Radio link adaption of a channel between a first network element and a second network element in a communication network

0001The invention relates to a method for radio link adaptation according to the preamble of claim 1, a network element according to the preamble of claim 8 and a communication network according to the preamble of claim 9.

0002In wireless communication systems like e.g. Universal Mobile Telecommunications System (UMTS), Third Generation Partnership Project Long Term Evolution (3GPP LTE) or Worldwide Interoperability for Microwave Access (WIMAX), information is transmitted by modulation of a carrier signal. Decoding errors are caused by distortions like e.g. interference or noise of the signal in the radio channel. Channel coding fights this problem by adding redundancy to the information.

0003The ratio R = Information/(Information + Redundancy) is called code rate. In LTE e.g., code rates range from 1/9 to 9/10 dependent on the modulation scheme.

0004Depending on the channel's signal-to-interference-and-noise ratio, parameters or parameter combinations defining the radio transmission over a channel between two network elements, such as power, modulation or coding schemes or combination of theses parameters, have a different performance, and are thus chosen dependent on values indicating the quality of the channel. Said values indicating the quality of the channel are typically called CQI values (CQI = channel quality indicator).

0005According to a common approach to select parameters or parameter combinations defining the radio transmission based on a CQI value, said parameters or parameter combinations defining the radio transmission available in the communication system are arranged in a table according to an increasing threshold value for the CQI value associated with a predefined transport capacity. The threshold value for the CQI value is chosen in such a way, that a corresponding transport capacity of the transmission is equal to a predefined threshold. The threshold value for the CQI value governs the usage of the parameters or parameter combinations defining the radio transmission. All parameters or parameter combinations for which the threshold value is lower than the actual CQI value are candidates that can be used for transmission. The parameter or parameter combination with the highest potential transport capacity will be employed. This scheme is denoted Table Driven Link Adaptation.

0006In the international patent application <patcit id="pcit0001" dnum="WO2005003784W"><text>PCT/IB2005/003784</text></patcit>, link adaptation by means of selection of a modulation and coding scheme (MCS) for high speed downlink packet access (HSDPA) is disclosed. The link adaptation is based on corrected CQI values, which take into account the age of the CQI values. As newer CQI reports generally reflect the current channel conditions more reliably than old ones, it is proposed to use a modulation and coding scheme that is more fault-tolerant, if the CQI report is relatively old. Thus, the corrected CQI value is equal to the most recent CQI value reduced by an offset value which is a function of the age of the most recent CQI value.

0007However, an offset value that is increasing with the age of the most recent CQI value is in most of the real situations not an adequate model, as normally, the channel quality is not deteriorating in the course of time. As a consequence, the corrected CQI value will be too low for old CQI values, and the chosen modulation and coding scheme is too fault-tolerant with a transport capacity below the optimum.

0008The object of the invention is thus to propose a method for radio link adaptation based on a corrected CQI value which is a better approximation of the actual CQI value and which is thus more appropriate. This object is achieved by a method for radio link adaptation according to the teaching of claim 1, a network element according to the teaching of claim 8, and a communication network according to the teaching of claim 9.

0009The main idea of the invention is to determine fluctuations of the CQI value for various points in time, and to use said fluctuations to calculate a corrected CQI value. Said corrected CQI value in turn is used to choose at least one parameter defining the radio transmission over the channel.

0010Further developments of the invention can be gathered from the dependent claims and the following description.

0011In the following the invention will be explained further making reference to the attached drawings.

0012<figref idref="f0001">Fig. 1</figref> schematically shows a cellular communication network comprising base stations and terminals in which the invention can be implemented.

0013<figref idref="f0002">Fig. 2</figref> schematically shows the LTE transport capacity of the transmission dependent on the signal-to-interference-and-noise ratio for different modulation and coding schemes.

0014<figref idref="f0003">Fig. 3</figref> schematically shows a processing chain that is used for calculation of a corrected CQI value according to the invention.

0015A communication network in which the invention can be implemented comprises various network elements like user terminals or base stations.

0016<figref idref="f0001">Fig. 1</figref> shows an example for such a network CN that comprises base stations BS1-BS8 and user terminals T1-T4.

0017Each of said user terminals T1-T4 is connected to one or multiple of said base stations BS1-BS8, which is symbolized by double arrows in <figref idref="f0001">fig. 1</figref>. The base stations BS1-BS8 are in turn connected to a core network, which is not shown in <figref idref="f0001">fig. 1</figref> for the sake of simplicity.

0018In an embodiment of the invention, the base station BS4 is connected to the base station BS5, and the base station BS6 is connected to the base station BS7 via radio connections, which is symbolized by dashed double arrows in <figref idref="f0001">fig. 1</figref>.

0019The user terminals T1-T4 comprise the functionality of a user terminal for transmission and reception of signaling and data messages in a network using radio transmission.

0020Furthermore, the user terminals T1-T4 comprise the functionality to determine CQI values based on measurements of a radio channel.

0021The base stations BS1-BS8 comprise the functionality of a base station of a network using radio transmission, i.e. they provide the possibility for user terminals to get connected to said network and for data exchange of said user terminals by means of radio transmission.

0022Said base stations BS1-BS8 also comprise the functionality to select at least one parameter defining the radio transmission over a radio channel dependent on a received CQI value of said radio channel.

0023Furthermore, a base station BS1-BS8 according to the invention is adapted to determine fluctuations of values indicating a quality of the channel, i.e. CQI values, measured at various points in time, calculate based on said fluctuations a corrected value indicating a quality of the channel, and choose at least one parameter defining the radio transmission over the channel based on said corrected value indicating a quality of the channel.

0024In an embodiment of the invention, the base stations BS1-BS8 comprise the functionality to determine CQI values based on measurements of a radio channel.

0025The base stations BS1-BS8 support a multitude of parameters or parameter combinations defining the radio transmission over a channel between two network elements, such as power, modulation or coding schemes or combination of theses parameters, like the modulation and coding schemes (MCS).

0026The selection of e.g. a modulation and coding scheme (MCS) is based on the channel quality information available as CQI values, which is in our embodiment depicted in <figref idref="f0001">fig. 1</figref> reported by the user terminals UE1-UE4 to the respective connected base station BS1, BS6 and BS7. In another embodiment, it is also possible, that CQI values are reported from a first base station BS4 to a second base station BS5. The modulation and coding schemes (MCS) available in the communication network are ordered in a table preferably according to an increasing threshold value for the CQI value that is associated with a predefined performance of the transmission. The threshold value for the CQI value is determined in such a way, that a corresponding transport capacity of the transmission, e.g. indicated in bits per symbol, is equal to a predefined threshold for the performance of the transmission. Alternatively, the threshold value for the CQI value is determined in such a way, that a corresponding frame or block error rate of the transmission is equal to a predefined threshold for the performance of the transmission. The threshold value for the CQI value governs the usage of a modulation and coding scheme. All modulation and coding schemes for which the threshold value for the CQI value is lower than the reported CQI value are candidates for transmission, and preferably, the modulation and coding scheme with the highest potential transport capacity will be employed. This scheme is denoted Table Driven Link Adaptation.

0027In another embodiment, for each reported CQI value, it can be calculated for which modulation and coding scheme the transport capacity of the transmission for the reported CQI value is higher than a predefined threshold for the performance of the transmission. Then, from all modulation and coding schemes with a transport capacity of the transmission for the reported CQI value higher than the predefined threshold for the performance of the transmission, the one with the highest potential transport capacity will be employed.

0028In <figref idref="f0002">fig. 2</figref>, the simulated LTE transport capacity of the transmission in bits per OFDM symbol is depicted dependent on the signal-to-interference-and-noise ratio (SINR) in dB for different modulation and coding schemes that are indicated by different numbers. The signal-to-interference-and-noise ratio is related to the channel quality indicator, i.e. a SINR value can be mapped on a CQI value. The modulation and coding schemes for which the LTE transport capacity is depicted are combinations of the modulation schemes Quadrature Phase Shift Keying (QPSK) and Quadrature Amplitude Modulation (QAM) with coding schemes with different coding rates R.

0029The predefined performance of the transmission, i.e. the required transport capacity of the transmission is depicted as a dotted horizontal line. The intersection point of said dotted line and the transport capacity of the transmission indicates for each modulation and coding scheme a threshold value for the SINR, which is indicated as vertical dotted lines. The SINR value must be equal to or higher than the threshold value for the SINR in order to achieve the required transport capacity with the respective modulation and coding scheme.

0030All modulation and coding schemes for which the threshold value for the SINR is lower than the current SINR value are candidates for transmission, and preferably, the modulation and coding scheme with the highest potential transport capacity will be employed. In the example depicted in <figref idref="f0002">fig. 2</figref>, the current SINR value is 5 dB, and the modulation and coding schemes numbered 9 and 10 are candidates for transmission. The modulation and coding scheme with number 10, i.e. QPSK with R=0.73 will be employed, as this is the scheme with the highest potential transport capacity in the vicinity of the SINR value of 5 dB.

0031As already described above, the modulation and coding schemes available in the communication network can be ordered in a table according to increasing threshold values for the SINR.

0032In another embodiment, the threshold value for the SINR of a modulation and coding scheme corresponds to the SINR value at which the simulated LTE transport capacity of the transmission for a modulation and coding scheme touches the envelope of all simulated LTE transport capacities, which is depicted as a dotted curve in <figref idref="f0002">fig. 2</figref>. Then, from all modulation and coding schemes with a threshold value for the SINR below the reported SINR value, the one with the highest threshold value for the SINR value is chosen for transmission.

0033According to the invention, fluctuations of CQI values measured by a network element UE1-UE4 or BS4-BS7 at various points in time are determined, said fluctuations are used for the calculation of a corrected CQI value in a network element UE1-UE4 or BS4-BS7, and said corrected CQI value is used to choose at least one parameter defining the radio transmission over the channel, which is preferably a power, a modulation or coding scheme or a combination of these parameters.

0034The at least one parameter defining the radio transmission over the channel is chosen based on said corrected CQI value in such a way, that either a corresponding frame or block error rate of the transmission is below a predefined threshold for the performance of the transmission, or a corresponding performance of the transmission, e.g. indicated in bits per symbol, is higher than a predefined threshold for the performance of the transmission.

0035A fluctuation as mentioned above can be the difference between two CQI values at different points in time. Preferably, said difference is periodically measured, and said fluctuations are differences between a first CQI value measured at point in time t1 and a second CQI value measured at point in time t1+Δt for various points in time t1.

0036Preferably, the difference can be subtracted from the most recent CQI value of said two CQI values, which leads to the corrected CQI value.

0037In an embodiment of the invention, said fluctuations are used for the calculation of a correction value, and said correction value is combined with the most recent CQI value to obtain the corrected CQI value.

0038Preferably, the most recent CQI value is added to said correction value multiplied by a factor and added to a bias, which results in the corrected CQI value.

0039In an embodiment of the invention, said correction value is the standard deviation of said differences.

0040<figref idref="f0003">Fig. 3</figref> shows a processing chain that is used for calculation of a corrected CQI value according to an embodiment of the invention and that is implemented in a network element as e.g. a user terminal T1-T4 or a base station BS1-BS8.

0041The network element uses CQI values to judge the quality of its radio channels. Therefore, the so-called self-similarity of the radio channel e.g. for a given round trip delay is analyzed. The self-similarity is derived from the behavior of the radio channel over time. It is calculated by comparing a sequence of CQI values against an identical but time-shifted copy of itself. The round trip delay RTD is defined as the time that elapses between the time of a CQI measurement in a network element and the time of a data transmission based on that CQI measurement, i.e. the round trip delay RTD is the shortest delay between the generation of a CQI value in a network element and the reception of a data packet which was transmitted based on said CQI value.

0042In a first step 1 in <figref idref="f0003">fig. 3</figref>, a CQI value at time t-RTD is determined by means of choosing the appropriate CQI value from a multitude of stored CQI values from the past.

0043In a second step 2, in order to find the self-similarity of a radio channel, the difference between a CQI value at time t and the CQI value at time t-RTD is calculated for multiple points in time t.

0044In a third step 3, said differences at multiple points in time t are stored, and the standard deviation of said differences is calculated.

0045In a fourth step 4, the standard deviation is multiplied by a factor f, and in a fifth step 5, a bias b is added to the standard deviation that has been multiplied by a factor f. This results after step 5 in a correction value for the CQI value. The default values, i.e. the normally used values, for factor f and bias b are f=-1 and b=0, which corresponds to the usage of the standard deviation as correction value.

0046In a sixth step 6, The calculated correction value for the CQI value is added to the most recent CQI value.

0047The corrected CQI value is used to look up e.g. a modulation and coding scheme in the link adaptation table as described above.

0048Simulations that have been performed show that by reducing the most recent CQI value by the standard deviation of the differences between the CQI value at time t and the CQI value at time t-RTD for multiple points in time t, the spectral efficiency in dense urban environments has been improved by about 10% compared with a table driven link adaptation algorithm according to the prior art.

0049It is to be expected, that the spectral efficiency can be further improved by means of performing simulations with different values for the factor f and the bias b in order to find out the most appropriate values for the factor f and the bias b.