CN103780362B - Link performance prediction method and system based on MMIB in a kind of LTE system - Google Patents

Abstract

The present invention provides an MMIB-based link performance prediction method and system in an LTE system, the method comprising: Step 101) The sending end obtains the equivalent signal-to-noise of the original code block of the current transmission code block and its retransmission code block The equivalent signal-to-noise ratio is obtained according to the signal-to-noise ratio of each symbol in the transmitted code block fed back by the receiving end; step 102) the transmitting end is based on the obtained equivalent signal-to-noise ratio of each code block and the rate matching strategy. The average number of repetitions of all symbols in the code block is mapped to obtain the average bit mutual information of the code block, and the code block includes the original code block and its retransmission code block; Step 103) The sender merges the original code block and the retransmission code block The average bit mutual information of the current transmission code block is obtained to obtain the equivalent average bit mutual information of the current transmission code block; Step 104) Under the additive Gaussian white noise channel, the block error rate is obtained according to the equivalent average bit mutual information mapping of the current transmission code block, and the comparison is completed Prediction of link status at the link layer.

Description

MMIB-based link performance prediction method and system in an LTE system

technical field

The present invention relates to link performance prediction technology in LTE, in particular to an MMIB-based link performance prediction method and system in an LTE system.

Background technique

The future broadband wireless communication system will meet the needs of multiple integrated services from voice to multimedia, requiring fast transmission of content on limited spectrum resources. Multiple-input multiple-output (MIMO) technology makes full use of space resources and uses multiple antennas to achieve multiple transmissions and multiple receptions. It can double the channel capacity without increasing spectrum resources and antenna transmission power. Orthogonal Frequency Division Multiplexing (OFDM) is a multi-carrier narrowband transmission technology, and its subcarriers are orthogonal to each other, which can efficiently use spectrum resources. The effective combination of the two can overcome the adverse effects brought by the multipath effect and frequency selective fading, improve the utilization rate of spectrum resources, and increase the system capacity. In order to further improve the reliability of data transmission, the hybrid automatic repeat request (HARQ) technology is adopted in the data link layer of the wireless communication system. HARQ is the combination of Forward Error Correction Coding (FEC) and Automatic Repeat Request (ARQ). FEC adds error correction redundancy information to the packet data to be transmitted, which can protect the integrity and reliability of data transmission within a certain range. However, when the error in the packet data exceeds the protection range of the error correction code, the ARQ is used to retransmit the data packet. MIMO-OFDM and HARQ together constitute the key technology of the next generation communication system (LTE). Therefore, research on MIMO-OFDM system technology including HARQ mechanism is a hot topic in current research.

In LTE system-level simulation and cross-layer optimization, in order to quickly evaluate system performance, accurate modeling and simulation of links is required. In current research, Link Performance Prediction (LEP) technology is generally used. The link performance prediction technology accurately predicts the real-time performance of the link through an algorithm with reasonable complexity. This performance index is usually the transmission error probability, such as the packet error rate (PER). In the past research work, several link performance prediction algorithms have been proposed, such as Equivalent Exponential Signal-to-Noise Ratio Mapping (EESM), Mean Real-time Channel Capacity (MIC), Received Bit Information Rate (RBIR), Average Bit Interaction information (MMIB), etc. Compared with other algorithms, the MMIB algorithm has higher prediction accuracy, and does not require the terminal to adopt a consistent modulation and coding method, making the algorithm more suitable for the adaptive modulation and coding process of LTE. But summarizing the current research work, there are still the following deficiencies. First of all, the current MMIB algorithm does not consider the adaptive HARQ mechanism. Therefore, the analysis of the changes in the MMIB calculation model brought about by the inclusion of the adaptive HARQ mechanism has become a problem to be solved by the new link prediction algorithm; secondly, the rate matching in LTE The mechanism makes the repetition or puncturing of the transmission code block, and then enables the receiving end to obtain different forms of decoding gain, which is not fully reflected in the current link sensing algorithm. The present invention will address these two shortcomings and propose a method based on MMIB's MIMO-OFDM HARQ system link prediction method realizes link performance prediction technology with relatively high complexity and high accuracy.

The full Chinese name corresponding to the English abbreviation is as follows:

AWGN: additive white Gaussian noise;

MMIB: mean bit mutual information;

MIMO: multiple input multiple output;

OFDM: Orthogonal Frequency Division Multiplexing;

HARQ: Hybrid Automatic Repeat Request.

Contents of the invention

The object of the present invention is to provide an MMIB-based link performance prediction method and system in an LTE system in order to overcome the above problems.

In order to achieve the above object, the present invention provides a link performance prediction method based on MMIB in an LTE system, the method is used to predict in real time the link status of the data link layer of the LTE system transmitting end and receiving end, and the method includes:

Step 101) The sending end obtains the equivalent signal-to-noise ratio of the original code block of the current transmission code block and its retransmission code block, and the equivalent signal-to-noise ratio is based on the original code block of the current transmission code block and its retransmission fed back by the receiving end The signal-to-noise ratio of each symbol in the code block is obtained;

Step 102) The sending end obtains the average bit mutual information of each code block according to the obtained equivalent signal-to-noise ratio of each code block and the average number of repetitions of all symbols in each code block when the rate matching strategy is adopted;

Step 103) The sending end merges the average bit mutual information of the original code block and the retransmitted code block obtained in the previous step, and finally obtains the equivalent average bit mutual information of the current transmission code block;

Step 104) Under the additive white Gaussian noise channel, the block error rate is obtained according to the equivalent average bit mutual information mapping of the current transmission code block, and the prediction of the link status of the link layer is completed.

In the above technical solution, the step 101) further includes:

When the transmission code block is retransmitted for the ith time, the order of modulation and coding is n ⁱ , the number of symbols is L ⁱ , and the signal-to-noise ratio vector of the symbols is Then the symbol equivalent signal-to-noise ratio of the transmitted code block is The formula is as follows:

where I _n (γ) is defined as follows,

I _n (γ) = nE _Y [Γ]

Among them, n is the order of modulation and coding, and A is a set of 2 ⁿ symbols; is a symbol set whose i bit is b, b takes the value of 0 or 1, Y~N(0,1), and N(0,1) represents a normal distribution function.

In the above technical solution, the step 102) further includes:

Step 102-1) is used to obtain the mutual information of the symbols contained in each code block:

When the equivalent signal-to-noise ratio of each symbol in the transmission code block is The modulation order of the modulation code is n ⁱ , and the average symbol repetition times of the transmission code block in the rate matching process is Then the mutual information formula of the transmission code block symbol is:

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

The first average bit mutual information of the transmission code block is further expressed as:

Wherein, n ⁱ is the modulation and coding order of the ith retransmission.

In the above technical solution, the step 103) further includes:

Step 103-1) Obtain the number of retransmissions of the code block using the HARQ strategy;

Step 103-2) Obtain the effective length of the data bits sent by the sending end after each retransmission through channel coding;

Step 103-3) According to the average bit mutual information, the number of retransmissions and the effective length, the following formula is used to combine the average bit mutual information to obtain the equivalent average bit mutual information:

Among them, C _i represents the effective length of the data in the i-th transmission, F represents the number of retransmissions, and MMIB _i represents the average bit mutual information in the i-th retransmission of the code block.

In the above technical solution, when the number of data bits is D, and it is assumed that the coding rate of the channel during the ^ith transmission is Then the equivalent average bit mutual information formula is:

Based on the above method, the present invention also provides a MMIB-based link performance prediction system in an LTE system, wherein the system includes:

The equivalent signal-to-noise ratio mapping module is used to obtain the equivalent signal-to-noise ratio of the original code block of the current transmission code block and its retransmission code block, and the equivalent signal-to-noise ratio is based on the original code of the current transmission code block fed back by the receiving end The signal-to-noise ratio of each symbol in the block and its retransmission code block is obtained;

The code block average bit mutual information mapping module is used to obtain the average bit mutual information of each code block according to the obtained equivalent signal-to-noise ratio of each code block and the average number of repetitions of all symbols in each code block when the rate matching strategy is adopted;

The equivalent average bit mutual information acquisition module is used to combine the obtained average bit mutual information of the original code block and the retransmitted code block, and finally obtain the equivalent average bit mutual information of the current transmission code block; and

The block error rate acquisition module is used to obtain the block error rate according to the equivalent average bit mutual information mapping of the current transmission code block under the additive Gaussian white noise channel, and complete the prediction of the link status of the link layer.

In the above technical solution, the equivalent signal-to-noise ratio mapping module obtains the equivalent signal-to-noise ratio of all symbols in the received code block according to the following formula:

where I _n (γ) is defined as follows:

I _n (γ) = nE _Y [Γ]

Among them, A is a set of 2 ⁿ symbols; is a set of symbols whose i bit is b, Y~CN(0,1).

In the above technical solution, the code block average bit mutual information mapping module further includes:

The inter-bit information acquisition submodule of the symbols in the code block is used to obtain the inter-bit information of the symbols contained in each code block, when the equivalent signal-to-noise ratio of the symbols in the received code block Assume that the average number of symbol repetitions of the code block in the rate matching process is Then the mutual information formula of symbols is:

The average bit mutual information acquisition submodule is used to obtain the average bit mutual information of the code block based on the mutual information of each symbol, and the first average bit mutual information of the transmission code block is further expressed as:

In the above technical solution, the equivalent average bit mutual information acquisition module further includes:

The information collection sub-module is used to obtain the number of retransmissions of the code block using the HARQ strategy and the effective length of the data bits sent by the sending end after each retransmission after channel coding;

The processing sub-module is used to combine the average bit mutual information according to the average bit mutual information, the number of retransmissions and the effective length output by the information collection sub-module to obtain the equivalent average bit mutual information by using the following formula:

Among them, C _i represents the effective length of the data in the i-th transmission, F represents the number of retransmissions, and MMIB _i represents the average bit mutual information in the i-th retransmission of the code block.

In the above technical solution, when the number of data bits is D, and it is assumed that the coding rate of the channel during the ^ith transmission is Then the equivalent average bit mutual information formula is:

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

The new MMIB link prediction algorithm has the following advantages: firstly, it fully considers the adaptive HARQ mechanism; secondly, the new algorithm takes into account the influence brought by the rate matching mechanism in LTE. These two points will make the prediction of the new link prediction algorithm more consistent with the actual system. Based on this new link prediction algorithm, on the one hand, it can obtain more accurate link status and reduce the complexity of simulation during system-level simulation; on the other hand, it can be based on this high-accuracy Link prediction algorithm Design various adaptive link algorithms, such as adaptive coding and modulation selection at the sending end.

Description of drawings

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

Fig. 2 is a schematic diagram of modeling of the rate matching process in LTE of the present invention;

FIG. 3 is a schematic diagram of the calculation model of the average bit mutual information of combined retransmission code blocks according to the present invention.

detailed description

Embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

The core of the link performance prediction technology is to provide a calculation model with reasonable complexity, which can obtain the error probability of wireless transmission under the channel condition according to the channel state of the physical layer. Usually the physical channel state is determined by a series of channel responses H and SNR, then 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)

Since a packet may consist of one or more coded data blocks (Block), the link-aware computing core is further expressed as shown in formula (2),

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

The idea of the MMIB algorithm is: under the AWGN channel, BLER and MMIB have a definite mapping relationship; therefore, firstly, the MMIB is derived based on the symbol signal-to-noise ratio (SNR), and then the BLER is obtained by using the mapping relationship between BLER and MMIB, and finally the wireless transmission error is obtained. Probability PER.

Based on the current MMIB algorithm, considering the HARQ mechanism in LTE, the present invention proposes a new MMIB link performance prediction algorithm as shown in Figure 1. The technical solution of the present invention is divided into four steps in sequence: equivalent SNR mapping, MMIB calculations, equivalent MMIB mappings, and MMIB-BLER mappings. Firstly, the signal-to-noise ratio of each symbol of the code block is mapped to the equivalent signal-to-noise ratio of the code block; the second step is based on the equivalent signal-to-noise ratio, the average number of symbol repetitions in rate matching, and the modulation order of the code block in this transmission. The MMIB of the code block; in the third step, the MMIB of the code block retransmitted multiple times is combined to obtain the equivalent MMIB including the adaptive HARQ mechanism; finally, the BLER is obtained according to the quantitative mapping relationship between the BLER and the MMIB under the AWGN channel.

1 Equivalent SNR mapping

Assume that the order of modulation and coding during the i-th (i=0,1,2,3) retransmission is ni ⁱ , the number of symbols in the transmission block is L ⁱ , and the SNR vector of the received symbol is Using Bit-Interleaved Coded Modulation (BICM) Capacity to Obtain Equivalent SNR As shown in formula (3),

Among them, the BICM capacity I _n (γ) is defined as follows,

I _n (γ) = nE _Y [Γ]

Among them, n is the order of modulation and coding, and A is a set of 2 ⁿ symbols; It is a set of symbols whose i bit is b, b takes the value of 0 or 1, Y～N(0,1), and N(0,1) is a normal distribution. When the source symbols are distributed with equal probability, the capacity of BICM is equal to the average mutual information of each symbol.

2 MMIB calculation

After the equivalent SNR of the transmission code block is obtained, the MMIB of the transmission code block is calculated based on the equivalent SNR. In the rate matching process of the LTE system, some transmission bits in the transmission code block are repeated or punctured, so that the receiving end can obtain different forms of decoding gain. In order to further accurately describe the impact of rate matching on the calculation of MMIB, the modeling of the rate matching process in LTE is shown in Figure 2: assuming that the original data bit length is D, the transmission code block length after channel coding is C, and L is the system allocation The resource block output length of , r _c is the channel coding rate, and this value is usually pre-acquired from the system settings. When L>C, in order to match the output length of the resource block, some or all of the data bits are repeated when the C-bit data passes through the rate matching module; when L<C, the C-bit data will be punctured to remove some redundant bits. It is worth mentioning that in the actual application of LTE, in order to reduce the complexity of encoding and decoding, when L<C, C is transmitted in fragments, that is, C is divided into several parts, so that the length of each part is always less than L, Therefore, there is always data duplication in LTE applications.

To sum up the description of the rate matching mechanism, there is a certain bit repetition in the transmitted data block, which will inevitably affect the mutual information of the received code block. The accumulated conditional mutual information (ACMI) can be used to calculate the mutual information when the symbol is repeated. As shown in formula (5),

Where γ _i is the SNR of each symbol, and M is the number of symbol repetitions.

In the previous module, the equivalent signal-to-noise ratio of the symbols in the received code block is obtained Assume that the average symbol repetition rate of the code block in the rate matching process is (This value is obtained by the specific rate matching algorithm), then the mutual information of the symbols is obtained as shown in formula (6):

The MMIB of the transmission code block is shown in formula (7),

3 Equivalent MMIB mapping

According to the content of the retransmission data packet and the combination and decoding methods of the receiving end, HARQ is divided into three types: Type I HARQ, Type II HARQ, and Type III HARQ. Adaptive, multi-channel parallel stop-and-wait Type I HARQ is commonly used in LTE practical applications. In this type of HARQ, when an uncorrectable error occurs in a received code block, the receiver requests the sender to retransmit the code block, and the retransmitted code block can be decoded independently or combined with previous code blocks.

In the LTE downlink transmission, due to the adoption of the adaptive HARQ mechanism, that is, when the transmitting end retransmits the code block, it can select the best modulation and coding scheme according to the current channel conditions to reduce the bit error rate to the greatest extent. Unlike previous link-aware algorithms, the impact of this adaptive mechanism on mutual information calculation will be considered. The MMIB calculation model after merging multiple retransmission code blocks is shown in Figure 3, and the MMIB of the combined code blocks after F retransmissions is regarded as the weighted average of the MMIB of F individual transmission code blocks, and each individual transmission code block has different MMIB and effective length C, the combined equivalent MMIB is calculated as shown in formula (8),

Assuming that the number of transmitted data bits is D, in the i ^th transmission, the channel coding rate is but Combining formulas (7) and (8), the equivalent MMIB can be further obtained as shown in formula (9),

4 MMIB-BLER mapping

Under the AWGN channel, there is a definite mapping relationship between MMIB and BLER as shown in formula (10),

Where b and c are constants related to the length of the code block and the coding rate.

Based on the MMIB obtained in the previous steps, the BLER of the system link can be obtained by using the mapping relationship between the MMIB and the BLER.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than limit them. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art should understand that modifications or equivalent replacements to the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention, and all of them should be included in the scope of the present invention. within the scope of the claims.

Claims (6)

1. a link performance prediction method based on MMIB in an LTE system, the method is used to predict the link status of the data link layer of the LTE system sending end and receiving end in real time, and the method includes: Step 101) The sending end obtains the equivalent signal-to-noise ratio of the original code block of the current transmission code block and its retransmission code block, and the equivalent signal-to-noise ratio is based on the original code block of the current transmission code block and its retransmission fed back by the receiving end The signal-to-noise ratio of each symbol in the code block is obtained; Step 102) The sending end obtains the average bit mutual information of each code block according to the obtained equivalent signal-to-noise ratio of each code block and the average number of repetitions of all symbols in each code block when the rate matching strategy is adopted; Step 103) The sending end merges the average bit mutual information of the original code block and the retransmitted code block obtained in the above steps, and finally obtains the equivalent average bit mutual information of the current transmission code block; Step 104) Under the additive Gaussian white noise channel, the block error rate is obtained according to the equivalent average bit mutual information mapping of the current transmission code block, and the prediction of the link status of the link layer is completed. 2. the link performance prediction method based on MMIB in the LTE system according to claim 1, is characterized in that, described step 103) further comprises: Step 103-1) Obtain the number of retransmissions of the code block using the HARQ strategy; Step 103-2) Obtain the effective length of the data bits sent by the sending end after each retransmission through channel coding; Step 103-3) According to the average bit mutual information, the number of retransmissions and the effective length, the following formula is used to combine the average bit mutual information to obtain the equivalent average bit mutual information: $\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; I</span>}\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; =</span>\frac{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; f</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; MMIB</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; f</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}$ Among them, C _i represents the effective length of the data in the i-th transmission, F represents the number of retransmissions, and MMIB _i represents the average bit mutual information in the i-th retransmission of the code block. 3. the link performance prediction method based on MMIB in the LTE system according to claim 2, is characterized in that, when described data bit number of bits is D, and assumes the coding code rate of channel when the ^ith transmission for Then the equivalent average bit mutual information formula is: ${\mathrm{<span\; class="notranslate">\; MMIB</span>}}_{\mathrm{<span\; class="notranslate">\; e</span>}\mathrm{<span\; class="notranslate">\; f</span>}\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; =</span>\frac{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; f</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; MMIB</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; f</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}}}}<span\; class="notranslate">\; .</span>$ 4. a link performance prediction system based on MMIB in an LTE system, is characterized in that, described system comprises: The equivalent signal-to-noise ratio mapping module is used to obtain the equivalent signal-to-noise ratio of the original code block of the current transmission code block and its retransmission code block, and the equivalent signal-to-noise ratio is based on the original code of the current transmission code block fed back by the receiving end The signal-to-noise ratio of each symbol in the block and its retransmission code block is obtained; The code block average bit mutual information mapping module is used to obtain the average bit mutual information of each code block according to the obtained equivalent signal-to-noise ratio of each code block and the average number of repetitions of all symbols in each code block when the rate matching strategy is adopted; The equivalent average bit mutual information acquisition module is used to combine the obtained average bit mutual information of the original code block and the retransmitted code block, and finally obtain the equivalent average bit mutual information of the current transmission code block; and The block error rate acquisition module is used to obtain the block error rate according to the equivalent average bit mutual information mapping of the current transmission code block under the additive Gaussian white noise channel, and complete the prediction of the link status of the link layer. 5. the link performance prediction system based on MMIB in the LTE system according to claim 4, is characterized in that, described equivalent average bit mutual information acquisition module further comprises: The information collection sub-module is used to obtain the number of retransmissions of the code block using the HARQ strategy and the effective length of the data bits sent by the sending end after each retransmission after channel coding; The processing sub-module is used to combine the average bit mutual information according to the average bit mutual information, the number of retransmissions and the effective length output by the information collection sub-module to obtain the equivalent average bit mutual information by using the following formula: $\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; I</span>}\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; =</span>\frac{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; f</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; MMIB</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; f</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}$ Among them, C _i represents the effective length of the data in the i-th transmission, F represents the number of retransmissions, and MMIB _i represents the average bit mutual information in the i-th retransmission of the code block. 6. the link performance prediction system based on MMIB in the LTE system according to claim 5, is characterized in that, when described data bit number of bits is D, and assumes the encoding code rate of channel when the ^ith transmission for Then the equivalent average bit mutual information formula is: ${\mathrm{<span\; class="notranslate">\; MMIB</span>}}_{\mathrm{<span\; class="notranslate">\; e</span>}\mathrm{<span\; class="notranslate">\; f</span>}\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; =</span>\frac{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; f</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; MMIB</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; f</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}}}}<span\; class="notranslate">\; .</span>$