WO2018228704A1

WO2018228704A1 - A network device, a user equipment and a method for wireless data transmission

Info

Publication number: WO2018228704A1
Application number: PCT/EP2017/064816
Authority: WO
Inventors: Majid NASIRI KHORMUJI; Jianqin Liu; Branislav M POPOVIC
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2017-06-16
Filing date: 2017-06-16
Publication date: 2018-12-20

Abstract

An apparatus, a network device and a method, comprising the network device receiving uplink information from a user equipment, wherein the user equipment is connected to a wireless network via multiple subbands; assigning a rank index over multiple subbands based on the uplink information; compressing the rank index; and sending the compressed rank index to the user equipment via the wireless network.

Description

A NETWORK DEVICE, A USER EQUIPMENT AND A METHOD FOR

WIRELESS DATA TRANSMISSION

BACKGROUND This disclosure relates to the field of wireless communications technologies, and in particular to wireless data transmission.

To improve the development of wireless communications technologies, a Long Term Evolution (LTE) project has been set up by the 3rd Generation

Partnership Project (3rd Generation Partnership Project, 3GPP). Multiple-Input Multiple-Output (MIMO) and Orthogonal Frequency Division Multiplexing

(OFDM) are among the technologies defined in the LTE project. Fifth- generation radio transmission 5G NR (Next Radio) has been set to develop the transmission technologies further.

A user equipment (UE) may be connected to the radio network via multiple antennas. A serving base station, for example an evolved Node B (eNB), provides connection to multiple users via multiple antennas. Each antenna may be assigned to a different transmission band in the multiplexed radio interface. The user equipment measures radio link transmission quality parameters and sends a report to the network, for example to the eNB. In the MIMO radio transmission model, different antennas and transmission bands may have some interference between each other; neighboring cells may have interference or the signal quality may be affected by other physical properties. In one example the eNB receives the report and as a response, sends a Rank Index (Rl) to the user equipment. In some instances, the Rank Index is also called as Rank Indicator. The eNB may configure the Rank Index periodically or during an attach procedure and send the Rank Index to the user equipment.

In one example of the LTE, two transmission schemes with single antenna transmission and closed-loop MIMO transmission are supported for PUSCH (Physical Uplink Shared Channel). A closed-loop MIMO is a codebook-based transmission comprising encoding information bits, wherein the eNB indicates the precoding matrix information (PMI) and rank index (Rl) information in the uplink (UL) grant. A single precoding matrix indicated by this PMI along with the number of layers indicated by Rl may be applied for the data transmission of the whole scheduled bandwidth by the user equipment. In other words, the precoding for uplink MIMO in the LTE is not frequency-selective. In the 5G NR, the system bandwidth may be increased to 1 00MHz or more in contrast with that of 20MHz in the LTE system. The main consequence of enlarging the uplink transmission bandwidth is that the channel properties can vary across the scheduled bandwidth. As one example, the rank and precoding matrix of different subbands in a large system bandwidth could be different. Uplink subband scheduling, also known as frequency-selective scheduling, with adaptive rank is hence considered for NR as a potential transmission candidate.

The rank index information is transmitted over the radio link with direct bit encoding, wherein the bits used refer to the rank index number. The rank index of each subband is reported independently from the other subbands. As an example of a user equipment having four transmit antennas, the rank of the uplink channel varies in the set {1 ,2,3,4}, which requires two bits to be reported from the eNB to the user equipment, encoded accordingly according to the set {00, 01 , 1 0, 1 1 }.

If the number of subbands is equal to N, then the direct solution needs to report Nlog( T) bits, where K denotes the maximum number of layers (i.e. the minimum number of user equipment antennas and eNB antennas). When there are four antennas at the user equipment and the eNB has a larger number of antennas, the number of bits to report the rank index becomes 2N - increasing linearly with the increasing number of subbands. The number of bits to be reported from the eNB to the UE may increase along the next generation transmission systems. The overhead on the radio transmission link becomes severe as the transmission bandwidth increases.

SUMMARY

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. As one aspect, this disclosure explains the procedure to reduce the feedback overhead of a rank index feedback report from a network device to the user equipment. The feedback overhead may be reduced by utilizing high correlation among rank indexes over different subbands. A first aspect discloses a network device for a wireless communication network. The network device comprises a transceiver configured to receive uplink information related to multiple subbands; and a processor configured to assign a rank index over said multiple subbands, to compress the rank index and to cause the transceiver to send the rank index. The uplink information may be an uplink pilot report from a user equipment connected to the wireless network via multiple subbands. The network device may be any component, node or a terminal connected to the telecommunication network. In one embodiment the network device may be a single eNB, an aggregate of network devices or a network eNB. The network device may receive the uplink information from the user equipment, wherein the user equipment is connected to a wireless network via multiple subbands. The compressed rank index may be sent to the user equipment. The feedback report having the rank index is sent from the network device to the user equipment as compressed data, reducing the amount of the radio transmission bandwidth used. In an implementation of the first aspect the processor of the network device is configured to determine a difference between the rank index and a previously sent rank index; and cause the transceiver to send the difference. The network device may be configured to obtain a previously sent rank index, compare the rank index with a previously sent rank index and detect a difference between the rank index and a previously sent rank index. In one example the network device may store the previously sent rank index to a memory and obtain the previously sent rank index from the memory. In one example the memory is an external component from where the previously sent rank index may be fetched. The rank index according to the implementation may be a sequence of rank indexes, wherein the sequence of rank indexes may comprise information on all antennas and/or subbands of the user equipment. In a further implementation of the first aspect, the processor of the network device is configured to determine a sequence of rank indexes of neighboring subbands within said multiple subbands; encode the sequence of rank indexes to a sequence of differences between two consecutive rank indexes in the sequence of rank indexes, wherein the difference is one of the group of -1 , 0, +1 ; decompose the sequence of differences between two consecutive rank indexes as a sum of two binary sequences, wherein the first binary sequence consists of differences 0 and -1 ; and the second binary sequence consists of differences 0 and +1 ; and encode the first binary sequence and the second binary sequence with enumerative source encoding. The enumerative source encoding is a well-known coding method, wherein the principles were published in "IEEE Transactions on Information Theory" and introduced by Thomas M. Cover in 1973. The differences between the neighboring subbands may be incremental, as the environment within the neighboring subbands regarding time, frequency or space may resemble each other.

In a further implementation of the first aspect the processor of the network device is configured to determine encoding information bits comprising multiple predefined rank indexes and encoded bits corresponding to the predefined rank indexes; compare the rank index with the encoded bits; determine the encoded bits corresponding to the rank index; and cause the transceiver to send the encoded bits. The encoding information bits may be obtained from the memory, wherein the encoding information bits comprise multiple permutations of the encoded bits. In one example of an embodiment the network device is configured to calculate and generate the encoding information bits according to the information obtained from the sequence of rank indexes.

In a further implementation of the first aspect the processor of the network device is configured to generate encoding information bits; encode, by using the encoding information bits, encoded bits corresponding to the rank index; and cause the transceiver to send the encoding information bits and the encoded bits. In one example the network device is configured to send the encoding information bits and the encoded bits to the user equipment. A second aspect discloses a user equipment for a wireless network, comprising a processor and a transceiver, wherein the processor is configured to cause the transceiver to: connect to the wireless network via multiple subbands; send uplink information related to said multiple subbands; and receive a compressed rank index over said multiple subbands; and wherein the processor is configured to decompress the compressed rank index. The user equipment may be regarded as a receiver-transmitter pair with the network device, wherein the network device resides as part of the wireless network.

In an implementation of the second aspect the processor of the user equipment is configured to cause the transceiver to: receive the compressed rank index as a difference between the rank index and a previously sent rank index; and wherein the processor is configured to decompress the compressed rank index using the difference and the previously sent rank index. The sequence of rank indexes may comprise information on all antennas of the user equipment. In a further implementation of the second aspect the processor of the user equipment is configured to: decode a sequence of rank indexes from a first binary sequence and a second binary sequence encoded with enumerative source encoding; wherein between two consecutive rank indexes in the sequence of rank indexes, the difference is one of the group of -1 , 0, +1 ; the first binary sequence consists of differences 0 and -1 ; and the second binary sequence consists of differences 0 and +1 . The user equipment may be configured to decode the encoded sequence of rank indexes according to one implementation of the first aspect.

In a further implementation of the second aspect the processor of the user equipment is configured to cause the transceiver to: receive encoding information bits comprising multiple predefined rank indexes and encoded bits corresponding to the predefined rank indexes; receive the encoded bits corresponding to the rank index; and decompress the rank index by using the encoded bits and the encoding information bits. A third aspect discloses a method, comprising: a network device receiving uplink information from a user equipment, wherein the user equipment is connected to a wireless network via multiple subbands; assigning a rank index over multiple subbands based on the uplink information; compressing the rank index; and sending the compressed rank index to the user equipment via the wireless network.

In an implementation of the third aspect the method comprises the network device: detecting a difference between the rank index and a previously sent rank index; sending the difference to the user equipment; and the user equipment calculating the rank index using the difference and the previously sent rank index.

In a further implementation of the third aspect the method comprises the network device: receiving a sequence of rank indexes of neighboring subbands within said multiple subbands; encoding the sequence of rank indexes to a sequence of differences between two consecutive rank indexes in the sequence of rank indexes, wherein the difference is one of the group of -1 , 0, +1 ;

decomposing the sequence of differences between two consecutive rank indexes as a sum of two binary sequences, wherein the first binary sequence consists of differences 0 and -1 ; and the second binary sequence consists of differences 0 and +1 ; and encoding the first binary sequence and the second binary sequence with enumerative source encoding.

In a further implementation of the third aspect the method comprises the network device determining encoding information bits comprising multiple predefined rank indexes and encoded bits corresponding to the predefined rank indexes; the network device sending the encoded bits and the encoding information bits to the user equipment; the user equipment receiving the encoded bits and the encoding information bits; and determining the rank index by using the encoded bits and the encoding information bits.

In a further implementation of the third aspect the method comprises the network device generating encoding information bits; encoding, by using the encoding information bits, encoded bits corresponding to the rank index;

sending the encoding information bits and the encoded bits to the user equipment; the user equipment receiving the encoding information bits and the encoded bits; and determining the rank index by decoding the encoded bits with the encoding information bits. The embodiment of providing the encoding information bits has the advantage of very low complexity. It can be easily encoded and decoded in order to recover the information of the rank for different subbands.

A fourth aspect discloses a computer program product comprising instructions, which when the program is executed by a computer, cause the computer to carry out the steps of the method according to the third aspect. In one embodiment the computer program comprises instructions configured to carry out the steps of the network device. In one embodiment the computer program comprises instructions configured to carry out the steps of the user equipment. The network device, the user equipment and the method described

hereinbefore reduce the overhead related to the feedback report of the rank of the channels which determines the number of layers for uplink data

transmission. The compressed rank index reduces the overhead. The saved resources, e.g. time-frequency resource elements, may be used for example to transmit further information. Regarding the downlink control information (DCI) of the Physical Downlink Control Channel (PDCCH), the message size of the downlink control information may vary. As one example of an embodiment, the worst case message size may be considered, and according to that design, the associated control region can be assigned. In one embodiment a dynamic message size may be used, wherein the size of the downlink control information is signalled. Reducing the number of bits of the control channel is beneficial, as a lower coding rate can be used, further extending the coverage of the control channel.

Many of the attendant features will be more readily appreciated as they become better understood by reference to the following detailed description considered in connection with the accompanying drawings. The embodiments described below are not limited to implementations which solve any or all the

disadvantages of known radio transmission systems. BRIEF DESCRIPTION OF THE DRAWINGS

The present description will be better understood from the following detailed description read in light of the accompanying drawings, wherein

FIG. 1 a schematically illustrates an example of a telecommunication system for practicing embodiments of the present invention;

FIG. 1 b schematically illustrates an example of a telecommunication system for practicing embodiments of the present invention;

FIG. 2 illustrates a simplified signalling flowchart of an example of one embodiment; FIG. 3 illustrates a simplified signalling flowchart of an example of one embodiment;

FIG. 4 illustrates a simplified signalling flowchart of an example of one embodiment;

FIG. 5 shows an example of a simulation according to an embodiment; FIG. 6 shows a table of simulation assumptions according to the simulation example;

FIG. 7 shows a table illustrating results according to the simulation example;

FIG. 8 shows a table of simulation results according to direct encoding;

FIG. 9 illustrates simulation of one example of an embodiment; FIG. 10 illustrates simulated values of one example of an embodiment;

FIG. 1 1 shows a table of simulation results of one example of an embodiment; and

FIG. 12 illustrates simulated values of one example of an embodiment.

Like reference numerals are used to designate like parts in the accompanying drawings. DETAILED DESCRIPTION

The detailed description provided below in connection with the appended drawings is intended as a description of the present examples and is not intended to represent the only forms in which the present example may be constructed or utilized. However, the same or equivalent functions and sequences may be accomplished by different examples.

FIG. 1 a schematically illustrates one simplified example of a telecommunication network, disclosing one exemplary embodiment of a network device and a user equipment. In this example, user equipments 130, 140, 150 are wirelessly connected to a network device 120. All user equipments 130, 140, 150 comprise similar components to the user equipment 130: at least one processor 131 and a memory 132 for storing instructions that, when the instructions are executed, cause the user equipment 130 to perform the functionality described hereinafter. A transceiver 133 is configured to provide a communication link to the telecommunication network via a radio link. The functionality of the transceiver 133 may be independent as the transceiver may possess a dedicated processor. The processor 131 may be used to cause the transceiver to execute specific functions. The network device 120 provides a serving cell 121 for the user equipments

130, 140, 150. In this example, a portion of a telecommunication network 100 is represented by a cloud, as the telecommunication network 100 may be implemented in various ways. The telecommunication network 100 may reside at least partially in a cloud computing environment. In one embodiment, a portion of the functionality of the network device 120 is distributed to an apparatus 1 10, which is connected via the telecommunication network 100 to the network device 120. The apparatus 1 10 comprises at least one processor 1 1 1 and a memory 1 12 for storing instructions that, when the instructions are executed, cause the apparatus 1 10 and the network device 120 to perform the functionality described herein. A transceiver 1 13 is configured to provide a communication link to the user equipments 130, 140, 150 via a radio link. Further, in one embodiment the apparatus 1 10 is configured to control the functionality of the network device 120.

The network device 120 may be an evolved Node B (eNB) providing a direct radio link to the user equipments 130, 140, 150, an aggregation of network devices, an access node or a network eNB. FIG. 1 b schematically illustrates another simplified example of a telecommunication network, disclosing one exemplary embodiment of the network device 120 and a user equipment 130, wherein the apparatus 1 10 resides at the network device 120. The apparatus 1 10 and its components may be defined as a portion of the network device 120. Examples of wireless communications technologies having embodiments disclosed herein are the LTE network or 5G NR network. Examples of transmission technologies, wherein the present disclosure is configured to operate, are Multiple- In put Multiple-Output (MIMO) and Orthogonal Frequency Division Multiplexing (OFDM). Although the present examples are described and illustrated herein as being implemented in a mobile telecommunication network, the network device, the apparatus and the user equipment as described are provided as an example and not a limitation. As those skilled in the art will appreciate, the present examples are suitable for application in a variety of different types of communication networks with multiple subbands, wherein the subbands may be arranged by frequency, time division

multiplexing, orthogonal frequency multiplexing or any combination thereof and the use of subbands benefits from using the rank index definition. The feedback report of the rank index of the channels determines the number of layers for uplink data transmission. In one embodiment the rank index according to the disclosure herein is applied at Physical Uplink Shared Channel (PUSCH) Multiple-Input Multiple-Output transmission in 5G NR network.

The rank index Rl reporting for multi-subbands reduces the downlink channel overhead, i.e. the number of bits to inform the user equipment UE about the rank index Rl, by utilizing rank correlation across the subbands. According to one example the rank indexes Rl over different subbands are related to the adjacent ones. The differential between the adjacent or neighboring subbands does not exist or is very limited. In many cases the rank index differential equals one rank value. The feedback overhead is reduced by generating compressed information about the rank index of multiple subbands.

FIG. 2 illustrates a simplified signalling flowchart of an example of one embodiment. The user equipment UE sends uplink pilot information, providing uplink information to the network device AN, arrow 210 - the uplink pilot information may be generated according to the measurements of the user equipment UE. The network device AN does channel estimation as a response to the uplink information, step 220. In step 230 the network device AN provides rank indexes over multiple subbands, wherein the number of subbands is related to the user equipment's capabilities. The network device AN

compresses the rank index, for example a sequence comprising multiple rank indexes, by utilizing correlation of neighboring subbands or correlation to previously sent rank index information, step 240. In step 250 the compressed rank index is sent to the user equipment, in this example by compressed encoding bits of the rank index over multiple subbands. In step 260 the user equipment decompresses the information and obtains the rank index over multiple subbands.

In an embodiment, differential reporting of the rank is used, and only the change in the rank of the channel with respect to the earlier subband is reported. Since the change in the rank is bounded and in some cases there is no change, the number of feedback bits associated to the rank index will be reduced. A compression method is used to encode the differential change for reporting of the rank index which amounts to a much lower number of encoding bits.

FIG. 3 illustrates a simplified signalling flowchart of an example of one embodiment. The user equipment UE sends uplink pilot information, providing uplink information to the network device AN, arrow 310 - the uplink pilot information may be generated according to the measurements of the user equipment UE. The network device AN does channel estimation as a response to the uplink information, step 320. In step 330 the network device AN provides rank indexes over multiple subbands. In step 340 the network device AN generates differential rank index information. According to step 350, in one embodiment the network device is configured to convert changes having a differential greater than +1/-1 to the closest change +1/-1 . In greater differentials the compression allows the rank index to move to the correct direction, wherein the optimized rank index value may be achieved after multiple iterations of updating the rank index in the user equipment UE.

After step 350, all differentials have one value of a group consisting of -1 , 0 and +1 .

In step 360 the differential sequence of rank indexes is decomposed into two binary sequences, wherein a first binary sequence consists of differences 0 and -1 . The second binary sequence consists of differences 0 and +1 . The two binary sequences are encoded in step 370 and transmitted to the user equipment UE as compressed encoding bits of the rank index over multiple subbands, arrow 380. In step 390 the user equipment UE decodes the rank index over multiple subbands. FIG. 4 illustrates a simplified signalling flowchart of an example of one embodiment. The user equipment UE sends uplink pilot information, providing uplink information to the network device AN, arrow 410. The network device AN does channel estimation as a response to the uplink information, step 420. In step 430 the network device AN provides rank indexes over multiple subbands. In step 440 the network device AN generates new encoding information bits for rank indexes and encodes the rank index using the new encoding information bits, step 450. The candidate rank of the channels in each subband, often comprising two candidate ranks, is remapped with the new encoding

information bits. The new encoding information bits are for example new bits encoding the selected rank indexes. Arrows 460, 470 illustrate the encoded and compressed rank indexes being sent to the user equipment UE along with the new encoding information bits for the rank index. The new encoding information bits may change depending on the particular realization of the rank index over multiple subbands, in other words the encoding information bits may be adaptive. Source compression of highly correlated rank distribution is applied to the generation of encoding information bits, notably reducing the amount of the feedback overhead. In step 480 the user equipment may decode the rank index using the encoding information bits. The correlation of the neighboring subbands' rank indexes is explained with the following examples of embodiments and simulations, wherein FIG. 5 shows an example of ranks of the uplink channels over 10 subbands for 6 user

equipments UE when the total uplink bandwidth is 100 MHz. Each subband is set to 10 MHz. The rank indexes, along with the number of uplink data layers, can vary over different subbands for each user equipment UE. The simulation assumptions are illustrated in the table of FIG. 6. The table according to FIG. 7 shows the throughput gain of uplink subband precoding with respect to wideband precoding (LTE baseline) when the number of antennas at the eNB is 8 and the number of antennas at the user equipment UE is two and four. In contrast to the baseline LTE system, for which only a fixed number of layers along with a precoding matrix is selected for all scheduled resources for each user equipment, for the subband precoding the number of layers with the precoding matrix is updated. The spectral efficiency gain of the subband precoding is notable, wherein more substantial effects may be achieved in 5G NR. The simulation also indicates high correlation of the rank indexes over different subbands.

It is observed that the ranks of the channel over the frequency subbands are highly correlated and for one of the first user equipments in the example of FIG. 5 the maximum change in the rank index is one bit. The rank index of the channel for the first user equipment over 10 subbands changes according to the table of FIG. 8. illustrating direct encoding without compression for the first user equipment in FIG. 5. By applying the direct encoding, the amount of the feedback results in 20 bits. However, since the changes in the rank index are highly correlated and the maximum change from one subband to the

neighboring subband is one, the network device, for example the eNB, can indicate the change in the rank index with only one bit.

FIG. 9 illustrates one example of an embodiment with differential encoding with respect to the earlier subband. The differential encoding amounts to 1 1 bits as compared to 20 bits following the direct encoding. The example illustrates a notable reduction in the feedback overhead. The encoding based on the differential rank index in explained hereinafter. An assumption of N uplink transmit subbands is used. The rank of each uplink subband channel is estimated in the user equipment from uplink reference signals. The ranks of all subband channels are fed back to the uplink

transmitter. If K transmit antennae at the user equipment are assumed, and there is an equal or a larger number of antennae at the uplink receiver, the highest rank of each subband channel for uplink MIMO transmission is at most

K. Consequently, ^1ο^(^κ) bits are needed per subband rank feedback, so in total Q_r = Nio_g2(K) _{bits are needed} f_{or the} feedback.

Assuming that the ranks of neighboring subbands cannot defer by more than +/-

I W-l

1 , the sequence ^^η'^η=ι of subband ranks is differentially encoded, where ^r° is the

< W-l rank of the first (lowest or highest) subband, to another sequence ^{½ «=1} , as -_l> n = \,...,N ^~ \ _ ^

The sequence ^" ^«=1 is a sparse ternary sequence, whose elements are mostly zeros. The rank is changing relatively slowly over subbands, occasionally +1 or -1 , referring to FIG. 9.

I W-l

The sequence ^" ^«=1 is decomposed as the sum of two binary sequences, one containing zeros and +1 s, and the other containing the same zeros and -1 s, i.e.

W-l W-l

-i _ +

(2)

W-l ( _ W-l

Sequences ^^«=1 and ?" >ⁿ⁼¹ are separately encoded using enumerative source encoding by Cover in "IEEE Transactions on Information Theory", vol. IT-19, pp. 73-77, Jan. 1973. According to the encoding algorithm, if the sorted positions of all ^M+ nonzero elements in the binary sequence

are given as a sequence ^ *=° , ^{1≤ lk ≤ N ~} ^ ^{lk < lk+1} , and the sorted positions of all ^M~ nonzero elements in the binary sequence * " >ⁿ⁼¹ are given as a sequence j i≤ⁱ _k≤N - i, ½ < ½+i _; then each sequence of indices of nonzero elements can be encoded by unique labels ^L+ and ^L~ , given as [1 , Eq. (4)] whe

is the extended binomial coefficient.

From (3) it follows that L ^ ^{M J} J and L ^ ^{M J} J . It means that

W-l

the number of bits needed to encode the sequence of all ranks ⁿ >*=o is

The improvement of spectral efficiency expressed in percentages obtained by this kind of encoding can be described as the ratio

In the simulations, it is observed that the ranks of neighbouring subbands cannot defer by more than +/- 1 . However, for potential cases where the change is greater than +/- 1 , the network device, such as the eNB, may just report the closest change. For example, for the cases of a change equal to +2 it may report +1 and for the case of a change equal to -2 it may report -1 . This is still beneficial since the baseline wideband reporting ignores to report any change.

FIG. 10 illustrates simulated values of one example of an embodiment, showing the probability mass function of the gain in the equation (5), G, when the number of subbands is N=5 and 10 and the number of user equipments' antennas is K=4. The simulation results show that there are three potential cases based on the transition M⁺ and ^ .

The average gain in feedback reduction can be computed in these simulations to be 56% and 61 % for N=5 and N=10, respectively. According to these exemplary simulations the overhead is reduced by more than a half. As the number of subbands increases, the compression gain increases, as the possibility to use the correlation in the compression improves.

When decoding indices of non-zero differentially encoded ranks, the decoder r -generates the set of indexes ^k'^k=0 from the received label r. The indexes \^sk =o can be found by the following algorithm: 1 for ^{k = 0} to M -1,

jN-x

P M-k while ^p>r. x = x + 1

'N-x end s_k =x +l r = r— p end

According to one embodiment, updated source coding is generated with new encoding information bits. FIG.11 shows one example of the embodiment by encoding with adaptive encoding information bits. The new encoding

information bits are transmitted along with the labeling to the user equipment. In this example the two observed rank labels are 00 and 01. The new encoding information bits are generated by relabeling 00 and 01 to 0 and 1. This relabeling should also be transmitted to the user equipment. The number of bits reduces to 10 bits for reporting the ranks and 6 bits for transmitting the encoding information bits, totalling in 1 6 bits. This example reduced the overhead by 20%, by applying compression on the feedback information. The compression reduces the amount of the feedback as the feedback of the rank index is highly correlated.

The overhead of the embodiment may be calculated according to the following model: for finding the candidate ranks that appear after channel estimation there exist ^M different possible ranks. The parameter M \_s supposed to be less than the total set of possible ranks - for example, for 4 antennas normally two ranks show up. To encode the observed rank for each subband

^^(^ bits are needed; and hence when the number of subbands is ^N . Since the dictionary varies depending on the channel realization, the encoding information bits need to be transmitted. At most are needed, where K is the highest rank which is equal to the number of antennas at the user equipment. _ _l M log₂{K)+ M log₂{M )+ N log₂{M )

N log₂(K) ₍₆₎

Figure 12 shows the probability mass function of the gain in equation (6), G, when the number of subbands is N=5 and 10 and the number of user equipments' antennas is K=4. The simulation results show that there are two potential cases, either M=2 (i.e. two potential different ranks) or M=1 (i.e. only one potential rank).

The average gain in feedback reduction becomes 24% and 41 % for N=5 and N=10. As the number of subbands increases, the compression gain increases accordingly, since the correlation effect is more significant. The encoder complexity may be evaluated by the following two examples. While one example implementation requires at most 15 integer multiplications, 5 integer divisions and 5 integer additions, another example implementation eliminates the multiplications and divisions by employing a 135-word lookup table, and requires 5 integer additions and 6 table look-ups. In one exemplary implementation, a binomial coefficient is rewritten and implemented as

, requiring computation of

^•^ ^{- 1} integer multiplications for the numerator and ^{y ~ 2} for the denominator along with one integer division whose result also is an integer. Hence, the computation of the label r in (1 ) (being the sum of ^M binomial coefficients),

-l M -2 requires *=° integer multiplications for the M numerators and *=° for the M - 1 denominators (the rightmost binomial does not require a division), along with M - l integer divisions and M - l integer additions.

Since the M - l denominators do not depend on the particular set of subband indexes ^*=° , they can be computed in advance and stored. Thus, for the value of ^{M =} 6 , the computation requires 15 integer multiplications along with 5 integer divisions and 5 integer additions. The 5 denominator values 720, 120, 24, 6, and 2 are stored.

A second exemplary implementation employs a look-up table approach. In each

of the M coefficients ^x takes one of the N values in the range °^> · · · ^> Ν I (since ^{1≤ Sk ≤ N} ) while ^y takes one of the M values in the range ^^{■■■■> M} (since ≤k≤M - ly Therefore there are NM relevant binomial coefficients for the computation. Moreover, since coefficients with ⁼ 0 or ^{y = 1} are trivial (

/°\ = 0 \ ) = ^x

\^yl and W ), a set of \^{N ~}^K^{M _ 1}) relevant binomial coefficients remains. These {^{N ~} ^\^{M _ 1}) non-trivial relevant coefficients are stored in a look-up table with indices ^x and ^y . The values N = 28 _and M = 6 |_eac| to a table containing 135 binomial coefficients. The computation thus requires 5 integer additions and 6 table look-ups in a 135-word table (where the wordlength is 19 bits as the largest relevant coefficient among the stored ones, ⁶ J , requires 19 bits). Finally note that this 135-word table of relevant binomial coefficients is computed with integer additions only (the entries being elements in the Pascal triangle). Therefore, the table size can directly be traded for additions, opening up a variety of implementations with small table sizes and addition-only computations.

The methods described herein may be performed by software in machine- readable form on a tangible storage medium, e.g. in the form of a computer program comprising computer program code means adapted to perform all the steps of any of the methods described herein when the program is run on a computer and where the computer program may be embodied on a computer- readable medium. Examples of tangible storage media include computer storage devices comprising computer-readable media, such as disks, thumb drives, memory etc., and do not only include propagated signals. Propagated signals may be present in a tangible storage medium, but propagated signals per se are not examples of tangible storage media. The software can be suitable for execution on a parallel processor or a serial processor such that the method steps may be carried out in any suitable order, or simultaneously.

Although at least a portion of the subject matter has been described in language specific to structural features and/or acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as examples of implementing the claims and other equivalent features and acts are intended to be within the scope of the claims.

The invention has been described in conjunction with various embodiments herein. However, other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other devices or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfil the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems.

Claims

A network device for a wireless communication network, said network device comprising:

a transceiver configured to receive uplink information related to multiple subbands; and

a processor configured to assign a rank index over said multiple subbands, to compress the rank index and to cause the transceiver to send the rank index.

The network device according to claim 1 , wherein the processor is configured to determine a difference between the rank index and a previously sent rank index; and

cause the transceiver to send the difference.

The network device according to claim 1 or claim 2, wherein the processor is configured to:

determine a sequence of rank indexes of neighboring subbands within said multiple subbands;

encode the sequence of rank indexes to a sequence of differences between two consecutive rank indexes in the sequence of rank indexes, wherein the difference is one of the group of -1 , 0, +1 ;

decompose the sequence of differences between two consecutive rank indexes as a sum of two binary sequences, wherein the first binary sequence consists of differences 0 and -1 ; and the second binary sequence consists of differences 0 and +1 ; and

encode the first binary sequence and the second binary sequence with enumerative source encoding.

The network device according to claim 1 , wherein the processor is configured to:

determine encoding information bits comprising multiple predefined rank indexes and encoded bits corresponding to the predefined rank indexes; compare the rank index with the encoded bits;

determine the encoded bits corresponding to the rank index; and cause the transceiver to send the encoded bits.

The network device according to claim 1 or claim 4, wherein the processor is configured to:

generate encoding information bits;

encode, by using the encoding information bits, encoded bits

corresponding to the rank index; and

cause the transceiver to send the encoding information bits and the encoded bits.

A user equipment for a wireless network, comprising a processor and a transceiver, wherein the processor is configured to cause the transceiver to:

connect to the wireless network via multiple subbands;

send uplink information related to said multiple subbands; and

receive a compressed rank index over said multiple subbands; and wherein

the processor is configured to decompress the compressed rank index.

The user equipment according to claim 6, wherein the processor is configured to cause the transceiver to :

receive the compressed rank index as a difference between the rank index and a previously sent rank index; and wherein

the processor is configured to decompress the compressed rank index using the difference and the previously sent rank index.

The user equipment according to claim 6 or claim 7, wherein the processor is configured to:

decode a sequence of rank indexes from a first binary sequence and a second binary sequence encoded with enumerative source encoding; wherein between two consecutive rank indexes in the sequence of rank indexes, the difference is one of the group of -1 , 0, +1 ;

the first binary sequence consists of differences 0 and -1 ; and

the second binary sequence consists of differences 0 and +1 .

9. The user equipment according to claim 6, wherein the processor is

configured to cause the transceiver to:

receive encoding information bits comprising multiple predefined rank indexes and encoded bits corresponding to the predefined rank indexes; receive the encoded bits corresponding to the rank index; and

decompress the rank index by using the encoded bits and the encoding information bits.

10. A method, comprising:

a network device receiving uplink information from a user equipment, wherein the user equipment is connected to a wireless network via multiple subbands;

assigning a rank index over multiple subbands based on the uplink information;

compressing the rank index; and

sending the compressed rank index to the user equipment via the wireless network.

1 1 . The method according to claim 10, comprising the network device:

detecting a difference between the rank index and a previously sent rank index;

sending the difference to the user equipment; and

the user equipment calculating the rank index using the difference and the previously sent rank index.

12. The method according to claim 10 or claim 1 1 , comprising the network device:

receiving a sequence of rank indexes of neighboring subbands within said multiple subbands; encoding the sequence of rank indexes to a sequence of differences between two consecutive rank indexes in the sequence of rank indexes, wherein the difference is one of the group of -1 , 0, +1 ;

decomposing the sequence of differences between two consecutive rank indexes as a sum of two binary sequences, wherein the first binary sequence consists of differences 0 and -1 ; and the second binary sequence consists of differences 0 and +1 ; and

encoding the first binary sequence and the second binary sequence with enumerative source encoding.

13. The method according to claim 10, comprising the network device

determining encoding information bits comprising multiple predefined rank indexes and encoded bits corresponding to the predefined rank indexes;

the network device sending the encoded bits and the encoding information bits to the user equipment;

the user equipment receiving the encoded bits and the encoding information bits; and

determining the rank index by using the encoded bits and the encoding information bits.

14. The method according to claim 10 or claim 13, comprising the network device generating encoding information bits;

encoding, by using the encoding information bits, encoded bits corresponding to the rank index;

sending the encoding information bits and the encoded bits to the user equipment;

the user equipment receiving the encoding information bits and the encoded bits; and

determining the rank index by decoding the encoded bits with the encoding information bits.

15. A computer program product comprising instructions, which when the program is executed by a computer, cause the computer to carry out the steps of the method of claims 10 to 14.