CN103780362A - MMIB-based link performance prediction method and system in LTE system - Google Patents

Abstract

The invention provides a MMIB-based link performance prediction method in an LTE system and a MMIB-based link performance prediction system in the LTE system. The method comprises the following steps that: step 101) a transmitting end obtains the equivalent signal-to-noise ratio of an original code block of a current transmission code block and a retransmission code block of the original code block, wherein the equivalent signal-to-noise ratio can be obtained according to signal-to-noise ratios of various symbols in a transmission code block fed back by a receiving end; step 102) the transmitting end performs mapping according to the obtained equivalent signal-to-noise ratio of each code block and the average repetition times of all symbols in the code blocks when a rate matching strategy is adopted, so as to obtain the mean mutual information per bit of the code blocks, wherein the code blocks include the original code block and the retransmission code block thereof; step 103) the transmitting end merges the mean mutual information per bit of the original code block and the retransmission code block so as to obtain the equivalent mean mutual information per bit of the current transmission code block; and step 104) mapping is performed according to the equivalent mean mutual information per bit of the current transmission code block under an additive Gauss white noise channel so as to obtain block error rate, and the prediction of the link status of a link layer can be completed.

Description

Link performance prediction method and system based on MMIB in a kind of LTE system

Technical field

The present invention relates to the link performance prediction technology in LTE, be specifically related to the link performance prediction method and system based on MMIB in a kind of LTE system.

Background technology

Following system of broadband wireless communication, by the multiple integrated service demand meeting from speech to multimedia, requires to realize the fast transport of content on limited frequency spectrum resource.Multiple-input and multiple-output (MIMO) technology is filled and is utilized space resources, and utilizes multiple antenna to realize MIMO, in the situation that not needing to increase frequency spectrum resource and antenna transmitted power, can improve exponentially channel capacity.OFDM (OFDM) is a kind of multicarrier narrow band transmission technology, mutually orthogonal between its subcarrier, can utilize efficiently frequency spectrum resource.The effective combination of the two can overcome the harmful effect that multipath effect and frequency selective fading are brought, and improves frequency spectrum resource utilization rate, increases power system capacity.In order further to improve the reliability of transfer of data, adopt in the data link layer of wireless communication system automatic repeat request (HARQ) technology of mixing.HARQ is the product that forward error correction coding (FEC) and HARQ (ARQ) combine.FEC adds error correction redundant information in grouped data waiting for transmission, within the specific limits integrality and the reliability of protected data transmission.But when the mistake occurring in grouped data exceeds the protection range of error correcting code, carry out the re-transmission of packet by ARQ.MIMO-OFDM and HARQ have formed the key technology of next generation communication system (LTE) jointly, and therefore, the MIMO-OFDM systems technology research that comprises HARQ mechanism is the hot issue of current research.

In the system-level emulation of LTE and cross-layer optimizing, for RES(rapid evaluation system) performance, need to carry out Accurate Model emulation to link, in current research, generally adopt link performance prediction (LEP) technology.Link performance prediction technology is the algorithm that has a reasonable complexity by, predicts accurately the instant performance of link, and these performance index are generally transmission error probability, as Packet Error Ratio (PER) etc.In research work in the past, several link performance prediction algorithms have been carried out, as equivalent index signal to noise ratio mapping (EESM), average Real-time Channel capacity (MIC), received bit information rate (RBIR), average bit mutual information (MMIB) etc.Compare other algorithm, MMIB algorithm has higher forecasting accuracy, and does not require that terminal adopts consistent modulation and coded system, makes the adaptive modulation and coding process of the more applicable LTE of this algorithm.But sum up current research work, still have the following disadvantages.First, current MMIB algorithm is not considered adaptive HARQ mechanism, and therefore, analysis package changes containing the MMIB computation model bringing after adaptive HARQ mechanism, becomes the problem that new link prediction algorithm need to solve; Secondly, LTE medium-rate matching mechanisms makes to transmit repetition or the perforation of code block, and then make receiving terminal obtain multi-form decoding gain, this is not also embodied fully in current link perception algorithm, the present invention will be for these 2 deficiencies, propose a kind of MIMO-OFDM HARQ system link Forecasting Methodology based on MMIB, realized compared with the link performance prediction technology of ground complexity and high accuracy.

It is as follows that corresponding Chinese full name write a Chinese character in simplified form in English:

AWGN: additive white Gaussian noise;

MMIB: average bit mutual information;

MIMO: multiple-input and multiple-output;

OFDM: OFDM;

HARQ: mix automatic repeat request.

Summary of the invention

The object of the invention is to, for overcoming the problems referred to above, the invention provides the link performance prediction method and system based on MMIB in a kind of LTE system.

To achieve these goals, the invention provides a kind of link performance prediction method based on MMIB in LTE system, the method is for the link circuit condition of the data link layer of real-time estimate LTE system transmitting terminal and receiving terminal, and described method comprises:

Step 101) transmitting terminal obtains the original code block of current transmission code block and retransmits the equivalent signal-to-noise ratio of code block, and this equivalent signal-to-noise ratio obtains according to the original code block of the current transmission code block of receiving terminal feedback and the signal to noise ratio that retransmits each symbol in code block thereof;

Step 102) transmitting terminal according to the equivalent signal-to-noise ratio of the above-mentioned each code block obtaining and while adopting rate-matched strategy in each code block the average number of repetition mapping of all symbols obtain the average bit mutual information of each code block;

Step 103) transmitting terminal merges the original code block that upper step obtains and the average bit mutual information that retransmits code block, finally obtains the average bit mutual information of equivalence of current transmission code block;

Step 104) under additive white Gaussian noise channel, shine upon and obtain Block Error Rate according to the average bit mutual information of equivalence of current transmission code block, complete the prediction to link layer link situation.

In technique scheme, described step 101) further comprise:

In the time that transmission code block retransmits for the i time, the exponent number of modulating-coding is n ⁱ, symbol numbers is L ⁱ, the signal to noise ratio vector of symbol is

the symbolic equivalent signal to noise ratio of this transmission code block

formula as follows:

${\mathrm{\γ}}_{\mathrm{eff}}^i=I_{n^i}^{-1}\left(\frac{1}{L^i}\underset{{\mathrm{\γ}}_j^i\∈{\mathrm{\γ}}^i}{\mathrm{\Σ}}{I}_{n^i}\left({\mathrm{\γ}}_j^i\right)\right)$

Wherein I _n(γ) be defined as follows,

I _n(γ)＝n-E _Y[Γ]

$\mathrm{\Γ}=\frac{1}{2^n}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{b=0}{\overset{1}{\mathrm{\Σ}}}\underset{z\∈A_b^i}{\mathrm{\Σ}}{\log }_2\frac{\underset{\widehat{\mathrm{\α}}\∈A}{\mathrm{\Σ}}\exp (-{|Y-\sqrt{\mathrm{\γ}}(\widehat{\mathrm{\α}}-z)|}^2)}{\underset{\widetilde{\mathrm{\α}}\∈A_b^i}{\mathrm{\Σ}}\exp (-{|Y-\sqrt{\mathrm{\γ}}(\widetilde{\mathrm{\α}}-z)|}^2)}$

Wherein, the exponent number that n is modulating-coding, A is 2 ⁿindividual assemble of symbol;

for the i bit assemble of symbol that is b, b value 0 or 1, Y～N (0,1), and N (0,1) represents normal distyribution function.

In technique scheme, described step 102) further comprise:

Step 102-1) for obtaining the step of mutual information of the symbol that each code block comprises; When the equivalent signal-to-noise ratio of each symbol in transmission code block is

the order of modulation of modulating-coding is n ⁱ, in rate-matched process, the Symbol average number of repetition of this transmission code block is

the mutual information formula of this transmission code block symbol is:

$I=I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\γ}}_{\mathrm{eff}}^i\right)$

Step 102-2) obtain the average bit mutual information of this code block for the mutual information based on transmission code block each symbol:

When the order of modulation of the modulating-coding adopting in this transmission is n ⁱtime, first of this transmission code block the average bit mutual information is further expressed as:

$\mathrm{MMIB}=\frac{I}{n^i}=\frac{I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\γ}}_{\mathrm{eff}}^i\right)}{n^i}.$

Wherein, n ⁱmodulating-coding exponent number while being the i time re-transmission.

In technique scheme, described step 103) further comprise:

Step 103-1) obtain the number of retransmissions that adopts HARQ strategy code-aiming block;

Step 103-2) data bit that sends when obtaining transmitting terminal and retransmitting the at every turn data effective length after via chnnel coding;

Step 103-3) adopt the each average bit mutual information of following formula merging to obtain the average bit mutual information of equivalence according to average bit mutual information, number of retransmissions and effective length:

$\mathrm{MMIB}=\frac{\underset{i=1}{\overset{F}{\mathrm{\Σ}}}{C}_i{\mathrm{MMIB}}_i}{\underset{i=1}{\overset{F}{\mathrm{\Σ}}}{C}_i}$

Wherein, C _idata effective length while representing the i time transmission, F represents number of retransmissions, MMIB _iaverage bit mutual information while representing to retransmit code block the i time.

In technique scheme, in the time that described data bit figure place is D, and hypothesis is at i ^thwhen inferior transmission, the encoder bit rate of channel is the average bit mutual information formula of equivalence is:

${\mathrm{MMIB}}_{\mathrm{eff}}=\frac{\underset{i=0}{\overset{F}{\mathrm{\Σ}}}\frac{1}{r_c^i}*{\mathrm{MMIB}}_i}{\underset{i=0}{\overset{F}{\mathrm{\Σ}}}\frac{1}{r_c^i}}.$

Also provide the link performance prediction system based on MMIB in a kind of LTE system based on said method the present invention, it is characterized in that, described system comprises:

Equivalent signal-to-noise ratio mapping block, for obtaining the original code block of current transmission code block and the equivalent signal-to-noise ratio of re-transmission code block thereof, in the original code block of the current transmission code block that this equivalent signal-to-noise ratio feeds back according to receiving terminal and re-transmission code block thereof, the signal to noise ratio of each symbol obtains;

The average bit mutual information of code block mapping block, obtains the average bit mutual information of each code block for the average number of repetition mapping of all symbols of each code block according to the equivalent signal-to-noise ratio of the each code block obtaining and while adopting rate-matched strategy;

The average bit mutual information acquisition module of equivalence, for merging the original code block obtaining and the average bit mutual information that retransmits code block, finally obtains the average bit mutual information of equivalence of current transmission code block; With

Block Error Rate acquisition module, obtains Block Error Rate for the average bit mutual information mapping of equivalence according to current transmission code block under additive white Gaussian noise channel, completes the prediction to link layer link situation.

In technique scheme, described equivalent signal-to-noise ratio mapping block obtains receiving the equivalent signal-to-noise ratio of all symbols in code block according to following formula:

${\mathrm{\γ}}_{\mathrm{eff}}^i=I_{n^i}^{-1}\left(\frac{1}{L^i}\underset{{\mathrm{\γ}}_j^i\∈{\mathrm{\γ}}^i}{\mathrm{\Σ}}{I}_{n^i}\left({\mathrm{\γ}}_j^i\right)\right)$

Wherein I _n(γ) be defined as follows:

I _n(γ)＝n-E _Y[Γ]

$\mathrm{\Γ}=\frac{1}{2^n}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{b=0}{\overset{1}{\mathrm{\Σ}}}\underset{z\∈A_b^i}{\mathrm{\Σ}}{\log }_2\frac{\underset{\widehat{\mathrm{\α}}\∈A}{\mathrm{\Σ}}\exp (-{|Y-\sqrt{\mathrm{\γ}}(\widehat{\mathrm{\α}}-z)|}^2)}{\underset{\widetilde{\mathrm{\α}}\∈A_b^i}{\mathrm{\Σ}}\exp (-{|Y-\sqrt{\mathrm{\γ}}(\widetilde{\mathrm{\α}}-z)|}^2)}$

Wherein, A is 2 ⁿindividual assemble of symbol;

for i bit be b assemble of symbol, Y～CN (0,1).

In technique scheme, the average bit mutual information of described code block mapping block further comprises:

In code block, the mutual bit information of symbol obtains submodule, for obtaining the mutual bit information of the symbol that each code block comprises, when receiving the equivalent signal-to-noise ratio of symbol in code block

the Symbol average number of repetition of supposing this code block in rate-matched process is

the mutual information formula of then symbol is:

$I=I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\γ}}_{\mathrm{eff}}^i\right)$

Average bit mutual information obtains submodule, obtains the average bit mutual information of code block for the mutual information based on each symbol, when the order of modulation of the modulating-coding adopting in this transmission is n ⁱtime, first of this transmission code block the average bit mutual information is further expressed as:

$\mathrm{MMIB}=\frac{I}{n^i}=\frac{I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\γ}}_{\mathrm{eff}}^i\right)}{n^i}.$

In technique scheme, the average bit mutual information of described equivalence acquisition module further comprises:

Information gathering submodule, the data effective length of the data bit sending while adopting the number of retransmissions of HARQ strategy code-aiming block and transmitting terminal at every turn to retransmit for obtaining after via chnnel coding;

Processing submodule, adopts the each average bit mutual information of following formula merging to obtain the average bit mutual information of equivalence for the number of retransmissions according to average bit mutual information and the output of information gathering submodule and effective length:

$\mathrm{MMIB}=\frac{\underset{i=1}{\overset{F}{\mathrm{\Σ}}}{C}_i{\mathrm{MMIB}}_i}{\underset{i=1}{\overset{F}{\mathrm{\Σ}}}{C}_i}$

Wherein, C _idata effective length while representing the i time transmission, F represents number of retransmissions, MMIB _iaverage bit mutual information while representing to retransmit code block the i time.

In technique scheme, in the time that described data bit figure place is D, and hypothesis is at i ^thwhen inferior transmission, the encoder bit rate of channel is the average bit mutual information formula of equivalence is:

${\mathrm{MMIB}}_{\mathrm{eff}}=\frac{\underset{i=0}{\overset{F}{\mathrm{\Σ}}}\frac{1}{r_c^i}*{\mathrm{MMIB}}_i}{\underset{i=0}{\overset{F}{\mathrm{\Σ}}}\frac{1}{r_c^i}}.$

Compared with prior art, technical advantage of the present invention is:

New MMIB link prediction algorithm has following advantage: first taken into full account adaptive HARQ mechanism; Secondly, new algorithm has been considered the impact that LTE medium-rate matching mechanisms brings.This two names a person for a particular job and makes the more realistic system of prediction of new link prediction algorithm.Based on this new link prediction algorithm, in the time carrying out system-level emulation, can obtain Link State more accurately on the one hand, reduce the complexity of emulation; On the other hand in actual applications can be based on this various link adaptation algorithms of link prediction algorithm design with high accuracy, as adaptive coding and modulating mode selection at transmitting terminal etc.

Accompanying drawing explanation

Fig. 1 is new link performance prediction method provided by the invention;

Fig. 2 is the rate-matched process model building schematic diagram in LTE of the present invention;

Fig. 3 is the average bit mutual information computation model schematic diagram that merging of the present invention retransmits code block.

Embodiment

Below in conjunction with accompanying drawing, embodiments of the present invention are further described.

The core of link performance prediction technology is to provide a computation model that has reasonable complexity and can obtain according to the channel status of physical layer the wireless transmission error probability under this channel condition.Conventionally physical channel state is determined by a series of channel response H and SNR, and the core of link performance prediction is to determine as formula (1),

PER=f (H ₁, γ ₁, H ₂, γ ₂... H _n, γ _n) (1), because a grouping may be made up of one or more coded data blocks (Block), therefore link perception is calculated the further expression of core as shown in Equation (2),

PER＝BLER(H ₁,γ ₁,H ₂,γ ₂,...H _n,γ _n)?（2）

The thinking of MMIB algorithm is: under awgn channel, BLER and MMIB have definite mapping relations; Therefore first derive MMIB based on symbol signal to noise ratio snr, and then utilize the mapping relations between BLER and MMIB to obtain BLER, finally obtain wireless transmission error probability P ER.

MMIB algorithm based on current, consider the HARQ mechanism in LTE, the present invention proposes new MMIB link performance prediction algorithm as shown in Figure 1, and technical scheme of the present invention is divided into successively and comprises four steps: equivalent signal-to-noise ratio mapping, MMIB calculate, equivalent MMIB shines upon and MMIB-BLER mapping.First each symbol signal to noise ratio of code block is mapped as to the equivalent signal-to-noise ratio of code block; The order of modulation of second step based on Symbol average number of repetition in equivalent signal-to-noise ratio, rate-matched and this transmission code block obtains the MMIB of code block; The 3rd step merges the code block MMIB repeatedly retransmitting, and obtains comprising the equivalent MMIB after adaptive HARQ mechanism; Finally obtain BLER according to mapping relations quantitative between BLER under awgn channel and MMIB.

1 equivalent signal-to-noise ratio mapping

Suppose i(i=0,1,2,3) exponent number of modulating-coding when inferior re-transmission is n ⁱ, the symbol numbers of transmission block is L ⁱ, the SNR vector of receiving symbol is

utilize Bit Interleave coded modulation (BICM) capacity to obtain equivalent signal-to-noise ratio

as shown in Equation (3),

${\mathrm{\γ}}_{\mathrm{eff}}^i=I_{n^i}^{-1}\left(\frac{1}{L^i}\underset{{\mathrm{\γ}}_j^i\∈{\mathrm{\γ}}^i}{\mathrm{\Σ}}{I}_{n^i}\left({\mathrm{\γ}}_j^i\right)\right)---\left(3\right)$

Wherein, Bit Interleave coded modulation BICM capacity I _n(γ) be defined as follows,

I _n(γ)＝n-E _Y[Γ]

$\mathrm{\Γ}=\frac{1}{2^n}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{b=0}{\overset{1}{\mathrm{\Σ}}}\underset{z\∈A_b^i}{\mathrm{\Σ}}{\log }_2\frac{\underset{\widehat{\mathrm{\α}}\∈A}{\mathrm{\Σ}}\exp (-{|Y-\sqrt{\mathrm{\γ}}(\widehat{\mathrm{\α}}-z)|}^2)}{\underset{\widetilde{\mathrm{\α}}\∈A_b^i}{\mathrm{\Σ}}\exp (-{|Y-\sqrt{\mathrm{\γ}}(\widetilde{\mathrm{\α}}-z)|}^2)}---\left(4\right)$

Wherein, the exponent number that n is modulating-coding, A is 2 ⁿindividual assemble of symbol;

for i bit be b assemble of symbol, b value 0 or 1, Y～N (0,1), N(0,1) be normal distribution.In the time that source symbol equiprobability distributes, Bit Interleave coded modulation BICM capacity is equal to the Average Mutual of each symbol.

2MMIB calculates

After obtaining transmitting code block equivalent signal-to-noise ratio, based on the MMIB of this equivalence snr computation transmission code block.In the rate-matched process of LTE system, the part transmitted bit in transmission code block is carried out to repetition or perforation, and then make receiving terminal obtain multi-form decoding gain.The impact of MMIB being calculated in order further to describe exactly rate-matched, to the modeling of LTE medium-rate matching process as shown in Figure 2: suppose that original data bits length is D, after chnnel coding, transmission code block length is C, the Resource Block output length that L is system assignment, r _cfor channel encoding rate, this numerical value obtains in advance conventionally from system arranges.Work as L>C, in order to mate Resource Block output length, when C Bit data process rate-matched module, have part or all of data bit to repeat; In the time of L<C, will bore a hole to C Bit data, destroy partial redundance bit.It is worth mentioning that, in the practical application of LTE, in order to reduce encoding and decoding complexity, in the time of L<C, C is carried out to burst transmission, be divided into some parts by C, make the length of every part always be less than L, therefore in LTE application, always have the repetition of data.

The to sum up description to rate-matched mechanism exists certain position to repeat in transmission data block, and this will inevitably impact the mutual information that receives code block, the mutual information can carry out compute sign repetition with accumulative total conditional mutual information (ACMI) time as shown in Equation (5),

$I=I_n\left(\underset{i=0}{\overset{M}{\mathrm{\&Sigma;}}}{\mathrm{\&gamma;}}_i\right)---\left(5\right)$

γ in formula _ifor the SNR of each symbol, M is symbol number of repetition.

In prior module, obtain receiving the equivalent signal-to-noise ratio of symbol in code block

the Symbol average repetition rate of supposing this code block in rate-matched process is

(this value is obtained by concrete speed matching algorithm), obtains the mutual information of symbol as shown in Equation (6):

$I=I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\&gamma;}}_{\mathrm{eff}}^i\right)$

When the order of modulation under the Modulation and Coding Scheme that this transmission adopts is n ⁱtime, the MMIB of this transmission code block as shown in Equation (7),

$\mathrm{MMIB}=\frac{1}{n^i}=\frac{I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\&gamma;}}_{\mathrm{eff}}^i\right)}{n^i}---\left(7\right)$

3 equivalent MMIB mappings

Merge the mode difference of decoding according to the difference of retransmission data packet may content and receiving terminal, HARQ is divided into three types: I type HARQ, II type HARQ, III type HARQ.What in LTE practical application, generally adopt self adaptation, multi-channel parallel the I type HARQ such as stops.In the HARQ of the type, when the code block that receives occur correcting wrong time, receiving terminal request transmitting terminal transmits this code block again, the code block of transmission can be decoded separately and also can be merged decoding with code block before again.

In the time of LTE downlink transmission, owing to having adopted adaptive HARQ mechanism, be that transmitting terminal is in the time resending this code block, can select best Modulation and Coding Scheme according to current channel conditions, farthest reduce the error rate, the present invention is different from previous link perception algorithm, will consider that this adaptation mechanism calculates the impact bringing on mutual information.Repeatedly retransmit MMIB computation model after code block as shown in Figure 3 by merging, after F re-transmission merged, the MMIB of code block regards the MMIB weighted average of F individual transmission code block as, each individual transmission code block has different MMIB and effective length C, equivalent MMIB after merging calculates as shown in Equation (8)

$\mathrm{MMIB}=\frac{\underset{i=1}{\overset{F}{\mathrm{\&Sigma;}}}{C}_i{\mathrm{MMIB}}_i}{\underset{i=1}{\overset{F}{\mathrm{\&Sigma;}}}{C}_i}---\left(8\right)$

Suppose that sending data bit figure place is D, at i ^thwhen inferior transmission, chnnel coding code check is

?

in conjunction with formula (7) and (8), further can obtain equivalent MMIB as shown in Equation (9),

${\mathrm{MMIB}}_{\mathrm{eff}}=\frac{\underset{i=0}{\overset{F}{\mathrm{\&Sigma;}}}\frac{1}{r_c^i}*\frac{I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\&gamma;}}_{\mathrm{eff}}^i\right)}{n^i}}{\underset{i=0}{\overset{F}{\mathrm{\&Sigma;}}}\frac{1}{r_c^i}}---\left(9\right)$

4MMIB-BLER mapping

Under awgn channel, between MMIB and BLER, there are definite mapping relations as shown in Equation (10),

${\mathrm{BLER}}_{\mathrm{BCR}}=\frac{1}{2}[1-\mathrm{erf}\left(\frac{x-b_{\mathrm{BCR}}}{\sqrt{2}*c_{\mathrm{BCR}}}\right)]---\left(10\right)$

Wherein b is the constant relevant to code block length and encoding rate with c.

The MMIB obtaining based on preceding step, utilizes the mapping relations between MMIB and BLER, can obtain the BLER of system link.

It should be noted last that, above embodiment is only unrestricted in order to technical scheme of the present invention to be described.Although the present invention is had been described in detail with reference to embodiment, those of ordinary skill in the art is to be understood that, technical scheme of the present invention is modified or is equal to replacement, do not depart from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of claim scope of the present invention.

Claims (10)

1. the link performance prediction method based on MMIB in LTE system, the method is for the link circuit condition of the data link layer of real-time estimate LTE system transmitting terminal and receiving terminal, and described method comprises:

Step 101) transmitting terminal obtains the original code block of current transmission code block and retransmits the equivalent signal-to-noise ratio of code block, and this equivalent signal-to-noise ratio obtains according to the original code block of the current transmission code block of receiving terminal feedback and the signal to noise ratio that retransmits each symbol in code block thereof;

Step 102) transmitting terminal according to the equivalent signal-to-noise ratio of the above-mentioned each code block obtaining and while adopting rate-matched strategy in each code block the average number of repetition mapping of all symbols obtain the average bit mutual information of each code block;

Step 103) transmitting terminal merges the original code block that obtains of above-mentioned steps and retransmits the average bit mutual information of code block, finally obtains the average bit mutual information of equivalence of current transmission code block;

Step 104) under additive white Gaussian noise channel, shine upon and obtain Block Error Rate according to the average bit mutual information of equivalence of current transmission code block, complete the prediction to link layer link situation.

2. the link performance prediction method based on MMIB in LTE system according to claim 1, is characterized in that described step 101) further comprise:

In the time that transmission code block retransmits for the i time, the exponent number of modulating-coding is n ⁱ, symbol numbers is L ⁱ, the signal to noise ratio vector of symbol is

the symbolic equivalent signal to noise ratio of this transmission code block

formula as follows:

${\mathrm{\&gamma;}}_{\mathrm{eff}}^i=I_{n^i}^{-1}\left(\frac{1}{L^i}\underset{{\mathrm{\&gamma;}}_j^i\&Element;{\mathrm{\&gamma;}}^i}{\mathrm{\&Sigma;}}{I}_{n^i}\left({\mathrm{\&gamma;}}_j^i\right)\right)$

Wherein I _n(γ) be defined as follows,

I _n(γ)＝n-E _Y[Γ]

$\mathrm{\&Gamma;}=\frac{1}{2^n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\underset{b=0}{\overset{1}{\mathrm{\&Sigma;}}}\underset{z\&Element;A_b^i}{\mathrm{\&Sigma;}}{\log }_2\frac{\underset{\widehat{\mathrm{\&alpha;}}\&Element;A}{\mathrm{\&Sigma;}}\exp (-{|Y-\sqrt{\mathrm{\&gamma;}}(\widehat{\mathrm{\&alpha;}}-z)|}^2)}{\underset{\widetilde{\mathrm{\&alpha;}}\&Element;A_b^i}{\mathrm{\&Sigma;}}\exp (-{|Y-\sqrt{\mathrm{\&gamma;}}(\widetilde{\mathrm{\&alpha;}}-z)|}^2)}$

Wherein, the exponent number that n is modulating-coding, A is 2 ⁿindividual assemble of symbol;

for the i bit assemble of symbol that is b, b value 0 or 1, Y～N (0,1), and N (0,1) represents normal distyribution function.

3. the link performance prediction method based on MMIB in LTE system according to claim 1, is characterized in that described step 102) further comprise:

Step 102-1) for obtaining the step of mutual information of the symbol that each code block comprises;

When the equivalent signal-to-noise ratio of each symbol in transmission code block is

the order of modulation of modulating-coding is n ⁱ, in rate-matched process, the Symbol average number of repetition of this transmission code block is

the mutual information formula of this transmission code block symbol is:

$I=I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\&gamma;}}_{\mathrm{eff}}^i\right)$

Step 102-2) obtain the average bit mutual information of this code block for the mutual information based on transmission code block each symbol:

When the order of modulation of the modulating-coding adopting in this transmission is n ⁱtime, first of this transmission code block the average bit mutual information is further expressed as:

$\mathrm{MMIB}=\frac{1}{n^i}=\frac{I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\&gamma;}}_{\mathrm{eff}}^i\right)}{n^i}.$

Wherein, n ⁱmodulating-coding exponent number while being the i time re-transmission.

4. the link performance prediction method based on MMIB in LTE system according to claim 1, is characterized in that described step 103) further comprise:

Step 103-1) obtain the number of retransmissions that adopts HARQ strategy code-aiming block;

Step 103-2) data bit that sends when obtaining transmitting terminal and retransmitting the at every turn data effective length after via chnnel coding;

Step 103-3) adopt the each average bit mutual information of following formula merging to obtain the average bit mutual information of equivalence according to average bit mutual information, number of retransmissions and effective length:

$\mathrm{MMIB}=\frac{\underset{i=1}{\overset{F}{\mathrm{\&Sigma;}}}{C}_i{\mathrm{MMIB}}_i}{\underset{i=1}{\overset{F}{\mathrm{\&Sigma;}}}{C}_i}$

Wherein, C _idata effective length while representing the i time transmission, F represents number of retransmissions, MMIB _iaverage bit mutual information while representing to retransmit code block the i time.

5. the link performance prediction method based on MMIB in LTE system according to claim 4, is characterized in that, in the time that described data bit figure place is D, and hypothesis is at i ^thwhen inferior transmission, the encoder bit rate of channel is

the average bit mutual information formula of equivalence is:

${\mathrm{MMIB}}_{\mathrm{eff}}=\frac{\underset{i=0}{\overset{F}{\mathrm{\&Sigma;}}}\frac{1}{r_c^i}*{\mathrm{MMIB}}_i}{\underset{i=0}{\overset{F}{\mathrm{\&Sigma;}}}\frac{1}{r_c^i}}.$

6. the link performance prediction system based on MMIB in LTE system, is characterized in that, described system comprises:

Equivalent signal-to-noise ratio mapping block, for obtaining the original code block of current transmission code block and the equivalent signal-to-noise ratio of re-transmission code block thereof, in the original code block of the current transmission code block that this equivalent signal-to-noise ratio feeds back according to receiving terminal and re-transmission code block thereof, the signal to noise ratio of each symbol obtains;

The average bit mutual information of code block mapping block, obtains the average bit mutual information of each code block for the average number of repetition mapping of all symbols of each code block according to the equivalent signal-to-noise ratio of the each code block obtaining and while adopting rate-matched strategy;

The average bit mutual information acquisition module of equivalence, for merging the original code block obtaining and the average bit mutual information that retransmits code block, finally obtains the average bit mutual information of equivalence of current transmission code block; With

Block Error Rate acquisition module, obtains Block Error Rate for the average bit mutual information mapping of equivalence according to current transmission code block under additive white Gaussian noise channel, completes the prediction to link layer link situation.

7. the link performance prediction system based on MMIB in LTE system according to claim 6, is characterized in that, described equivalent signal-to-noise ratio mapping block obtains receiving the equivalent signal-to-noise ratio of all symbols in code block according to following formula:

${\mathrm{\&gamma;}}_{\mathrm{eff}}^i=I_{n^i}^{-1}\left(\frac{1}{L^i}\underset{{\mathrm{\&gamma;}}_j^i\&Element;{\mathrm{\&gamma;}}^i}{\mathrm{\&Sigma;}}{I}_{n^i}\left({\mathrm{\&gamma;}}_j^i\right)\right)$

Wherein I _n(γ) be defined as follows:

I _n(γ)＝n-E _Y[Γ]

$\mathrm{\&Gamma;}=\frac{1}{2^n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\underset{b=0}{\overset{1}{\mathrm{\&Sigma;}}}\underset{z\&Element;A_b^i}{\mathrm{\&Sigma;}}{\log }_2\frac{\underset{\widehat{\mathrm{\&alpha;}}\&Element;A}{\mathrm{\&Sigma;}}\exp (-{|Y-\sqrt{\mathrm{\&gamma;}}(\widehat{\mathrm{\&alpha;}}-z)|}^2)}{\underset{\widetilde{\mathrm{\&alpha;}}\&Element;A_b^i}{\mathrm{\&Sigma;}}\exp (-{|Y-\sqrt{\mathrm{\&gamma;}}(\widetilde{\mathrm{\&alpha;}}-z)|}^2)}$

Wherein, A is 2 ⁿindividual assemble of symbol;

for i bit be b assemble of symbol, Y～CN (0,1).

8. the link performance prediction system based on MMIB in LTE system according to claim 6, is characterized in that, the average bit mutual information of described code block mapping block further comprises:

In code block, the mutual bit information of symbol obtains submodule, for obtaining the mutual bit information of the symbol that each code block comprises, when receiving the equivalent signal-to-noise ratio of symbol in code block the Symbol average number of repetition of supposing this code block in rate-matched process is

the mutual information formula of then symbol is:

$I=I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\&gamma;}}_{\mathrm{eff}}^i\right)$

Average bit mutual information obtains submodule, obtains the average bit mutual information of code block for the mutual information based on each symbol, when the order of modulation of the modulating-coding adopting in this transmission is n ⁱtime, first of this transmission code block the average bit mutual information is further expressed as:

$\mathrm{MMIB}=\frac{1}{n^i}=\frac{I_{n^i}\left(N_{\mathrm{rep}}^i{\mathrm{\&gamma;}}_{\mathrm{eff}}^i\right)}{n^i}.$

9. the link performance prediction system based on MMIB in LTE system according to claim 6, is characterized in that, the average bit mutual information of described equivalence acquisition module further comprises:

Information gathering submodule, the data effective length of the data bit sending while adopting the number of retransmissions of HARQ strategy code-aiming block and transmitting terminal at every turn to retransmit for obtaining after via chnnel coding;

Processing submodule, adopts the each average bit mutual information of following formula merging to obtain the average bit mutual information of equivalence for the number of retransmissions according to average bit mutual information and the output of information gathering submodule and effective length:

$\mathrm{MMIB}=\frac{\underset{i=1}{\overset{F}{\mathrm{\&Sigma;}}}{C}_i{\mathrm{MMIB}}_i}{\underset{i=1}{\overset{F}{\mathrm{\&Sigma;}}}{C}_i}$

Wherein, C _idata effective length while representing the i time transmission, F represents number of retransmissions, MMIB _iaverage bit mutual information while representing to retransmit code block the i time.

10. the link performance prediction system based on MMIB in LTE system according to claim 9, is characterized in that, in the time that described data bit figure place is D, and hypothesis is at i ^thwhen inferior transmission, the encoder bit rate of channel is

the average bit mutual information formula of equivalence is:

${\mathrm{MMIB}}_{\mathrm{eff}}=\frac{\underset{i=0}{\overset{F}{\mathrm{\&Sigma;}}}\frac{1}{r_c^i}*{\mathrm{MMIB}}_i}{\underset{i=0}{\overset{F}{\mathrm{\&Sigma;}}}\frac{1}{r_c^i}}.$

Images