CN101415208A - Self-adapting modulation encode method adapting for mobile multicast system - Google Patents

Abstract

The invention discloses an adaptive modulation and coding method applicable to a mobile multicast system, and relates to a method for selecting a modulation and coding mode for each multicast group in the mobile multicast system. During each scheduling period of the system, a base station performs the adaptive modulation and coding after clustering traffic data, and then sends the data after passing through a power controller; a user decodes and demodulates the received information, outputs decoding and demodulation results in groups on one hand, and feeds back signal-interference-noise ratio interval serial numbers extracted from channel estimation results to the base station on the other hand, and the base station establishes a first-order Markov chain based on the signal-interference-noise ratio interval serial numbers fed back by each user in the first 20 scheduling periods to predict the signal-interference-noise ratio interval serial number of the user in the next scheduling period, and then the base station selects the modulation and coding mode for each multicast group according to a maximization principle of the system spectrum utilization rate. By predicting the interval serial numbers with the Markov, the method can greatly reduce the feedback cost, and can obtain higher system frequency spectrum utilization rate.

Description

Be adapted to the adaptive modulation coding method of mobile multicast system

Technical field

The present invention relates to a kind of adaptive modulation coding method that is applicable to mobile multicast system, belong to wireless communication field.

Technical background

Present mobile communication needs the higher transmission rate and the availability of frequency spectrum to realize the multimedia service transmission, link adaptation techniques can improve the message transmission rate and the availability of frequency spectrum effectively, be one of key technology of present and future mobile communication system, and Adaptive Modulation and Coding is a kind of main link adaptation techniques.In the descending ofdm system that transmits in the multicast mode, introduce the Adaptive Modulation and Coding technology and can guarantee under certain error rate condition the entire system performance to be greatly improved, this has wide practical use in the systems such as wireless sensor network at multimedia broadcast multicast.

The Adaptive Modulation and Coding scheme that proposes mainly concentrates on unicast system at present.List of references " Yang J; Tin N; Khandani A K.Adaptive modulation and coding in3G wireless systems; VTC 2002 Fall.Vancouver; 2002:544-548 " is selected modulation coding mode based on fixed threshold, and its threshold value and channel type are closely related; List of references " Choi J; Lee Y H.Improved AMC using adaptive SIR thresholds inOFDM based wireless systems; IEEE WCNC 2006; 1289-1292 " proposes the improvement Adaptive Modulation and Coding scheme in a kind of ofdm system, this scheme can be adjusted the Signal to Interference plus Noise Ratio thresholding adaptively, this compares with fixed threshold Adaptive Modulation and Coding scheme, and systematic function obtains very big raising, but the adjustment more complicated of Signal to Interference plus Noise Ratio thresholding; List of references " Feng H; Fan C; Lv T.A threshold optimizing method based on Markov in AMCcombined with HARQ; WiCOM 2006; 1-5 " proposes a kind of Adaptive Modulation and Coding scheme that is applicable to the communication system of band HARQ, this scheme is modeled as the single order Markov process with wireless channel, utilizes Markov process to determine the Signal to Interference plus Noise Ratio thresholding of every kind of modulation coding mode, and theory analysis and numerical simulation all obtain effect preferably; List of references " Diaz J; Bar-Ness Y; YeHoon Lee.A new approach to joint AMC and power allocation forMIMO-OFDM; Telecommunications 2006 International Conference on Internet and WebApplications and Services/Advanced International Conference on; 38-42 " proposes a kind of new adaptive modulation coding method, this method utilizes the imperfect information of channel feedback that channel is classified as different modulation types, and a kind of new power distribution method is proposed on this basis, receive effect preferably; Adaptive Modulation and Coding in list of references " Hwang C S; Kim Y.An adaptive modulation method for multicast communications of hierarchical data in wirelessnetworks.IEEE ICC; vol 2; 2002:896-900 " seminar's broadcast system, Adaptive Modulation and Coding is combined with the subcarrier packet allocation, analyzed the entire system throughput based on fixed threshold; Adaptive Modulation in list of references " SungK K; Chung G K.Throughput analysis of band AMC scheme in broadband wireless OFDMsystem.Wireless Communications and Networking Conference, 2006 " seminar's broadcast system; List of references " model morning; Chen Meiya; Su Lijun etc.; Adaptive Modulation and Coding system threshold adjustment algorithm research; Beijing University of Post ﹠ Telecommunication's journal. the 29th volume 4 phase .2006.8:49-53 " the AR forecast model is incorporated in the prediction of pilot tone, and analyzed the prediction of under different delay situation instantaneous value and the AR model prediction influence to systematic function, but above adaptive modulation coding method all requires the Signal to Interference plus Noise Ratio of feedback user, can consume a lot of backward channel capacities like this.

Summary of the invention

Technical problem: at the Adaptive Transmission problem in the mobile multicast system, the present invention proposes a kind of adaptive modulation coding method that is adapted to mobile multicast system of simple possible, and this method can effectively improve the system spectrum utilance.

Technical scheme: the present invention proposes a kind of adaptive modulation coding method that is adapted to mobile multicast system, and in the method, Adaptive Modulation and Coding is carried out after with traffic data packets in the base station, sends data through behind the power controller; The user deciphers demodulation with the information that receives, to decipher demodulation result grouping output on the one hand, carry out channel estimating on the other hand, and the feedback information that channel estimating is obtained feeds back to the base station by feedback channel, channel estimating is carried out according to feedback information in the base station, upgrades modulation coding mode in each dispatching cycle according to channel prediction result then.In the adaptive modulation coding method that the present invention carried, after the user carries out channel estimating, extract the interval sequence number of Signal to Interference plus Noise Ratio, and the interval sequence number that will extract feeds back to the base station, the Forecasting Methodology of interval sequence number is to set up Markov model prediction next of user's the interval sequence number of Signal to Interference plus Noise Ratio according to the interval sequence number of Signal to Interference plus Noise Ratio of mobile subscriber's preceding 20 dispatching cycles of feedback dispatching cycle, and definite method of multicast group modulation coding mode is to be each multicast group selection modulation coding mode according to multicast group availability of frequency spectrum maximization principle.For each multicast group, the performing step of this method is:

The first step: the user is according to Signal to Interference plus Noise Ratio value that channel estimating obtained, compare with the minimum Signal to Interference plus Noise Ratio value of every kind of modulation coding mode, thereby the interval sequence number of the Signal to Interference plus Noise Ratio that draws this user should the interval sequence number feed back to the base station by special-purpose based on feedback link then;

Second step: the base station is a forecasting sequence with the interval sequence number of Signal to Interference plus Noise Ratio of preceding 20 dispatching cycles of the feedback of each user in the multicast group, sets up the interval sequence number of Signal to Interference plus Noise Ratio that the Markov forecast model is predicted next of user dispatching cycle;

The 3rd step: the base station is target according to the interval sequence number predicted value of the Signal to Interference plus Noise Ratio of each user in the multicast group to the maximum with throughput of system and selects modulation coding mode for this multicast group.

In the above-mentioned adaptive modulation coding method that is adapted to mobile multicast system, in second step next dispatching cycle the user the prediction of the interval sequence number of Signal to Interference plus Noise Ratio undertaken by weighting Markov Forecasting Methodology, its concrete steps comprise:

1. with each user's forecasting sequence s ₁, s ₂..., s ₂₀Promptly the interval sequence number value of this user preceding 20 dispatching cycles of feedback is regarded one as with { 1,2, N} is the single order Markov chain of its state, N is the kind number of the modulation coding mode that system supported, statistics goes on foot the number of times of transferring to the j state from the i state through r, transfers to the transition probability of i state through the r step as the i state with the ratio of the total degree of i state appearance in this number of times statistical value and the forecasting sequence With transition probability

For element is formed r step state-transition matrix $P^{\left(r\right)}=\left(\begin{array}{cccc}{\mathrm{\π}}_{\mathrm{1,1}}^{\left(r\right)}& {\mathrm{\π}}_{\mathrm{1,2}}^{\left(r\right)}& \·\·\·& {\mathrm{\π}}_{1,N}^{\left(r\right)}\\ {\mathrm{\π}}_{\mathrm{2,1}}^{\left(r\right)}& {\mathrm{\π}}_{\mathrm{2,2}}^{\left(r\right)}& \·\·\·& {\mathrm{\π}}_{2,N}^{\left(r\right)}\\ \·\·\·& \·\·\·& \·\·\·& \·\·\·\\ {\mathrm{\π}}_{N,1}^{\left(r\right)}& {\mathrm{\π}}_{N,2}^{\left(r\right)}& \·\·\·& {\mathrm{\π}}_{N,N}^{\left(r\right)}\end{array}\right),$ I wherein, j ∈ 1,2 ..., N}, r=1,2 ..., 6;

2. the r of Markov chain step memory coefficient ω in asking 1. _r

3. according to the state of user transfer matrix, with interval sequence number predicted value $\hat{n}=\arg \left\{\underset{n\∈\{\mathrm{1,2},\·\·\·,N\}}{\max }\underset{r\′=1}{\overset{6}{\mathrm{\Σ}}}{\mathrm{\ω}}_{r\′}{\mathrm{\π}}_{s_{21-r\′},n}^{(r\′)}\right\}$ Predicted value as the interval sequence number of next this user's Signal to Interference plus Noise Ratio dispatching cycle;

In the above-mentioned adaptive modulation coding method that is adapted to mobile multicast system, the system of selection of multicast group modulation coding mode of the step of the 3rd in the implementation step may further comprise the steps:

1. the base station obtains the corresponding availability of frequency spectrum according to the predicted value of the interval sequence number of each user's Signal to Interference plus Noise Ratio in the multicast group;

The availability of frequency spectrum when 2. calculating each multicast group and adopting n kind modulation coding mode $R\left(n\right)=R_n\underset{i\′=n}{\overset{N}{\mathrm{\Σ}}}{m}_{i\′},$ R _nThe availability of frequency spectrum when selecting n kind modulation coding mode for each user, m _{I '}Be the number of users that is i ' of interval sequence number predicted value among all users of multicast group, $i\′\∈\{\mathrm{1,2},\·\·\·,N\};$

3. base station selected R (1), R (2) ..., the modulation coding mode of the maximum correspondence is as the modulation coding mode of the next one this multicast group dispatching cycle among the R (N).

In the above-mentioned adaptive modulation coding method that is adapted to mobile multicast system, calculate the r step memory coefficient ω of Markov chain _rMethod be:

1) calculates forecasting sequence s ₁, s ₂..., s ₂₀Average $\stackrel{\‾}{s}=\frac{1}{20}\underset{l=1}{\overset{20}{\mathrm{\Σ}}}{s}_l,$ Variance $\mathrm{\σ}=\sqrt{\frac{1}{19}\underset{l=1}{\overset{20}{\mathrm{\Σ}}}{(s_l-\stackrel{\‾}{s})}^2};$

2) calculate r step auto-correlation coefficient: ${\mathrm{\λ}}_r=\frac{\underset{l=1}{\overset{20-r}{\mathrm{\Σ}}}(s_l-\stackrel{\‾}{s})(s_{l+r}-\stackrel{\‾}{s})}{\underset{l=1}{\overset{20}{\mathrm{\Σ}}}{(s_l-\stackrel{\‾}{s})}^2};$

3) r is gone on foot the auto-correlation coefficient standardization, obtains remembering coefficient: ${\mathrm{\ω}}_r=\frac{\left|{\mathrm{\λ}}_r\right|}{\left|{\mathrm{\λ}}_1\right|+\left|{\mathrm{\λ}}_2\right|+\·\·\·+\left|{\mathrm{\λ}}_6\right|}.$

Beneficial effect: compared with prior art, the interval sequence number Forecasting Methodology of the Signal to Interference plus Noise Ratio that the present invention proposes step is simple and clear, be easy to realize, and only the interval sequence number of user's Signal to Interference plus Noise Ratio is fed back to the base station, do not need the Signal to Interference plus Noise Ratio value of user feedback channel, this has saved the channel capacity of feedback channel greatly; Use the Markov method and predict each user's Signal to Interference plus Noise Ratio wayside signaling, do not need oversize forecasting sequence just can receive satisfied result; Because being the multicast group in the system, employing availability of frequency spectrum maximization principle selects modulation coding mode, so system's employing adaptive modulation coding method that the present invention carried, can effectively improve the system spectrum utilance; In addition, the adaptive modulation coding method that the present invention carried is applicable to general unicast system, has wide application space.

Description of drawings

Fig. 1 is an Adaptive Modulation and Coding system configuration schematic diagram.

Fig. 2 is mobile subscriber's Signal to Interference plus Noise Ratio and availability of frequency spectrum graph of a relation.

Fig. 3 selects the flow chart of modulation coding mode for multicast group.

Fig. 4 is the interval sequence number prediction of a mobile subscriber's Signal to Interference plus Noise Ratio flow chart.

Embodiment

The present invention proposes a kind of adaptive modulation coding method that is adapted to mobile multicast system, and in the method, Adaptive Modulation and Coding is carried out after with traffic data packets in the base station, sends data through behind the power controller; The user deciphers demodulation with the information that receives, to decipher demodulation result grouping output on the one hand, carry out channel estimating on the other hand, and the feedback information that channel estimating is obtained feeds back to the base station by feedback channel, channel estimating is carried out according to feedback information in the base station, upgrades modulation coding mode in each dispatching cycle according to channel prediction result then.In the adaptive modulation coding method that the present invention carried, after the user carries out channel estimating, extract the interval sequence number of Signal to Interference plus Noise Ratio, and the interval sequence number that will extract feeds back to the base station, the Forecasting Methodology of interval sequence number is to set up Markov model prediction next of user's the interval sequence number of Signal to Interference plus Noise Ratio according to the interval sequence number of Signal to Interference plus Noise Ratio of mobile subscriber's preceding 20 dispatching cycles of feedback dispatching cycle, and definite method of multicast group modulation coding mode is to be each multicast group selection modulation coding mode according to multicast group availability of frequency spectrum maximization principle.For each multicast group, the performing step of this method is:

The first step: the user is according to Signal to Interference plus Noise Ratio value that channel estimating obtained, compare with the minimum Signal to Interference plus Noise Ratio value of every kind of modulation coding mode, thereby the interval sequence number of the Signal to Interference plus Noise Ratio that draws this user should the interval sequence number feed back to the base station by special-purpose based on feedback link then;

Second step: the base station is a forecasting sequence with the interval sequence number of Signal to Interference plus Noise Ratio of preceding 20 dispatching cycles of the feedback of each user in the multicast group, sets up the interval sequence number of Signal to Interference plus Noise Ratio that the Markov forecast model is predicted next of user dispatching cycle;

The 3rd step: the base station is target according to the interval sequence number predicted value of the Signal to Interference plus Noise Ratio of each user in the multicast group to the maximum with throughput of system and selects modulation coding mode for this multicast group.

In the above-mentioned adaptive modulation coding method that is adapted to mobile multicast system, in second step next dispatching cycle the user the prediction of the interval sequence number of Signal to Interference plus Noise Ratio undertaken by weighting Markov Forecasting Methodology, its concrete steps comprise:

1. with each user's forecasting sequence s ₁, s ₂..., s ₂₀Promptly the interval sequence number value of this user preceding 20 dispatching cycles of feedback is regarded one as with { 1,2, N} is the single order Markov chain of its state, N is the kind number of the modulation coding mode that system supported, statistics goes on foot the number of times of transferring to the j state from state through r, transfers to the transition probability of i state through the r step as the i state with the ratio of the total degree of i state appearance in this number of times statistical value and the forecasting sequence With transition probability

For element is formed r step state-transition matrix $P^{\left(r\right)}=\left(\begin{array}{cccc}{\mathrm{\π}}_{\mathrm{1,1}}^{\left(r\right)}& {\mathrm{\π}}_{\mathrm{1,2}}^{\left(r\right)}& \·\·\·& {\mathrm{\π}}_{1,N}^{\left(r\right)}\\ {\mathrm{\π}}_{\mathrm{2,1}}^{\left(r\right)}& {\mathrm{\π}}_{\mathrm{2,2}}^{\left(r\right)}& \·\·\·& {\mathrm{\π}}_{2,N}^{\left(r\right)}\\ \·\·\·& \·\·\·& \·\·\·& \·\·\·\\ {\mathrm{\π}}_{N,1}^{\left(r\right)}& {\mathrm{\π}}_{N,2}^{\left(r\right)}& \·\·\·& {\mathrm{\π}}_{N,N}^{\left(r\right)}\end{array}\right),$ I wherein, j ∈ 1,2 ..., N}, r=1,2 ..., 6;

2. the r of Markov chain step memory coefficient ω in asking 1. _r

3. according to the state of user transfer matrix, with interval sequence number predicted value $\hat{n}=\arg \left\{\underset{n\∈\{\mathrm{1,2},\·\·\·,N\}}{\max }\underset{r\′=1}{\overset{6}{\mathrm{\Σ}}}{\mathrm{\ω}}_{r\′}{\mathrm{\π}}_{s_{21-r\′},n}^{(r\′)}\right\}$ Predicted value as the interval sequence number of next this user's Signal to Interference plus Noise Ratio dispatching cycle;

In the above-mentioned adaptive modulation coding method that is adapted to mobile multicast system, the system of selection of multicast group modulation coding mode of the step of the 3rd in the implementation step may further comprise the steps:

1. the base station obtains the corresponding availability of frequency spectrum according to the predicted value of the interval sequence number of each user's Signal to Interference plus Noise Ratio in the multicast group;

The availability of frequency spectrum when 2. calculating each multicast group and adopting n kind modulation coding mode $R\left(n\right)=R_n\underset{i\′=n}{\overset{N}{\mathrm{\Σ}}}{m}_{i\′},$ R _nThe availability of frequency spectrum when selecting n kind modulation coding mode for each user, m _{I '}Be the number of users that is i ' of interval sequence number predicted value among all users of multicast group, i ' ∈ 1,2 ..., N};

3. base station selected R (1), R (2) ..., the modulation coding mode of the maximum correspondence is as the modulation coding mode of the next one this multicast group dispatching cycle among the R (N).

In the above-mentioned adaptive modulation coding method that is adapted to mobile multicast system, calculate the r step memory coefficient ω of Markov chain _rMethod be:

4) calculate forecasting sequence s ₁, s ₂..., s ₂₀Average $\stackrel{\‾}{s}=\frac{1}{20}\underset{l=1}{\overset{20}{\mathrm{\Σ}}}{s}_l,$ Variance $\mathrm{\σ}=\sqrt{\frac{1}{19}\underset{l=1}{\overset{20}{\mathrm{\Σ}}}{(s_l-\stackrel{\‾}{s})}^2};$

5) calculate r step auto-correlation coefficient: ${\mathrm{\λ}}_r=\frac{\underset{l=1}{\overset{20-r}{\mathrm{\Σ}}}(s_l-\stackrel{\‾}{s})(s_{l+r}-\stackrel{\‾}{s})}{\underset{l=1}{\overset{20}{\mathrm{\Σ}}}{(s_l-\stackrel{\‾}{s})}^2};$

6) r is gone on foot the auto-correlation coefficient standardization, obtains remembering coefficient: ${\mathrm{\ω}}_r=\frac{\left|{\mathrm{\λ}}_r\right|}{\left|{\mathrm{\λ}}_1\right|+\left|{\mathrm{\λ}}_2\right|+\·\·\·+\left|{\mathrm{\λ}}_6\right|}.$

With reference to the accompanying drawings, specific embodiments of the present invention is made explanation in more detail.

Consider the descending ofdm system that transmit in the multicast mode cellular mobile communication list sub-district, the non-real-time data business is transmitted to the user in the base station, suppose some mobile subscribers that distribute in the sub-district, user with identical services requirement forms multicast group, user in each multicast group uses same subchannel to receive the identical services that passes under the base station simultaneously, establishes channel conditions and remains unchanged in each Transmission Time Interval (TTI).The base station adopts Adaptive Modulation and Coding (AMC) technology in the renewal of each TTI according to channel condition information (CSI) the realization multicast group modulation coding mode (MCS) of subcarrier, to improve the system spectrum utilance.

The structure of Adaptive Modulation and Coding baseband transmission system as shown in Figure 1, Adaptive Modulation and Coding is carried out after with traffic data packets in the base station, sends data through behind the power controller; The user deciphers demodulation with the information that receives, carry out channel estimating then, and the result of channel estimating fed back to the base station by feedback channel, channel estimating is carried out according to feedback information in the base station, is that multicast group is upgraded MCS according to channel prediction result at each TTI then.Supposing the system is supported N MCS, is followed successively by MCS from low to high by speed ₁, MCS ₂..., MCS _N, corresponding message transmission rate is respectively R ₁, R ₂..., R _N, the minimum Signal to Interference plus Noise Ratio value of this N MCS correspondence is T ₁, T ₂..., T _N, the Signal to Interference plus Noise Ratio value of establishing all users all is not less than T ₁, use 1,2 respectively ..., N is to interval [T ₁, T ₂), [T ₂, T ₃) ..., [T _N, ∞) carry out label, as shown in Figure 2.At each TTI, the mobile subscriber in the sub-district uses the residing interval sequence number of self Signal to Interference plus Noise Ratio value | log ₂(N) |+1 bit information feeds back to the base station, wherein | and x| represents to be no more than | the maximum integer of x|; The base station is a forecasting sequence with the interval sequence number of Signal to Interference plus Noise Ratio of preceding 20 dispatching cycles of the feedback of each user in the multicast group, sets up the interval sequence number of Signal to Interference plus Noise Ratio that the Markov forecast model is predicted next of user dispatching cycle; Be target according to the interval sequence number predicted value of the Signal to Interference plus Noise Ratio of each user in the multicast group to the maximum with throughput of system then and select modulation coding mode for this multicast group.The base station be multicast group select modulation coding mode process as shown in Figure 3, wherein the interval sequence number forecasting process of Markov is as shown in Figure 4.

Relevant document is pointed out, can obtain enough accurate result with single order Markov model to Channel Modeling, so we considers the single order Markov process.The interval sequence number of the Signal to Interference plus Noise Ratio of travelling carriage has N value, and the process of transferring to another value from a value is at random, and has state transition probability, thus can set up one have N state 1,2 ..., the Markov chain of N}.The as if statistics data has the value of the interval sequence number of continuous 20 TTI Signal to Interference plus Noise Ratio, and value is respectively s ₁, s ₂..., s ₂₀∈ 1,2 ..., N}, definition Be that the i state goes on foot the probability of transferring to the j state through r, then

${\mathrm{\π}}_{i,j}^{\left(r\right)}\stackrel{\mathrm{\Δ}}{=}\frac{N_{i,j}^{\left(r\right)}}{N_i}---\left(1\right)$

Wherein

Be illustrated in state transitions sequence s ₁, s ₂..., s ₂₀The middle i of appearance state goes on foot the number of times of transferring to the j state, N through r _iBe illustrated in the number of times that the i state occurs in this state transitions sequence, and then the r of this Markov chain step state-transition matrix is

$P^{\left(r\right)}=\left(\begin{array}{cccc}{\mathrm{\π}}_{\mathrm{1,1}}^{\left(r\right)}& {\mathrm{\π}}_{\mathrm{1,2}}^{\left(r\right)}& \·\·\·& {\mathrm{\π}}_{1,N}^{\left(r\right)}\\ {\mathrm{\π}}_{\mathrm{2,1}}^{\left(r\right)}& {\mathrm{\π}}_{\mathrm{2,2}}^{\left(r\right)}& \·\·\·& {\mathrm{\π}}_{2,N}^{\left(r\right)}\\ \·\·\·& \·\·\·& \·\·\·& \·\·\·\\ {\mathrm{\π}}_{N,1}^{\left(r\right)}& {\mathrm{\π}}_{N,2}^{\left(r\right)}& \·\·\·& {\mathrm{\π}}_{N,N}^{\left(r\right)}\end{array}\right)---\left(2\right)$

The present invention uses 6 step Markov predictions, then through the interval sequence number of the next TTI Signal to Interference plus Noise Ratio of the mobile subscriber of Markov chain prediction is

$\hat{n}=\arg \left\{\underset{n\∈\{\mathrm{1,2},\·\·\·,N\}}{\max }\underset{r\′=1}{\overset{6}{\mathrm{\Σ}}}{\mathrm{\ω}}_{r\′}{\mathrm{\π}}_{s_{21-r\′},n}^{(r\′)}\right\}---\left(3\right)$

ω wherein ₁, ω ₂..., ω ₆Be the memory coefficient, closely related with the state transition probability of Markov chain.Usually, data were bigger to the contribution of predicted value than past data afterwards, so the present invention determines the memory coefficient with the correlation of forecasting sequence.Determine memory coefficient ω ₁, ω ₂..., ω ₆Step as follows:

(1) sequence of calculation s ₁, s ₂..., s ₂₀Average s and standard deviation sigma, average wherein $\stackrel{\‾}{s}=\frac{1}{20}\underset{l=1}{\overset{20}{\mathrm{\Σ}}}{s}_l,$ Variance $\mathrm{\σ}=\sqrt{\frac{1}{19}\underset{l=1}{\overset{20}{\mathrm{\Σ}}}{(s_l-\stackrel{\‾}{s})}^2}$

(2) calculate each rank auto-correlation coefficient λ _r, wherein

${\mathrm{\λ}}_r=\frac{\underset{i=1}{\overset{20-r}{\mathrm{\Σ}}}(s_i-\stackrel{\‾}{s})(s_{i+r}-\stackrel{\‾}{s})}{\underset{i=1}{\overset{20}{\mathrm{\Σ}}}{(s_i-\stackrel{\‾}{s})}^2},r=\mathrm{1,2},\·\·\·,6---\left(4\right)$

(3) to each rank auto-correlation coefficient λ _rStandardization obtains remembering coefficient

${\mathrm{\ω}}_r=\frac{\left|{\mathrm{\λ}}_r\right|}{\left|{\mathrm{\λ}}_1\right|+\left|{\mathrm{\λ}}_2\right|+\·\·\·+\left|{\mathrm{\λ}}_6\right|},r=\mathrm{1,2},\·\·\·,6---\left(5\right)$

The base station is selected MCS according to the interval sequence number of Signal to Interference plus Noise Ratio of all travelling carriage feedbacks in the multicast group for this multicast group.If certain multicast group has K user, this K user in the interval sequence number predicted value of the Signal to Interference plus Noise Ratio of certain TTI is respectively

${\hat{n}}_2,\·\·\·,{\hat{n}}_K\∈\{\mathrm{1,2},\·\·\·,N\},$ The value of the interval sequence number of this K Signal to Interference plus Noise Ratio is added up in the base station, and establishing the interval sequence number value of this K Signal to Interference plus Noise Ratio is 1,2 ..., the number of N is respectively m ₁, m ₂..., m _N, then obviously have following formula to set up

$\underset{i\&prime;=1}{\overset{N}{\mathrm{\&Sigma;}}}{m}_n=K,0\&le;m_1,{\mathrm{<figure-callout\; id="2"\; label="m"\; filenames="A200810236193E00131.png"\; state="\{\{state\}\}">m</figure-callout>}}_2,\&CenterDot;\&CenterDot;\&CenterDot;,m_N\&le;K---\left(6\right)$

At this moment, if select MCS _nAs the MCS of this multicast group downlink transfer, then this multicast group downstream spectrum utilance is

$R\left(n\right)=R_n\underset{i\&prime;=n}{\overset{N}{\mathrm{\&Sigma;}}}{m}_{i\&prime;}---\left(7\right)$

With this multicast group downstream spectrum utilance R (1), R (2) ..., R (N) compares, and selects the modulation coding mode of the modulation coding mode of the maximum correspondence as the next one this multicast group dispatching cycle, also is

$n^{*}=\arg \left\{\underset{\{\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,N\}}{\max }R\left(n\right)\right\}---\left(8\right).$

Claims (4)

1. be adapted to the adaptive modulation coding method of mobile multicast system, in the method, Adaptive Modulation and Coding is carried out after with traffic data packets in the base station, sends data through behind the power controller; The user deciphers demodulation with the information that receives, to decipher demodulation result grouping output on the one hand, carry out channel estimating on the other hand, and the feedback information that channel estimating is obtained feeds back to the base station by feedback channel, channel estimating is carried out according to feedback information in the base station, upgrades modulation coding mode in each dispatching cycle according to channel prediction result then.It is characterized in that, after the user carries out channel estimating, extract the interval sequence number of Signal to Interference plus Noise Ratio, and the interval sequence number that will extract feeds back to the base station, the Forecasting Methodology of interval sequence number is to set up Markov model prediction next of user's the interval sequence number of Signal to Interference plus Noise Ratio according to the interval sequence number of Signal to Interference plus Noise Ratio of mobile subscriber's preceding 20 dispatching cycles of feedback dispatching cycle, and definite method of multicast group modulation coding mode is to be each multicast group selection modulation coding mode according to multicast group availability of frequency spectrum maximization principle.For each multicast group, the performing step of this method is:

The first step: the user is according to Signal to Interference plus Noise Ratio value that channel estimating obtained, compare with the minimum Signal to Interference plus Noise Ratio value of every kind of modulation coding mode, thereby the interval sequence number of the Signal to Interference plus Noise Ratio that draws this user should the interval sequence number feed back to the base station by special-purpose based on feedback link then;

Second step: the base station is a forecasting sequence with the interval sequence number of Signal to Interference plus Noise Ratio of preceding 20 dispatching cycles of the feedback of each user in the multicast group, sets up the interval sequence number of Signal to Interference plus Noise Ratio that the Markov forecast model is predicted next of user dispatching cycle;

The 3rd step: the base station is target according to the interval sequence number predicted value of the Signal to Interference plus Noise Ratio of each user in the multicast group to the maximum with throughput of system and selects modulation coding mode for this multicast group.

2. the adaptive modulation coding method that is adapted to mobile multicast system as claimed in claim 1, it is characterized in that, in second step next dispatching cycle the user the prediction of the interval sequence number of Signal to Interference plus Noise Ratio undertaken by weighting Markov Forecasting Methodology, its concrete steps comprise:

1. with each user's forecasting sequence s ₁, s ₂..., s ₂₀Promptly the interval sequence number value of this user preceding 20 dispatching cycles of feedback is regarded one as with { 1,2, N} is the single order Markov chain of its state, N is the kind number of the modulation coding mode that system supported, statistics goes on foot the number of times of transferring to the j state from the i state through r, transfers to the transition probability of i state through the r step as the i state with the ratio of the total degree of i state appearance in this number of times statistical value and the forecasting sequence

With transition probability

For element is formed r step state-transition matrix $P^{\left(r\right)}=\left(\begin{array}{cccc}{\mathrm{\&pi;}}_{\mathrm{1,1}}^{\left(r\right)}& {\mathrm{\&pi;}}_{\mathrm{1,2}}^{\left(r\right)}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{\&pi;}}_{1,N}^{\left(r\right)}\\ {\mathrm{\&pi;}}_{\mathrm{2,1}}^{\left(r\right)}& {\mathrm{\&pi;}}_{\mathrm{2,2}}^{\left(r\right)}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{\&pi;}}_{2,N}^{\left(r\right)}\\ \&CenterDot;\&CenterDot;\&CenterDot;& \&CenterDot;\&CenterDot;\&CenterDot;& \&CenterDot;\&CenterDot;\&CenterDot;& \&CenterDot;\&CenterDot;\&CenterDot;\\ {\mathrm{\&pi;}}_{N,1}^{\left(r\right)}& {\mathrm{\&pi;}}_{N,2}^{\left(r\right)}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{\&pi;}}_{N,N}^{\left(r\right)}\end{array}\right),$ I wherein, j ∈ 1,2 ..., N}, r=1,2 ..., 6;

2. the r of Markov chain step memory coefficient ω in asking 1. _r

3. according to the state of user transfer matrix, with interval sequence number predicted value $\hat{n}=\arg \left\{\underset{n\&Element;\{\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,N\}}{\max }\underset{r\&prime;=1}{\overset{6}{\mathrm{\&Sigma;}}}{\mathrm{\&omega;}}_{r\&prime;}{\mathrm{\&pi;}}_{s_{21-r\&prime;,n}}^{(r\&prime;)}\right\}$ Predicted value as the interval sequence number of next this user's Signal to Interference plus Noise Ratio dispatching cycle.

3. the adaptive modulation coding method that is adapted to mobile multicast system as claimed in claim 1 is characterized in that, the system of selection of multicast group modulation coding mode of the step of the 3rd in the implementation step may further comprise the steps:

1. the base station obtains the corresponding availability of frequency spectrum according to the predicted value of the interval sequence number of each user's Signal to Interference plus Noise Ratio in the multicast group;

The availability of frequency spectrum when 2. calculating each multicast group and adopting n kind modulation coding mode $R\left(n\right)=R_n\underset{i\&prime;=n}{\overset{N}{\mathrm{\&Sigma;}}}{m}_{i\&prime;},$ R _nThe availability of frequency spectrum when selecting n kind modulation coding mode for each user, Be the number of users that is i ' of interval sequence number predicted value among all users of multicast group, i ' ∈ 1,2 ..., N};

3. base station selected R (1), R (2) ..., the modulation coding mode of the maximum correspondence is as the modulation coding mode of the next one this multicast group dispatching cycle among the R (N).

4. the adaptive modulation coding method that is adapted to mobile multicast system as claimed in claim 2 is characterized in that, calculates the r step memory coefficient ω of Markov chain _rMethod be:

1) calculates forecasting sequence s ₁, s ₂..., s ₂₀Average $\stackrel{\&OverBar;}{s}=\frac{1}{20}\underset{l=1}{\overset{20}{\mathrm{\&Sigma;}}}{s}_l$ , variance $\mathrm{\&sigma;}=\sqrt{\frac{1}{19}\underset{l=1}{\overset{20}{\mathrm{\&Sigma;}}}{(s_l-\stackrel{\&OverBar;}{s})}^2};$

2) calculate r step auto-correlation coefficient: ${\mathrm{\&lambda;}}_r=\frac{\underset{l=1}{\overset{20-r}{\mathrm{\&Sigma;}}}(s_l-\stackrel{\&OverBar;}{s})(s_{l+r}-\stackrel{\&OverBar;}{s})}{\underset{l=1}{\overset{20}{\mathrm{\&Sigma;}}}{(s_l-\stackrel{\&OverBar;}{s})}^2};$

3) r is gone on foot the auto-correlation coefficient standardization, obtains remembering coefficient: ${\mathrm{\&omega;}}_r=\frac{\left|{\mathrm{\&lambda;}}_r\right|}{\left|{\mathrm{\&lambda;}}_1\right|+\left|{\mathrm{\&lambda;}}_2\right|+\&CenterDot;\&CenterDot;\&CenterDot;+\left|{\mathrm{\&lambda;}}_6\right|}.$

Images