t,\mathrm{\τ})---\left(4\right)$

N in the formula (4) represents different transmission modes, and the impulse response function of each transmission mode is made up of three parts: all square function P of power time-delay section _n(τ), certainty phase function D _n(τ), Stochastic Modulation function ψ _n(τ); Wherein, the time delay expansion in power time-delay section function table reference road; The certainty phase function characterizes the Doppler frequency shift of channel; The Stochastic Modulation function then characterizes Doppler's expansion of channel;

The realization of the randomness of channel part will be divided into following three steps and finish so:

(1) realization of all square functions of power time delay section

Power delay profile is that DPP has determined the shape that retarding power distributes, by τ _c, σ _c, σ _τDetermine with four parameters such as peak power A; With P (τ) expression time-delay Power Spectrum Distribution function, broadband short wave channel mode time-delay Power Spectrum Distribution is that Gamma distributes, and its mathematical form is as follows:

$p\left(\mathrm{\τ}\right)=A\frac{{\mathrm{\α}}^{\mathrm{\α}+1}}{\mathrm{\Δ\Γ}(\mathrm{\α}+1)}{z}^{\mathrm{\α}}{e}^{-\mathrm{\αz}}---\left(5\right)$

$z=\frac{(\mathrm{\τ}-{\mathrm{\τ}}_c)}{\mathrm{\Δ}}+1---\left(6\right)$

Δ=τ wherein _c-τ _l, be used for controlling width, then z is expressed as:

z＝(τ-τ _l)/(τ _c-τ _l)＞0 (7)

Parameter alpha and τ _lWidth and symmetry that control distributes, they depend on time delay expansion and threshold level, in order to obtain their occurrence, make two parameters satisfy formula (8);

lns _v＝α(lnz _L+1-z _L)＝α(lnz _U+1-z _U) (8)

z _L＝(τ _L-τ _l)/(τ _c-τ _l)＞0 (9)

z _U＝(τ _U-τ _l)/(τ _c-τ _l)＞0 (10)

At first

lnz _L-z _L＝lnz _U-z _U

(1-z _L)/(z _U-1)＝(τ _c-τ _L)/(τ _U-τ _c)＜1 (11)

Obtain z according to following formula associating equation group with Newton iteration method _L, and then obtain parameter alpha and τ _lOccurrence;

α＝(lnz _L+1-z _L) ^-1lnS _v

S _v＝A _fl/A

τ _l＝τ _c-(τ _c-τ _L)/(1-z _L)＜τ _L

Under the condition of knowing A, α, z, obtain time-delay Power Spectrum Distribution function so;

(2) realization of certainty phase function

The certainty phase function is that DPF is used for characterizing the size that doppler frequency moves, usefulness function D (it is defined as follows for t, τ) expression:

D(t，τ)＝exp 2π[f _s+m(τ-τ _c)]t

m＝(f _s-f _sL)/(τ _c-τ _L) (12)

Have according to Euler's formula:

D(t，τ)＝COS2π[f _s+m(τ-τ _c)]t+isin 2π[f _s+m(τ-τ _c)]t (13)

Wherein: m is that Doppler frequency-shift is with respect to the rate of change of time-delay τ; f _sBe that time-delay is at τ=τ _cThe time the Doppler frequency shift value; f _SLBe time-delay τ=τ _LThe time the Doppler frequency shift value;

(3) realization of Stochastic Modulation function

For the decline of analog channel impulse response, (t τ) is made of a large amount of time serieses of answering at random Stochastic Modulation function ψ; On each postpones τ, makes up two independently in vain gaussian random sequence represent multiple seasonal effect in time series real part and imaginary part respectively; Therefore each multiple seasonal effect in time series whose amplitude obeys Rayleigh distributes; For the power spectrum width that limits random sequence to reach the Doppler spread spectrum that emulation needs, the width of filter is the Doppler extension width; The Stochastic Modulation function in implementation procedure, adopt discrete functional form with the Stochastic Modulation function representation be ψ (n, τ);

ψ(n，τ)＝x(n，τ)+iy(n，τ) (14)

For each fixing delay τ, be expressed as:

ψ _n＝x _n+iy _n(n＝0，1，2.....) (15)

x _nAnd y _nBe independent random variables, constituted by different random sequences respectively, the following (ρ of its production process _n, ρ ' _nBe respectively two kinds of different randomizers and produce the gaussian random sequence of independent zero-mean):

x _n＝ρ _n+(x _n-1-ρ _n)λ (16)

y _n＝ρ′ _n+(y _n-1-ρ′ _n)λ (17)

X wherein ₀=(1-λ) ρ ₀, y ₀=(1-λ) ρ ' ₀(18)

λ＝exp[-(Δt)σ _f] (19)

Doppler's expansion is divided into two kinds of situations: Gaussian shape and Lorentzian shape; So σ in the formula (19) _fGet different values:

Gaussian σ _f＝σ _D{2π/(-lnS _v)} ^1/2

Lorentizian σ _f＝2πσ _D{S _v/(1-S _v)} ^1/2

σ _DBe the half-power bandwidth of Doppler expansion, S _v=A _Fl/ A, A _FLBe the signal threshold level.

Images