CN101958765A - Channel quality indication predicting and compensating method and system - Google Patents
Channel quality indication predicting and compensating method and system Download PDFInfo
- Publication number
- CN101958765A CN101958765A CN2010102932801A CN201010293280A CN101958765A CN 101958765 A CN101958765 A CN 101958765A CN 2010102932801 A CN2010102932801 A CN 2010102932801A CN 201010293280 A CN201010293280 A CN 201010293280A CN 101958765 A CN101958765 A CN 101958765A
- Authority
- CN
- China
- Prior art keywords
- prediction
- cqi
- autoregressive
- moving average
- sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 71
- 230000008569 process Effects 0.000 claims abstract description 27
- 238000005311 autocorrelation function Methods 0.000 claims description 83
- 238000003745 diagnosis Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000005070 sampling Methods 0.000 claims description 10
- 238000002360 preparation method Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 2
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 claims 3
- 238000004891 communication Methods 0.000 abstract description 6
- 238000004422 calculation algorithm Methods 0.000 abstract description 4
- 230000005540 biological transmission Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000004088 simulation Methods 0.000 description 7
- 230000006978 adaptation Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000010267 cellular communication Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005562 fading Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a channel quality indication predicting and compensating method and a channel quality indication predicting and compensating system. The method comprises the following steps of: accumulating CQI values of the same subband or full bandwidth at same intervals to form a CQI time sequence; identifying an autoregressive summarization moving average model, namely according to the CQI time sequence, identifying the process of the autoregressive summarization moving average model and determining parameters in the autoregressive summarization moving average model; estimating the parameters of the autoregressive summarization moving average model to determine the autoregressive summarization moving average model; and predicting the CQI values based on the determined autoregressive summarization moving average model. The method can effectively reduce the difference of the original CQI values and a true value at the current moment so as to save communication resources and energy and improve the system communication capacity; and the method also has the advantages of low complexity of algorithms and convenient implementation.
Description
Technical Field
The present invention relates to the field of mobile communications technologies, and in particular, to a method and a system for predicting and compensating channel quality indicator.
Background
In a cellular communication system, the quality of the signal received by a user depends on the channel quality of the cell, the noise level and the interference level of the neighbouring cells. Under a certain transmission power, in order to optimize the capacity and coverage of the system, the data transmission rate of the transmitter should be able to match the signal quality received by the UE, i.e. more data is transmitted and less data is transmitted in contrast when the signal quality received by the UE is good. This is because the low-order modulation coding method can resist higher interference and noise, but the transmission rate is low; and the high-order modulation coding method has high transmission rate but is sensitive to interference and noise. This technique is known as a link adaptation technique based on adaptive modulation and coding.
Referring to fig. 1, in an LTE communication system, an eNodeB selects an appropriate modulation scheme and coding rate according to a CQI (Channel Quality Indicator) fed back by a UE, so as to improve a transmission rate and throughput of the system. The CQI is an indication of a supportable data rate given in consideration of the signal-to-noise level of the current channel and the performance of the receiver. Although the LTE standard designs the necessary signaling for the inter-operation between eNodeB and UE regarding CQI, the problem of how eNodeB specifically selects the appropriate modulation scheme and coding rate is not described in the standard, and the problem of how eNodeB maximizes the system capacity by using the received CQI is left to the equipment manufacturer. That is, the CQI is actually a "recommendation" for adaptive modulation and coding operations performed by the eNodeB, and the eNodeB may select a corresponding modulation and coding scheme according to the CQI value fed back by the UE, or may select the modulation and coding scheme according to comprehensive consideration of various situations.
It should be noted that, due to the uplink transmission delay of the CQI and the processing delay of the eNodeB on the CQI, there is a difference between the CQI used by the eNodeB in performing link adaptation and scheduling and the actual CQI at the current time. For fast fading channels or situations where the signal-to-noise ratio at the receiving end changes faster, the UE is likely to face completely different channel conditions than the previous reporting time. This would result in a waste of power or frequency resources for the case of a good channel, while it is likely to trigger a HARQ process for the case of a bad channel, resulting in a waste of time resources. Therefore, it is necessary to predict and compensate the CQI value at the acting time based on the received CQI value.
Disclosure of Invention
The invention aims to provide a method and a system for predicting and compensating channel quality indication, so that a compensated CQI value is closer to a true value of the current moment under a certain condition, and the capacity of the system is further improved.
In one aspect, the invention discloses a method for predicting and compensating channel quality indication, comprising the following steps: a data preparation step, in which CQI values of the same sub-band or the same time interval of the full bandwidth are accumulated to form a CQI time sequence; identifying an autoregressive summation moving average model, namely identifying the process of the autoregressive summation moving average model according to the CQI time sequence and determining parameters in the autoregressive summation moving average model; a parameter estimation step, namely estimating parameters of the autoregressive summation moving average model to determine the autoregressive summation moving average model; and a predicting and compensating step, namely predicting and compensating the CQI value based on the determined autoregressive summation moving average model.
In the above prediction and compensation method, it is preferable that: a model diagnosis step of performing diagnosis check on the autoregressive summation moving average model determined in the parameter estimation step, determining whether the model is appropriate, and if so, executing the prediction and compensation step; if not, the model is discarded.
In the above prediction and compensation method, preferably, in the prediction and compensation step, the CQI value is predicted by: a predicted time calculation step, namely determining CQI feedback and processing time delay and time intervals among sampling points of CQI; calculating a value obtained by dividing the CQI feedback and the processing time delay by the time interval between each sampling point of the CQI, and taking the value as predicted time; a compensation step, namely determining a prediction step length and judging whether the prediction time is an integer: if so, taking the prediction time as a prediction step length, and taking a result obtained after prediction as a compensation result of the CQI at the current moment; if not, the prediction step length determines the upper rounding and the lower rounding of the prediction time, two times of prediction are respectively carried out, and the prediction result obtained according to the upper rounding prediction step length is recorded as X1And recording the prediction result obtained by taking down the whole prediction step length as X2And performing linear interpolation on the two prediction results to finally obtain a compensation result of the CQI at the current moment: x1+aX2Where a is the fractional part of the prediction time.
In the above prediction and compensation method, preferably, in the identifying step of the autoregressive-sum moving average model, the autoregressive-sum moving average model is ARIMA (p, d, q), and p, d, q are determined as follows: calculating the estimated values of the autocorrelation function and the partial autocorrelation function of the CQI time sequence, the estimated values of the autocorrelation function and the partial autocorrelation function of the sequence after first-order difference of the CQI time sequence, and the estimated values of the autocorrelation function and the partial autocorrelation function of the sequence after second-order difference of the CQI time sequence; judging the attenuation properties of the autocorrelation function of the CQI time sequence, the autocorrelation function of the sequence after the first-order difference of the CQI time sequence and the autocorrelation function of the sequence after the second-order difference of the CQI time sequence, and determining d, wherein d is a difference order required by an ARIMA (p, d, q) model to reach stability, and the value of d is 0, 1 or 2; and determining the properties of the autocorrelation function and the partial autocorrelation function of the d-order differential sequence of the CQI time sequence so as to determine the orders p and q of the autoregressive and moving average operator.
In the prediction and compensation method, preferably, in the parameter estimation step, the parameter estimation is performed according to the first p + q +1 autocovariance of the sequence after the d-order difference of the CQI time sequence.
In the prediction and compensation method, preferably, in the model diagnosis step, a model is diagnosed and tested by using a residual error of a sequence after d-order difference of the CQI time series; if the model is appropriate, the residual error becomes closer to a white noise sequence as the length of the sequence increases.
On the other hand, the invention also discloses a system for predicting and compensating the channel quality indication, which comprises the following steps: the device comprises a data preparation module, an identification module of an autoregressive sum moving average model, a parameter estimation module and a prediction and compensation module. The data preparation module is used for accumulating CQI values of the same sub-band or the same time interval of the full bandwidth to form a CQI time sequence; the identification module of the autoregressive summation moving average model is used for identifying the process of the autoregressive summation moving average model according to the CQI time sequence and determining parameters in the autoregressive summation moving average model; the parameter estimation module is used for estimating parameters of the autoregressive summation moving average model so as to determine the autoregressive summation moving average model; and the prediction and compensation module is used for predicting and compensating the CQI value based on the determined autoregressive summation moving average model.
In the above prediction and compensation system, preferably, a parameter estimation module and a prediction and compensation module are further provided with: a model diagnosis module for performing diagnosis test on the autoregressive summation moving average model determined in the parameter estimation step, determining whether the model is suitable, and if so, executing the prediction and compensation step; if not, the model is discarded.
In the above prediction and compensation system, preferably, the prediction and compensation module further includes: prediction time meterThe computing unit is used for determining the time intervals between the CQI feedback and processing time delay and each sampling point of the CQI; calculating a value obtained by dividing the CQI feedback and the processing time delay by the time interval between each sampling point of the CQI, and taking the value as predicted time; the compensation unit is used for determining a prediction step length and judging whether the prediction time is an integer: if so, taking the prediction time as a prediction step length, and taking a result obtained after prediction as a compensation result of the CQI at the current moment; if not, the prediction step length determines the upper rounding and the lower rounding of the prediction time, two times of prediction are respectively carried out, and the prediction result obtained according to the upper rounding prediction step length is recorded as X1And recording the prediction result obtained by taking down the whole prediction step length as X2And performing linear interpolation on the two prediction results to finally obtain a compensation result of the CQI at the current moment: x1+aX2Where a is the fractional part of the prediction time.
In the above prediction and compensation system, preferably, the identification module of the autoregressive-sum moving average model further includes: a calculation unit and a determination unit. The calculation unit is used for calculating the estimated values of the autocorrelation function and the partial autocorrelation function of the CQI time sequence, the estimated values of the autocorrelation function and the partial autocorrelation function of the CQI time sequence after first-order difference, and the estimated values of the autocorrelation function and the partial autocorrelation function of the CQI time sequence after second-order difference; the determining unit is used for judging the attenuation properties of the autocorrelation function of the CQI time sequence, the autocorrelation function of the sequence after the first-order difference of the CQI time sequence and the autocorrelation function of the sequence after the second-order difference of the CQI time sequence, and determining d, wherein d is a difference order required by the ARIMA (p, d, q) model to reach stability, and the value of d is 0, 1 or 2; and determining the properties of the autocorrelation function and the partial autocorrelation function of the d-order differential sequence of the CQI time sequence so as to determine the orders p and q of the autoregressive and moving average operator.
In the prediction and compensation system, preferably, in the parameter estimation module, the parameter estimation is performed according to the first p + q +1 autocovariance of the sequence after the d-order difference of the CQI time sequence.
In the prediction and compensation system, preferably, the model diagnosis module performs diagnosis and inspection on the model by using a residual error of a sequence after d-order difference of the CQI time series; if the model is appropriate, the residual error becomes closer to a white noise sequence as the length of the sequence increases.
Compared with the prior art, the method has the following advantages:
the invention utilizes the ARIMA process to model according to the previous CQI sequence value to predict the possible change of the CQI in the feedback and processing time delay, further obtains the compensation result of the CQI directly or through interpolation, and can effectively reduce the difference between the original CQI value and the real value at the current moment, thereby saving communication resources and energy, improving the communication capacity of the system, and having the advantages of low algorithm complexity and convenient realization.
Drawings
Fig. 1 is an illustration of feedback and processing delay for CQI in the prior art;
FIG. 2 is a flowchart illustrating steps of a method for predicting and compensating for CQI in accordance with an embodiment of the present invention;
FIG. 3a is a diagram illustrating an autocorrelation function of a CQI sequence in an embodiment of a method for predicting and compensating a channel quality indicator according to the present invention;
FIG. 3b is a diagram illustrating a partial autocorrelation function of a CQI sequence in an embodiment of a method for predicting and compensating a channel quality indicator according to the present invention;
FIG. 4 is a graph comparing system throughput after using the CQI prediction and compensation method provided by the present invention on an LTE simulation platform with system throughput when not using the CQI prediction and compensation method;
FIG. 5 is a block diagram of an embodiment of a system for predicting and compensating for channel quality indication according to the present invention;
FIG. 6 is a block diagram of a compensation module in an embodiment of the system for predicting and compensating for channel quality indicator according to the present invention;
fig. 7 is a schematic structural diagram of an identification module of a regression-sum moving average model in an embodiment of the system for predicting and compensating channel quality indicator according to the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
The invention provides a method for predicting and compensating CQI fed back by UE (user equipment) by an eNodeB (evolved Integrated Moving Average) in an LTE (long term evolution) communication system and an existing algorithm for modeling and predicting by using an Autoregressive summation Moving Average Model, and particularly relates to a method for predicting and compensating CQI by using an Autoregressive summation Moving Average Model.
The ARIMA model is a model established by converting a non-stationary time series into a stationary time series through a difference step (this step can be omitted), and then regressing the hysteresis value of the dependent variable and the present value and the hysteresis value of the random error term. The ARIMA model is generally represented as an ARIMA (p, d, q) model, where p, d and q are integers greater than or equal to 0, representing the order of the auto-regressive, differential and moving average portions of the model, respectively. The ARIMA Model includes a Moving Average Model (MA), an Autoregressive Model (AR), an Autoregressive Moving Average Model (ARMA), and an ARIMA Model, depending on whether the original sequence is stable or not and the difference of the portions included in the regression. ARIMA models are commonly used for analysis and prediction of time series.
The basic theory of the autoregressive summation moving average process (ARIMA) used in the present invention is illustrated below:
an autoregressive model:
in the autoregressive model, the current value of the process is represented as a finite linear combination of the past values of the process plus an impact at. Recording the process value of z at equal time intervals t, t-1, t-2, …t，zt-1，zt-2…. And use
Noting the deviation from the mean μ, i.e.
Then
It is called an Autoregressive (AR) process of order p.
Sliding average model:
if it is not
Linearly dependent on a finite past value of q a, i.e.
Referred to as a q-order Moving Average (MA) process.
Autoregressive moving average model:
if autoregressive and moving average are incorporated into a model, this is the autoregressive moving average model (ARMA)
The autoregressive moving average model comprises p + q +2 parametersμ；φ1，…，φp；θ1，…θq；
Autoregressive sum moving average model:
the above three models are mainly directed to stationary models, and in practical applications, many sequences show non-stationary characteristics, but they are stationary sequences by d-order differences, and in practice, d is usually 0, 1 or at most 2. Thus, the autoregressive sum moving average process (ARIMA) of order (p, d, q) is defined as:
wt＝φ1wt-1+…+φpwt-p+at-θ1at-1-…-θqat-q
wherein,
represents a pair ztStep d of (1). If z is used when d is equal to 0tμ instead of wtThen the model becomes the autoregressive moving average model.
Method embodiment
Referring to fig. 2, fig. 2 is a flowchart illustrating steps of an embodiment of a method for predicting and compensating for channel quality indication according to the present invention, including the following steps:
a data preparation step 210, accumulating the CQI values of the same sub-band or the same time interval of the full bandwidth to form a CQI time sequence; an identification step 220 of the autoregressive summation moving average model, namely identifying the process of the autoregressive summation moving average model according to the CQI time sequence and determining parameters in the autoregressive summation moving average model; a parameter estimation step 230 of estimating parameters of the autoregressive-sum moving average model to determine the autoregressive-sum moving average model; a prediction and compensation step 240, predicting the CQI value based on the determined autoregressive-sum moving average model.
Further, between the parameter estimation step 230 and the prediction and compensation step 240, a model diagnosis step is further provided, which performs a diagnosis check on the autoregressive summation moving average model determined in the parameter estimation step, determines whether the model is suitable, and if so, performs the prediction and compensation step; if not, the model is discarded.
The above steps are clearly illustrated by an example.
(1) Accumulating CQI data with the same bandwidth and the same measuring time interval to form a CQI time sequence, when the length of the sequence is less than a certain value, not processing, when the length of the sequence is more than the value, using the algorithm, namely entering the next step.
(2) Determining p, d, q:
and calculating the estimation values of the autocorrelation function and the partial autocorrelation function of the CQI time sequence and the estimation values of the autocorrelation function and the partial autocorrelation function after first-order difference and second-order difference of the CQI sequence. If the estimated autocorrelation function is not decaying quickly, the sequence is considered unbalanced and a difference is needed, otherwise it is considered stationary. D in ARIMA (p, d, q) is set to the differential order required for achieving the plateau. The value of d is generally 0, 1 and 2, and correspondingly, only the first 20 values of the autocorrelation function estimated value of the original sequence, the first order or the second order difference sequence are needed to be examined.
After the d value is determined, the order p and q of the autoregressive and moving average operators are determined by studying the general properties of the autocorrelation function and the partial autocorrelation function of the sequence after d-order difference. Specifically, the following principle is followed: the autocorrelation of the p-order autoregressive process is tail-tailing, while its partial autocorrelation function is tail-truncated after a lag p-step; the autocorrelation function of the q-order moving average process is truncated after a lag q-step, while its partial autocorrelation function is trailing; a mixing process comprising p-order autoregressive and q-order moving averages, whose autocorrelation functions are the exponential and/or decaying sinusoids of the mixture, and whose partial autocorrelation functions are the exponential and/or decaying sinusoids after p-q steps, both of which are smeared.
(3) Parameters in the ARIMA model are estimated: in this embodiment, the parameter estimation is based on the first p + q +1 autocovariance cj (j ═ 0, 1, …, (p + q)) of the time series after the d-order difference, and is obtained by the following steps:
the autoregressive parameters can be estimated by the following p linear equations:
note zt′＝φ(B)ztAnd the process is treated as the following moving average process:
z′t＝θ(B)at
z′tof autocovariance c'jCan be formed by ztAutocovariance c ofjIs expressed by that, for j equal to 0, 1, …, q has
Wherein phi is0＝-1。
Utilizing the autocovariance estimated value c 'obtained in the previous step'jBy the following iterative procedure, an estimate of the moving average parameter can be obtained,
in which the theta is agreed00, and θ is set at the start of the iteration1，θ2，…，θqEqual to zero. Each time obtaining thetajAndthe latest value of (a) is substituted into the next calculation until the result is stable.
(4) The model was tested diagnostically using the residuals of the sequence. When the length of the sequence increases, the residual error, if the model is appropriate
It is getting closer to the white noise sequence.
(5) And (3) predicting and compensating the CQI by using the model determined by the steps: if the value of the prediction time is an integer, the prediction step length is directly taken, and the result obtained after prediction is used as the compensation result of the CQI at the current moment; if the value of the prediction time is not an integer, the prediction step length is respectively rounded up and rounded down, the prediction is respectively carried out twice, and the prediction result obtained according to the rounded up prediction step length is recorded as X1And recording the prediction result obtained by taking down the whole prediction step length as X2And performing linear interpolation on the two prediction results to finally obtain a compensation result of the CQI at the current moment: x1+aX2Where a is the fractional part of the prediction time.
It should be noted that, in the above process, the steps 1) to 4) are all performed offline, and the step 6) is a specific implementation on a product or a simulation platform.
In an embodiment of the present invention, the method provided by the present invention is simulated on an LTE link simulation platform. The simulation conditions were as follows: bandwidth: 1.4MHz, carrier frequency: 2.1MHz, CQI uplink transmission and processing delay: 6ms, CQI feedback time interval: 2ms, channel model: VehA, UE speed: 30km/h, number of simulation frames: 8000.
fig. 3 is a diagram of an estimation of an autocorrelation function and a partial autocorrelation function of a CQI sequence for a certain subband in a 1.4M simulated bandwidth, which can be determined to be approximately stationary by observation, with the partial autocorrelation function being cut off after a two-step lag, and the autocorrelation function being smeared. Therefore, consider modeling the sequence as an ARIMA (2, 0, 0) sequence.
And estimating the parameters by using a least square method when the sequence is added with one CQI value since the length of the CQI sequence is more than 100. And then, performing three-step prediction by using the model, and performing link adaptation and scheduling by using a prediction result as a compensated CQI value. Fig. 4 is a comparison graph of system throughput after using the CQI prediction and compensation method provided by the present invention on an LTE simulation platform and system throughput when not using the CQI prediction and compensation method, where a curve a represents a variation trend of system throughput after using the method provided by the present invention to compensate CQI, and a curve b represents a variation trend of system throughput when not using CQI compensation. As shown in fig. 4, the improvement of the simulation result of the system throughput is obtained after the CQI is compensated by using the method provided by the present invention.
System embodiment
Referring to fig. 5, the present invention further provides a schematic structural diagram of an embodiment of a system for predicting and compensating channel quality indication, including:
a data preparation module 51, configured to accumulate CQI values of the same subband or the same full bandwidth and the same time interval to form a CQI time sequence; the identification module 52 of the autoregressive summation moving average model is used for identifying the process of the autoregressive summation moving average model according to the CQI time sequence and determining parameters in the autoregressive summation moving average model; a parameter estimation module 53, configured to estimate parameters of the auto-regressive summation moving average model to determine the auto-regressive summation moving average model; and a prediction and compensation module 54 for predicting and compensating the CQI value based on the determined autoregressive-sum moving average model.
In one, the parameter estimation module and the prediction and compensation module further comprise a model diagnosis module, which performs diagnosis check on the autoregressive summation moving average model determined in the parameter estimation step, determines whether the model is suitable, and if so, executes the prediction and compensation step; if not, the model is discarded. In one embodiment, a model is subjected to diagnostic test by using the residual error of the CQI time series; if the model is appropriate, the residual error becomes closer to a white noise sequence as the length of the sequence increases.
Referring to fig. 6, in another embodiment, the compensation module 54 includes: prediction time calculation unit 61 and compensation unit 62:
the predicted time calculation unit 61 is used for determining the time intervals between the sampling points of the CQI and the processing time delay of the CQI feedback; calculating a value obtained by dividing the CQI feedback and the processing time delay by the time interval between each sampling point of the CQI, and taking the value as predicted time; the compensation unit 62 is configured to determine a prediction step size, and determine whether the prediction time is an integer: if so, taking the prediction time as a prediction step length, and taking a result obtained after prediction as a compensation result of the CQI at the current moment; if not, the prediction step length determines the upper rounding and the lower rounding of the prediction time, two times of prediction are respectively carried out, and the prediction result obtained according to the upper rounding prediction step length is recorded as X1And recording the prediction result obtained by taking down the whole prediction step length as X2And performing linear interpolation on the two prediction results to finally obtain a compensation result of the CQI at the current moment: x1+aX2Where a is the fractional part of the prediction time。
Referring to fig. 7, in one embodiment, the identification module 52 of the autoregressive-sum moving average model includes a calculation unit 71 and a determination unit 72. The calculating unit 71 is configured to calculate estimated values of an autocorrelation function and a partial autocorrelation function of the CQI time series, and estimated values of an autocorrelation function and a partial autocorrelation function of the CQI time series after first-order difference and second-order difference; the determining unit 72 is configured to determine an attenuation property of the estimated autocorrelation function, where d is a difference order required for an ARIMA (p, d, q) model to reach a stationary state; and determining the properties of the autocorrelation function and the partial autocorrelation function of the d-order differential sequence to determine the orders p and q of the autoregressive and moving average operator.
Specifically, the estimated values of the autocorrelation function and the partial autocorrelation function of the time series of the CQI and the estimated values of the autocorrelation function and the partial autocorrelation function of the first order difference and the second order difference of the time series of the CQI are calculated. If the estimated autocorrelation function is not decaying quickly, the sequence is considered unbalanced and a difference is needed, otherwise it is considered stationary. D in ARIMA (p, d, q) is set to the differential order required for achieving the plateau. The value of d is generally 0, 1 and 2, and correspondingly, only the first 20 values of the autocorrelation function estimated value of the original sequence, the first order or the second order difference sequence are needed to be examined.
After the d value is determined, the order p and q of the autoregressive and moving average operators are determined by studying the general properties of the autocorrelation function and the partial autocorrelation function of the sequence after d-order difference. Specifically, the following principle is followed: the autocorrelation of the p-order autoregressive process is tail-tailing, while its partial autocorrelation function is tail-truncated after a lag p-step; the autocorrelation function of the q-order moving average process is truncated after a lag q-step, while its partial autocorrelation function is trailing; a mixing process comprising p-order autoregressive and q-order moving averages, whose autocorrelation functions are the exponential and/or decaying sinusoids of the mixture, and whose partial autocorrelation functions are the exponential and/or decaying sinusoids after p-q steps, both of which are smeared.
In one embodiment, the parameter estimation module is based on the first p + q +1 autocovariance cj (j ═ 0, 1, …, (p + q)) of the time-series d-order differenced sequence, and is obtained by the following steps:
the autoregressive parameters can be estimated by the following p linear equations:
note zt′＝φ(B)ztAnd the process is treated as the following moving average process:
z′t＝θ(B)at
z′tof autocovariance c'jCan be formed by ztAutocovariance c ofjIs expressed by that, for j equal to 0, 1, …, q has
Wherein phi is0＝-1。
Utilizing the autocovariance estimated value c 'obtained in the previous step'jBy the following iterative procedure, an estimate of the moving average parameter can be obtained,
in which the theta is agreed00, and θ is set at the start of the iteration1，θ2，…，θqEqual to zero. Each time obtaining thetajAnd
the latest value of (a) is substituted into the next calculation until the result is stable.
The present invention provides a method and a system for predicting and compensating channel quality indicator, and a specific embodiment is applied in the present disclosure to explain the principle and the implementation of the present invention, and the description of the above embodiment is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (12)
1. A method for predicting and compensating Channel Quality Indication (CQI), comprising the steps of:
a data preparation step, in which CQI values of the same sub-band or the same time interval of the full bandwidth are accumulated to form a CQI time sequence;
identifying an autoregressive summation moving average model, namely identifying the process of the autoregressive summation moving average model according to the CQI time sequence and determining parameters in the autoregressive summation moving average model;
a parameter estimation step, namely estimating parameters of the autoregressive summation moving average model to determine the autoregressive summation moving average model;
and a predicting and compensating step, namely predicting and compensating the CQI value based on the determined autoregressive summation moving average model.
2. The prediction and compensation method of claim 1, wherein between the parameter estimation step and the prediction and compensation step, there are further provided:
a model diagnosis step of performing diagnosis check on the autoregressive summation moving average model determined in the parameter estimation step, determining whether the model is appropriate, and if so, executing the prediction and compensation step; if not, the model is discarded.
3. The prediction and compensation method of claim 1 or 2, wherein in the prediction and compensation step, the CQI value is predicted by:
a predicted time calculation step, namely determining CQI feedback and processing time delay and time intervals among sampling points of CQI; calculating a value obtained by dividing the CQI feedback and the processing time delay by the time interval between each sampling point of the CQI, and taking the value as predicted time;
a compensation step, namely determining a prediction step length and judging whether the prediction time is an integer:
if so, taking the prediction time as a prediction step length, and taking a result obtained after prediction as a compensation result of the CQI at the current moment;
if not, the prediction step length determines the upper rounding and the lower rounding of the prediction time, two times of prediction are respectively carried out, and the prediction result obtained according to the upper rounding prediction step length is recorded as X1And recording the prediction result obtained by taking down the whole prediction step length as X2And performing linear interpolation on the two prediction results to finally obtain a compensation result of the CQI at the current moment: x1+aX2Where a is the fractional part of the prediction time.
4. The prediction and compensation method of claim 3, wherein in the step of identifying the autoregressive-summed moving average model, the autoregressive-summed moving average model is ARIMA (p, d, q), and p, d, q are determined by:
calculating the estimated values of the autocorrelation function and the partial autocorrelation function of the CQI time sequence, the estimated values of the autocorrelation function and the partial autocorrelation function of the sequence after first-order difference of the CQI time sequence, and the estimated values of the autocorrelation function and the partial autocorrelation function of the sequence after second-order difference of the CQI time sequence;
judging the attenuation properties of the autocorrelation function of the CQI time sequence, the autocorrelation function of the sequence after the first-order difference of the CQI time sequence and the autocorrelation function of the sequence after the second-order difference of the CQI time sequence, and determining d, wherein d is a difference order required by an ARIMA (p, d, q) model to reach stability, and the value of d is 0, 1 or 2;
and determining the properties of the autocorrelation function and the partial autocorrelation function of the d-order differential sequence of the CQI time sequence so as to determine the orders p and q of the autoregressive and moving average operator.
5. The prediction and compensation method of claim 4, wherein in the parameter estimation step, parameter estimation is performed according to the first p + q +1 autocovariance of the sequence after d-order difference of the CQI time sequence.
6. The prediction and compensation method of claim 5, wherein in the model diagnosis step, the model is diagnosed and checked by using a residual error of a sequence after d-order difference of the CQI time series; if the model is appropriate, the residual error becomes closer to a white noise sequence as the length of the sequence increases.
7. A system for predicting and compensating for channel quality indication, comprising:
a data preparation module for accumulating CQI values of the same sub-band or the same time interval of the full bandwidth to form a CQI time sequence;
the identification module of the autoregressive summation moving average model is used for identifying the process of the autoregressive summation moving average model according to the CQI time sequence and determining parameters in the autoregressive summation moving average model;
the parameter estimation module is used for estimating parameters of the autoregressive summation moving average model so as to determine the autoregressive summation moving average model;
and the prediction and compensation module is used for predicting and compensating the CQI value based on the determined autoregressive summation moving average model.
8. The prediction and compensation system of claim 7, further comprising, between the parameter estimation module and the prediction and compensation module:
a model diagnosis module for performing diagnosis test on the autoregressive summation moving average model determined in the parameter estimation step, determining whether the model is suitable, and if so, executing the prediction and compensation step; if not, the model is discarded.
9. The prediction and compensation system of claim 7 or 8, wherein the prediction and compensation module further comprises:
the prediction time calculation unit is used for determining CQI feedback and processing time delay and time intervals among sampling points of CQI; calculating a value obtained by dividing the CQI feedback and the processing time delay by the time interval between each sampling point of the CQI, and taking the value as predicted time;
the compensation unit is used for determining a prediction step length and judging whether the prediction time is an integer:
if so, taking the prediction time as a prediction step length, and taking a result obtained after prediction as a compensation result of the CQI at the current moment;
if not, the prediction step length determines the upper rounding and the lower rounding of the prediction time, two times of prediction are respectively carried out, and the prediction result obtained according to the upper rounding prediction step length is recorded as X1Obtained by predicting the whole step length from the take-downThe prediction result of (2) is recorded as X2And performing linear interpolation on the two prediction results to finally obtain a compensation result of the CQI at the current moment: x1+aX2Where a is the fractional part of the prediction time.
10. The prediction and compensation system of claim 9, wherein the identification module of the autoregressive-sum moving average model further comprises:
a calculating unit, configured to calculate estimated values of an autocorrelation function and a partial autocorrelation function of the CQI time series, estimated values of an autocorrelation function and a partial autocorrelation function of the CQI time series after first-order difference, and estimated values of an autocorrelation function and a partial autocorrelation function of the CQI time series after second-order difference;
a determining unit, configured to determine an attenuation attribute of an autocorrelation function of the CQI time series, an autocorrelation function of the CQI time series after first-order difference, and an autocorrelation function of the CQI time series after second-order difference, and determine d, which is a difference order required for an ARIMA (p, d, q) model to reach stability, and a value of d is 0, 1, or 2; and determining the properties of the autocorrelation function and the partial autocorrelation function of the d-order differential sequence of the CQI time sequence so as to determine the orders p and q of the autoregressive and moving average operator.
11. The prediction and compensation system of claim 10, wherein the parameter estimation module performs parameter estimation according to the first p + q +1 autocovariance of the d-order differentiated sequence of the CQI time series.
12. The prediction and compensation system of claim 11, wherein the model diagnosis module performs a diagnostic check on the model using a residual error of the sequence of d-order differences of the CQI time series; if the model is appropriate, the residual error becomes closer to a white noise sequence as the length of the sequence increases.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102932801A CN101958765A (en) | 2010-09-27 | 2010-09-27 | Channel quality indication predicting and compensating method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102932801A CN101958765A (en) | 2010-09-27 | 2010-09-27 | Channel quality indication predicting and compensating method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101958765A true CN101958765A (en) | 2011-01-26 |
Family
ID=43485891
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010102932801A Pending CN101958765A (en) | 2010-09-27 | 2010-09-27 | Channel quality indication predicting and compensating method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101958765A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014187337A1 (en) * | 2013-05-22 | 2014-11-27 | 华为技术有限公司 | Method and base station for adjusting channel quality indicator |
| CN104794112A (en) * | 2014-01-16 | 2015-07-22 | 中国移动通信集团山西有限公司 | Time series processing method and device |
| CN105828445A (en) * | 2016-05-23 | 2016-08-03 | 宇龙计算机通信科技(深圳)有限公司 | User equipment dispatching method and base station |
| CN107408269A (en) * | 2015-01-06 | 2017-11-28 | 摩根士丹利 | Nuisance trips of merchandising switch |
| CN109756310A (en) * | 2017-11-02 | 2019-05-14 | 北京大学 | CQI prediction technique in a kind of lte-a system |
| CN112867162A (en) * | 2021-01-25 | 2021-05-28 | 中国联合网络通信集团有限公司 | Slice resource allocation method and device |
| CN113965291A (en) * | 2020-07-20 | 2022-01-21 | 中兴通讯股份有限公司 | Communication control method, base station, terminal, and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1866810A (en) * | 2006-03-28 | 2006-11-22 | 华为技术有限公司 | Method and apparatus for realizing high-speed downlink packet dispatching |
| CN101442807A (en) * | 2008-12-30 | 2009-05-27 | 北京邮电大学 | Method and system for distribution of communication system resource |
| WO2010090787A1 (en) * | 2009-02-04 | 2010-08-12 | Acorn Technologies, Inc. | Two-steps least squares time domain channel estimation for ofdm systems |
-
2010
- 2010-09-27 CN CN2010102932801A patent/CN101958765A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1866810A (en) * | 2006-03-28 | 2006-11-22 | 华为技术有限公司 | Method and apparatus for realizing high-speed downlink packet dispatching |
| CN101442807A (en) * | 2008-12-30 | 2009-05-27 | 北京邮电大学 | Method and system for distribution of communication system resource |
| WO2010090787A1 (en) * | 2009-02-04 | 2010-08-12 | Acorn Technologies, Inc. | Two-steps least squares time domain channel estimation for ofdm systems |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104184538A (en) * | 2013-05-22 | 2014-12-03 | 华为技术有限公司 | Channel quality indicator (CQI) adjusting method and base station |
| WO2014187337A1 (en) * | 2013-05-22 | 2014-11-27 | 华为技术有限公司 | Method and base station for adjusting channel quality indicator |
| CN104184538B (en) * | 2013-05-22 | 2017-12-29 | 华为技术有限公司 | A kind of method of adjustment of channel quality instruction and base station |
| CN104794112A (en) * | 2014-01-16 | 2015-07-22 | 中国移动通信集团山西有限公司 | Time series processing method and device |
| CN104794112B (en) * | 2014-01-16 | 2018-05-22 | 中国移动通信集团山西有限公司 | Time Series Processing method and device |
| CN107408269B (en) * | 2015-01-06 | 2021-03-16 | 摩根士丹利服务集团公司 | Transaction abnormal cut-off switch |
| CN107408269A (en) * | 2015-01-06 | 2017-11-28 | 摩根士丹利 | Nuisance trips of merchandising switch |
| CN105828445A (en) * | 2016-05-23 | 2016-08-03 | 宇龙计算机通信科技(深圳)有限公司 | User equipment dispatching method and base station |
| CN105828445B (en) * | 2016-05-23 | 2019-06-11 | 宇龙计算机通信科技(深圳)有限公司 | A kind of method of scheduling user's set and base station |
| CN109756310B (en) * | 2017-11-02 | 2020-10-20 | 北京大学 | CQI prediction method in LTE-A system |
| CN109756310A (en) * | 2017-11-02 | 2019-05-14 | 北京大学 | CQI prediction technique in a kind of lte-a system |
| CN113965291A (en) * | 2020-07-20 | 2022-01-21 | 中兴通讯股份有限公司 | Communication control method, base station, terminal, and storage medium |
| CN113965291B (en) * | 2020-07-20 | 2024-05-17 | 中兴通讯股份有限公司 | Communication control method, base station, terminal and storage medium |
| CN112867162A (en) * | 2021-01-25 | 2021-05-28 | 中国联合网络通信集团有限公司 | Slice resource allocation method and device |
| CN112867162B (en) * | 2021-01-25 | 2023-04-07 | 中国联合网络通信集团有限公司 | Slice resource allocation method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101958765A (en) | Channel quality indication predicting and compensating method and system | |
| Qin et al. | Effective-SNR estimation for wireless sensor network using Kalman filter | |
| US10827412B2 (en) | Ultra-reliable communication reliability and detection in mobile networks | |
| CN101291517B (en) | Measurement processing method and device | |
| EP2959609A1 (en) | Methods and network nodes in a wireless communication network | |
| CN102664843A (en) | Method and device for single-frequency interference detection and elimination | |
| CN111435926B (en) | MIMO system channel prediction method, device, medium and equipment | |
| CN108476419A (en) | Device for wireless communications and method, parameter optimization device and method | |
| CN105763273A (en) | Cognitive radio spectrum sensing method | |
| KR101699826B1 (en) | Method an system for user speed estimation in wireless networks | |
| CN104753636B (en) | The feedback method that channel quality indicates in communication system and device, communication terminal | |
| CN102316506A (en) | Method and equipment for determining speed range of mobile station | |
| CN112422212B (en) | Data self-adaptive wireless communication channel prediction method, storage medium and equipment | |
| Hekkala et al. | Cooperative spectrum sensing study using welch periodogram | |
| CN104270328A (en) | Method for estimating signal-to-noise ratio in real time | |
| CN103139787B (en) | A kind of disturbance restraining method of drive test data and relevant apparatus | |
| CN101404546B (en) | Frequency spectrum occupancy rate measuring method based on signal moment characteristics | |
| US11223502B2 (en) | Method for predicting the channel between a transmitter/receiver and a connected vehicle | |
| CN113507130A (en) | Power grid state estimation method and system of real-time data communication system based on DPMU | |
| CN101790184B (en) | Method and device for estimating communication quality and base station | |
| Amara et al. | Channel quality prediction in LTE: How far can we look ahead under realistic assumptions? | |
| EP2662999A1 (en) | Receiver, reception method, and computer program | |
| CN101594321A (en) | A kind of method and apparatus of channel estimating | |
| Parashar et al. | Shaping probability density function of quantization noise in fixed point systems | |
| CN102257778B (en) | Channel quality estimation method, device and wireless receiver |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110126 |