Control Messages in Wireless Communication

Abstract

In LTE TDD with Carrier Aggregation (CA), a two bit sequence number or downlink assignment indicator (DAI) can be attached to downlink control information (DCI) messages relating to packets transmitted on some or all of successive carrier frequencies in a given subframe. The numerical progression of the sequence numbers can be used to determine the number of ACK/NACK bits (ACK/NACK codebook size) needed to acknowledge reception of the packets. The receiver determines the ACK/NACK codebook size to within a certain granularity or other constraints, depending on the received sequence of DAI values. The granularity or constraint(s) may be pre-determined, or configured or indicated by the sequence. This allows a more accurate and reliable common understanding between the transmitter and receiver concerning the number of DCI messages transmitted.

Claims

1. A wireless communication method comprising, at a transmitter: determining a number of control messages, the control messages meeting at least one criterion and for transmission in a sequence to a receiver; associating a sequence number with each control message in the sequence, wherein the sequence number of at least one control message in the sequence depends on a constraint applied to the number of acknowledgements to be transmitted from the receiver following reception of one or more of said number of control-messages; and transmitting the determined number of control messages and the associated sequence numbers; and, at a receiver: receiving one or more of said control messages and associated sequence numbers; determining the number of acknowledgements to be transmitted corresponding to said received control messages, wherein said determining of the number of acknowledgements to be transmitted is in dependence on said sequence numbers and said constraint applied to the number of said acknowledgements to be transmitted; and transmitting said determined number of acknowledgements.

2. The method according to claim 1 wherein the receiver is configured for wireless communication via a plurality of carriers and each control message relates to a respective carrier.

3. The method according to claim 1 wherein the at least one criterion employed in said determining comprises one or more of: the control message is transmitted within a defined time window; the control message is received within a defined time window; the control message is for downlink scheduling; the control message is transmitted with a particular message format; the control message is transmitted via a particular channel.

4. The method according to claim 3 wherein wireless communication is scheduled within frames each consisting of a succession of subframes, and wherein the at least one criterion comprises that the control message is for downlink scheduling relating to the same subframe.

5. The method according to claim 1 wherein the sequence number associated with successive control messages in the sequence is incremented by an amount in dependence on said constraint applied to the number of said acknowledgements.

6. The method according to claim 1 wherein determining the number of acknowledgements to be transmitted is in dependence on said sequence numbers comprises distinguishing a numerical progression of said sequence numbers.

7. The method according to claim 1 wherein, in said associating, the sequence number associated with successive control messages in the sequence is in dependence on the total number of control messages in the sequence.

8. The method according to claim 1 wherein said constraint is that the number of acknowledgements to be transmitted corresponds to the number of control messages being a multiple of N, wherein N is 2, 4 or 8.

9. The method according to claim 1 wherein said constraint is that the number of acknowledgements to be transmitted corresponds to the number of control messages being equal to A+BN, wherein A is 1, 2, 3 or 4, B is 2 or 4 and N is an integer greater than or equal to zero.

10. The method according to claim 1 wherein the numerical progression of the sequence numbers is given by:
SQI(i)=f(i,c)mod n where SQI is a sequence number; i identifies the control message in the sequence of control messages; c is the determined number of control messages in the sequence; f(i,c) is a function of i and c; and n is the number of different possible sequence numbers.

11. The method according to claim 1 further comprising, at the receiver: receiving control messages and associated sequence numbers from the transmitter; making a determination, on the basis of the sequence numbers, whether at least one control message in the sequence has been missed and how many control messages have been missed; and notifying the determination to the transmitter.

12. The method according to claim 1 wherein: the wireless communication method is applied to an LTE-based wireless communication system; the transmitter is a base station and the receiver is a terminal in the wireless communication system; the control messages are downlink control information and the sequence number is a downlink assignment index.

13. A transmitter in a wireless communication system configured to: determine a number of control messages, the control messages meeting at least one criterion and for transmission in a sequence to a receiver; associate a sequence number with each control message in the sequence, wherein the sequence number of at least one control message in the sequence depends on a constraint applied to the number of acknowledgements to be transmitted from the receiver following reception of one or more of said number of control messages; and transmit the determined number of control messages and the associated sequence numbers to the receiver.

14. A receiver in a wireless communication system, configured to: receive one or more of a number of control messages from a transmitter, wherein a sequence number is associated with each control message in the sequence and the sequence number of at least one control message in the sequence depends on a constraint applied to the number of acknowledgements to be transmitted from the receiver following reception of one or more of said number of control messages; determine the number of acknowledgements to be transmitted corresponding to said received control messages, in dependence on said sequence numbers and said constraint applied to the number of said acknowledgements to be transmitted; and transmit said determined number of acknowledgements.

15. A wireless communication system comprising the transmitter according to claim 13; and a receiver in a wireless communication system, configured to: receive one or more of a number of control messages from a transmitter, wherein a sequence number is associated with each control message in the sequence and the sequence number of at least one control message in the sequence depends on a constraint applied to the number of acknowledgements to be transmitted from the receiver following reception of one or more of said number of control messages; determine the number of acknowledgements to be transmitted corresponding to said received control messages, in dependence on said sequence numbers and said constraint applied to the number of said acknowledgements to be transmitted; and transmit said determined number of acknowledgements.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0119] Reference is made, by way of example only, to the accompanying drawings in which:

[0120] FIG. 1 illustrates a wireless communication system employing carrier aggregation (CA);

[0121] FIG. 2 illustrates channels at each of a plurality of protocol layers in LTE;

[0122] FIG. 3 shows how LTE physical channels are allocated to a PCell and SCell;

[0123] FIG. 4 is a flowchart of a process performed at a transmitter in embodiments of the present invention;

[0124] FIG. 5 is a flowchart of a process performed at a receiver in embodiments of the present invention;

[0125] FIG. 6 illustrates functional blocks of a terminal (UE) which may be employed in embodiments of the present invention; and

[0126] FIG. 7 illustrates functional blocks of base station equipment (eNB) which may be employed in embodiments of the present invention.

DETAILED DESCRIPTION

[0127] To recap some of the most relevant features from the introduction, a UE can receive a DL assignment with respect to any CC activated for that UE. The DL assignment is conveyed to the UE by a DCI message. Each DCI is transmitted on PDCCH or EPDDCH, and indicates the presence of one or more transport blocks on PDSCH, transmitted via a specific CC. The UE sends ACK/NACK for each received transport block of which it is notified via a DCI message.

[0128] In the simplest case there would be one ACK/NACK bit per DCI (assuming each DCI schedules, on a respective carrier, one data codeword or transport block to be transmitted in the downlink via PDSCH to a given UE). Another case would be where there are two ACK/NACK bits per DCI, such as when MIMO is used and assuming each DCI schedules two downlink codewords. In principle a mixed case would be possible, where some DCIs in a subframe schedule one codeword to a UE and some DCIs schedule two codewords. In this case a simplification might be applied at the UE to transmit two ACK/NACK bits for every DCI. Thus, in many cases the ACK/NACK codebook size is directly proportional to the number of DCIs.

[0129] However, in LTE TDD bundling may be required to fit the ACK/NACKs of a plurality of subframes with respect to the same CC, within the available number of ACK/NACK bits that can be transmitted in one subframe. Bundling is typically applied if the number of codewords which need to be acknowledged in a given subframe exceeds the number of ACK/NACK bits. The bundling means that one ACK/NACK bit acknowledges more than one codeword (which all need to have been received correctly for the bit to be ACK). This case typically arises in TDD where codewords received at the UE in a number of DL subframes all need to be acknowledged in the same UL subframe (e.g. with asymmetrical DL/UL subframe allocation with more DL subframes than UL subframes).

[0130] Such bundling can also be applied to simplify the mixed case mentioned above, where some DCIs in a subframe schedule one codeword to a UE and some DCIs schedule two codewords. The UE would transmit one ACK/NACK bit for every DCI, and bundling is applied for DCIs scheduling two codewords (i.e. only 1 ACK/NACK bit for two codewords).

[0131] The ACK/NACKs to be sent in the case of bundling are partly governed by downlink assignment information (DAI) informed to the UE by the eNB. The existing DAI can be viewed as a sequence number for a single CC which is incremented in a subframe where a DL assignment is transmitted to the UE. The DAI value is included in the DCI.

TABLE-US-00002 TABLE 1 2 bit DAI in LTE TDD Number of successive Most likely Approximate subframes uncorrectable probability of covered by a pattern of occurrence of given missed DCIs most likely ACK/NACK Sequence of 2 (where X uncorrectable message and Sequence of 2 bit bit DAI values indicates a pattern of with a DCI DAI values (binary) (decimal) missed DCI) missed DCIs 4 00, 01, 10, 11 0, 1, 2, 3 D, D, D, X 0.01 5 00, 01, 10, 11, 00 0, 1, 2, 3, 0 D, D, D, D, X 0.01 6 00, 01, 10, 11, 00, 01 0, 1, 2, 3, 0, 1 D, D, D, D, D, X 0.01 7 00, 01, 10, 11, 00, 01, 10 0, 1, 2, 3, 0, 1, 2 D, D, D, D, D, D, X 0.01

[0132] As a an example, if in given a set of subframes the UE receives a sequence of DAI values [0, 1, 2, 0, 1] with respect to a given carrier, it can deduce that at least one DCI has been missed. The shortest possible transmitted sequence that could have been sent is [0, 1, 2, 3, 0, 1]. In general the UE can detect and correct any sequences of up to 3 missed DCI in succession. A special case is where the last DCI of a sequence is missed (indicated by X in Table 1), and the UE would not be able to identify this. Note that the UE also knows the number of subframes that have elapsed and this information can help in the correction process, but for simplicity this aspect is not considered in the above example.

[0133] Assuming that the probability of a failing to detect a DCI message is fixed, and corresponds to some process which is uniform and uncorrelated between subframes and CCs, we can ascribe this a value such as p=0.01 as a typical design assumption. Then we can approximate the probability of occurrence of missing only the last DCI in a sequence as being approximately p also (as an illustration of this, for 4 subframes each with a DCI, the probability of three correct DCIs followed by one missed DCI would be P=(1p)(1p)(1p)p=0.990.990.990.01 which is approximately 0.01).

[0134] This case is the uncorrectable pattern with the smallest number of missed DCI. All the other uncorrectable patterns contain more missed DCI and so are much less likely. Thus the probability that a given sequence of DCI has one or more missing DCI, and cannot be corrected is also about p (i.e. about 0.01).

[0135] Note that here we consider a pattern of missed DCI to be correctable if at least the correct number of DL assignments can be identified (even if it might not be possible to deduce exactly which assignments were lost). It will be understood that the missed DCI are correctable in the sense that the eNB can be notified of the problem, allowing the DCI in question to be retransmitted.

[0136] A mechanism to detect loss of the last DCI in a sequence is proposed in 3GPP R1-152569 HARQ-ACK transmission for up to 32 CCs, CATT, 3GPP TSG-RAN WG1 Meeting #81 Fukuoka, Japan, 25-29 May 2015.

[0137] An additional bit is added to the DAI field and set to 1 for the last DCI in a time-domain sequence (otherwise set to 0). This solves the problem, but at the cost of an additional bit of overhead per DCI.

[0138] The invention is based on the recognition that in LTE CA the total number of DCIs transmitted to a UE on different CCs in a subframe is known at the transmitter. This information can be used to improve the robustness of the DAI scheme. In contrast, the number of packets to be transmitted in successive subframes in the time domain is not necessarily known in advance.

[0139] In the prior art the DAI value in a given DCI depends on the DAI in the previous DCI message.

[0140] In embodiments of this invention, the DAI value also depends on the total number of DCI messages in a sequence of DCI messages. Thus, in general the sequence of DAI values attached to successive DCIs in the frequency domain depends on the number of DCIs transmitted to the UE in that subframe.

[0141] Specific embodiments of the invention can be used to improve the reliability of the understanding between the transmitter and receiver of the number of DCIs transmitted, and the number of ACK/NACK payload bits to be transmitted by the receiver. As a particular example, for 2 bits of DAI, the ACK/NACK payload corresponds to the bits for a number of DCIs constrained to be a multiple of 4. Thus the granularity of the ACK/NACK payload size (or in other words, the codebook size) can be constrained to a particular value. As a further example the DAI values in a sequence may be incremented by an amount which indicates a constraint on, or the granularity of, the ACK/NACK codebook size (e.g. a multiple of 2 or 4) to the receiver.

[0142] For the general case the DAI value is given by

DAI(i)=f(i,c)mod nEqn 1

where
i identifies the DCI in the sequence of DCIs
c is the total number of DCIs in the sequence
f(i,c) is a function of i and c
n is the number of different DAI values.

[0143] Note that here, whether DCIs are successive in the frequency domain refers to the resources (e.g. CCs) to which the DCI refer, which may or may not be the same as the frequency relationship of the transmitted DCI themselves. More generally, the ordering need not be in terms of frequency provide that the UE and eNB have a common understanding of the order.

[0144] The basic process as performed at the transmitter (base station) is shown in FIG. 4.

[0145] Step S100 is a step of scheduling the downlink for a given UE in a given subframe, using a plurality of CCs configured for the UE.

[0146] In S102 the base station determines how many DCI (control messages) in total are needed to inform the UE of the DL assignment.

[0147] S104 is to calculate sequence numbers for each DCI, based on the total number of DCI just determined and based on the constraint(s) or granularity as described below in various embodiments.

[0148] In step S106 the base station attaches the sequence numbers to DCI and transmits the resulting control messages to the UE. The process is repeated for the next subframe in which the base station has a DL assignment for the UE on any of the CCs.

EMBODIMENTS

[0149] The embodiments are based on LTE.

[0150] The receiver can use the received DAI values to determine the size of the ACK/NACK codebook constrained with a given granularity of X, i.e. the number of ACK/NACK bits corresponding to a number of DCIs is constrained to be a multiple of X. Here, typical values for X would be 2 or 4. Effectively, the number of ACK/NACK bits is rounded up to correspond to a multiple of X DCIs. According to an embodiment applied to a 2 bit DAI with X=4 (least for 4 or more DCIs) up to three successive DCIs can be missed before the transmitter and receiver would have a misunderstanding over the size of the ACK/NACK codebook.

[0151] It might be thought desirable to have an ACK/NACK codebook size always exactly corresponding to the number of scheduled DCIs (i.e. X=1), but this fine granularity may not always be useful in practice. For example, there may be a limited set of ACK/NACK payload sizes available for transmission. Therefore sacrificing some granularity for increased robustness may be acceptable.

[0152] Note that in LTE a DCI may schedule a transmission of 1 or 2 codewords, so with 1 ACK/NACK bit per codeword, the corresponding number of ACK/NACK bits may be 1 or 2. However, for the purpose of this invention we assume that the number of ACK/NACK bits per CC is known to both transmitter and receiver.

First Embodiment

[0153] A first embodiment is based on LTE where a two bit DAI field is included in each DCI.

[0154] For a set of ordered carriers on which DCI is transmitted, the DAI is incremented for successive DCIs, starting with a pre-determined value such as zero. The receiver assumes that the ACK/NACK codebook size is a multiple of 2. The ACK/NACK codebook size corresponds to the number of ACK/NACK bits to be sent in acknowledgement of received DCIs (control channel messages) plus DCIs assumed by the receiver to have been transmitted but not received.

[0155] For this embodiment, in Eqn 1

f(i,c)=i

so DAI (i)=i mod n, where for example n=4 for a 2-bit DAI field.

[0156] The ordering of the set of carriers may be based on frequency, but this is not essential. Some examples of the DAI sequences for the first embodiment are shown in Table 2. It will be noted that, in contrast to Table 1 above, the case of DCIs for a set of carriers and not just one carrier is being considered here. Note that there is no restriction on the number of successive DCIs for which the invention may be applied.

TABLE-US-00003 TABLE 2 Improved 2 bit DAI for LTE CA (assuming codebook size is a multiple of two) Main error case Sequence of ACK/NACK (leading to Probability Number of 2 bit DAI codebook wrong of the successive Sequence of 2 bit values size (no of codebook main error DCIs DAI values (binary) (decimal) DCIs) size) case 1 00 0 2 The DCI is 0.01 missed 2 00, 01 0, 1 2 Both DCI's 10.sup.4 are missed 3 00, 01, 10 0, 1, 2 4 The last 0.01 DCI is missed 4 00, 01, 10, 11 0, 1, 2, 3 4 The last 10.sup.4 two DCIs are missed 5 00, 01, 10, 11, 00 0, 1, 2, 3, 0 6 The last 0.01 DCI is missed 6 00, 01, 10, 11, 00, 01 0, 1, 2, 3, 0, 1 6 The last 10.sup.4 two DCIs are missed 7 00, 01, 10, 11, 00, 01, 10 0, 1, 2, 3, 0, 1, 2 8 The last 0.01 DCI is missed

[0157] Table 2 shows the probability of the main error cases, and it can be seen that there is generally a significant improvement over the prior art result in Table 1. As already mentioned, the ACK/NACK codebook in terms of the number of ACK/NACK bits is equivalent to (proportional to) the number of DCIs. In a system such as LTE there may be one or two ACK/NACK bits per received DCI. In the Tables the ACK/NACK codebook size is expressed as the number of DCIs.

[0158] In Table 2 and the other Tables, error cases are included where no DCIs are received. In a system such as LTE this would result in no transmission of ACK/NACK bits, but we have included it here as special case of the wrong codebook size.

[0159] In a variation of the first embodiment the receiver assumes that the ACK/NACK codebook size is a multiple of 4.

TABLE-US-00004 TABLE 3 Improved 2 bit DAI for LTE CA (assuming codebook size is a multiple of four) Main error case Sequence of 2 (leading to Probability Number of bit DAI ACK/NACK wrong of the successive Sequence of 2 bit values codebook size codebook main error DCIs DAI values (binary) (decimal) (no of DCIs) size) case 1 00 0 4 The DCI is 0.01 missed 2 00, 01 0, 1 4 Both DCI's 10.sup.4 are missed 3 00, 01, 10 0, 1, 2 4 All three 10.sup.6 DCIs are missed 4 00, 01, 10, 11 0, 1, 2, 3 4 All four 10.sup.8 DCIs are missed 5 00, 01, 10, 11, 00 0, 1, 2, 3, 0 8 The last DCI 0.01 is missed 6 00, 01, 10, 11, 00, 01 0, 1, 2, 3, 0, 1 8 The last two 10.sup.4 DCIs are missed 7 00, 01, 10, 11, 00, 01, 10 0, 1, 2, 3, 0, 1, 2 8 The last 10.sup.6 three DCIs are missed

[0160] Table 3 shows a further improvement compared with Table 2 but at the cost of coarser granularity in the codebook size.

[0161] FIG. 5 shows a flowchart of steps performed at the UE in this embodiment. In step S200, the UE receives the control messages from the base station in which DCI have sequence numbers attached in the form of the 2-bit DAI values, and obtains a list of DAI values from the corresponding detected DCIs. The UE checks whether any DAI values are missing from the sequence in the list, which the UE can judge by comparing the DAI values with the sequences in Table 2, for example. If no DAI value is found to be missing (S200, NO) the UE proceeds to step S204. If a missing DAI value is detected, in S202 the UE adds the missing DAI value or values to the list. In step S204 the UE checks whether the first DAI value in the sequence is equal to the correct value pre-determined, in this example, zero. If so (S204, YES) the flow proceeds to S208. Otherwise, (S204, NO) this indicates that at least one DAI value is missing, and the missing value is added to the start of the list in S206. In step S208, the UE checks whether the number of DAIs in the resulting list corresponds to a valid number of ACK/NACK bits, bearing in mind the granularity referred to above. If the number of DAIs is a valid number (S208, YES) the flow proceeds to S212. Otherwise, (S208, NO), the UE adds a DAI value at the end of the list of DAI values in step S210 before proceeding to S212. In step S212, the UE determines the ACK/NACK bit values to be transmitted, taking into account the detected DCIs (data packets actually received) and the list of DAI values (including the DAIs missing from the sequence, which correspond to non-received DCIs). In step 214, after any necessary coding of the ACK/NACK bits, the UE sends the appropriate ACK/NACKs to the eNB. The UE waits until a new set of ACK/NACKs need to be transmitted and the process then returns to the start.

[0162] It should be noted that in the above procedure, when missing values are added to the list they can be marked in some way so that the DAIs actually received can be distinguished from those which are detected as missing.

[0163] In a further variation of this embodiments (which may also be applied to the other embodiments), as a special case, if only one DCI is received a different transmission format for the ACK/NACK feedback may be used, which can carry just one or two bits.

Second Embodiment

[0164] The second embodiment is like the first embodiment in that the DAI is incremented for successive DCIs, however, the starting value is zero when the number of DCIs is even and 1 when the number of DCIs is odd. The receiver assumes that the ACK/NACK codebook size is a multiple of two.

TABLE-US-00005 TABLE 4 Improved 2 bit DAI for LTE CA with codebook size a multiple of 2 and starting point of the sequence depending on the number of DCIs Main error case Sequence of 2 (leading to Probability Number of bit DAI ACK/NACK wrong of the successive Sequence of 2 bit values codebook size codebook main error DCIs DAI values (binary) (decimal) (no of DCIs) size) case 1 01 1 2 The DCI is 0.01 missed 2 00, 01 0, 1 2 Both DCI's 10.sup.4 are missed 3 01, 10, 11 1, 2, 3 4 The last two 10.sup.4 DCIs are missed 4 00, 01, 10, 11 0, 1, 2, 3 4 All four 10.sup.8 DCIs are missed 5 01, 10, 11, 00, 01 1, 2, 3, 0, 1 6 The last two 10.sup.4 DCIs are missed 6 00, 01, 10, 11, 00, 01 0, 1, 2, 3, 0, 1 6 The last two 10.sup.4 DCIs are missed 7 01, 10, 11, 00, 01, 10, 11 1, 2, 3, 0, 1, 2, 3 8 The first 10.sup.6 three DCIs are missed

[0165] The provision of different starting points for the sequence depending on the number of DCIs provides a further improvement as shown in Table 4.

[0166] In a variation of the second embodiment the receiver assumes that the ACK/NACK codebook size is a multiple of four and the starting value DAI is zero except when the number of DCIs is 1 plus a multiple of four, when the starting value is 1.

TABLE-US-00006 TABLE 5 Improved 2 bit DAI for LTE CA with codebook size a multiple of 4 and starting point of the sequence depending on the number of DCIs Main error case Sequence of 2 (leading to Probability Number of bit DAI ACK/NACK wrong of the successive Sequence of 2 bit values codebook size codebook main error DCIs DAI values (binary) (decimal) (no of DCIs) size) case 1 01 1 4 The DCI is 0.01 missed 2 00, 01 0, 1 4 Both DCI's 10.sup.4 are missed 3 00, 01, 10 0, 1, 2 4 All three 10.sup.6 DCIs are missed 4 00, 01, 10, 11 0, 1, 2, 3 4 All four 10.sup.8 DCIs are missed 5 01, 10, 11, 00, 01 1, 2, 3, 0, 1 8 The last two 10.sup.4 DCIs are missed 6 00, 01, 10, 11, 00, 01 0, 1, 2, 3, 0, 1 8 The last two 10.sup.4 DCIs are missed 7 00, 01, 10, 11, 00, 01, 10 0, 1, 2, 3, 0, 1, 2 8 The last 10.sup.6 three DCIs are missed

Third Embodiment

[0167] The third embodiment is like the first embodiment except that from the UE perspective, the counting increment indicates information about the codebook size.

[0168] In this example, the increment indicates whether the codebook size is a multiple of 2 (increment by 1 up) or 4 (increment by 3). In other words the progression in the sequence numbers indicates the granularity of the codebook size to the receiver. The pre-determined starting DAI value is zero.

[0169] As a special case with a 2 bit DAI, because of the mod 4 operation, a counting increment of 1 leads to counting up and a counting increment of 3 leads to counting down.

TABLE-US-00007 TABLE 6 Improved 2 bit DAI for LTE CA with DAI counting increment indicating whether the codebook size is a multiple of 2 or 4 Main error case Sequence of 2 (leading to Probability Number of bit DAI ACK/NACK wrong of the successive Sequence of 2 bit values codebook size codebook main error DCIs DAI values (binary) (decimal) (no of DCIs) size) case 1 00 0 2 The DCI is 0.01 missed 2 00, 01 0, 1 4 Both DCI's 10.sup.4 are missed 3 00, 11, 10 0, 3, 2 4 The last two 10.sup.4 DCIs are missed 4 00, 11, 10, 01 0, 3, 2, 1 4 The last 10.sup.6 three DCIs are missed 5 00, 01, 10, 11, 00 0, 1, 2, 3, 0 6 At least ~10.sup.6 three DCIs are missed 6 00, 01, 10, 11, 00, 01 0, 1, 2, 3, 0, 1 6 At least four ~10.sup.8 DCIs are missed 7 00, 11, 10, 01, 00, 11, 10 0, 3, 2, 1, 0, 3, 2 8 The last 10.sup.6 three DCIs are missed

[0170] In a variation of the third embodiment with a 2 bit DAI the counting increment indicates whether the codebook size is a multiple of 4 (increment 1) or 8 (increment 3). The reliability of determining the codebook size is improved, at the cost of coarser granularity.

TABLE-US-00008 TABLE 7 Improved 2 bit DAI for LTE CA with DAI counting increment indicating whether the codebook size is a multiple of 4 or 8 Main error case Sequence of 2 (leading to Probability Number of bit DAI ACK/NACK wrong of the successive Sequence of 2 bit values codebook size codebook main error DCIs DAI values (binary) (decimal) (no of DCIs) size) case 1 00 0 4 The DCI is 0.01 missed 2 00, 01 0, 1 4 Both DCI's 10.sup.4 are missed 3 00, 01, 10 0, 1, 2 4 All three 10.sup.6 DCIs are missed 4 00, 01, 10, 11 0, 1, 2, 3 4 All four 10.sup.8 DCIs are missed 5 00, 11, 10, 01, 00 0, 3, 2, 1, 0 8 At least ~10.sup.6 three DCIs are missed 6 00, 11, 10, 01, 00, 11 0, 3, 2, 1, 0, 3 8 At least four ~10.sup.8 DCIs are missed 7 00, 11, 10, 01, 00, 11, 10 0, 3, 2, 1, 0, 3, 2 8 At least four ~10.sup.8 DCIs are missed

[0171] In a further variation of the third embodiment with a 3 bit DAI the counting increment indicates the ACK/NACK codebook size as follows:

TABLE-US-00009 TABLE 8 Counting increment indicates ACK/NACK codebook size Counting increment Codebook size 1 1 + 4N 3 2 + 4N 5 3 + 4N 7 4 + 4N

[0172] In Table 8, N is an integer {0, 1, 2, 3} and the DAI value is modulo 8.

[0173] In order to better distinguish between potential error cases, the starting value of the DAI counter is set to the value of the counting increment.

[0174] The reliability of determining the codebook size is improved, with fine granularity, at the cost of an additional DAI bit.

TABLE-US-00010 TABLE 9 Improved 3 bit DAI for LTE CA with DAI counting increment indicating codebook size for 3 DAI bits Main error case Number Sequence of ACK/NACK (leading Probability of 3 bit DAI codebook to wrong of the successive Sequence of 3 bit DAI values values size (no of codebook main error DCIs (binary) (decimal) DCIs) size) case 1 001 1 1 The DCI 0.01 is missed 2 011, 110 3, 6 2 Both 10.sup.4 DCI's are missed 3 101, 010, 111 5, 2, 7 3 All three 10.sup.6 DCIs are missed 4 111, 110, 101, 100 7, 6, 5, 4 4 At least ~10.sup.6 three DCIs are missed 5 001, 010, 011, 100, 101 1, 2, 3, 4, 5 5 At least ~10.sup.8 four DCIs are missed 6 011, 110, 001, 100, 111, 010 3, 6, 1, 4, 7, 2 6 At least ~10.sup.8 four DCIs are missed 7 101, 010, 111, 100, 001, 110, 011 5, 2, 7, 4, 1, 6, 3 7 At least ~10.sup.8 four DCIs are missed

[0175] The results in Table 9 show the best performance, along with achieving fine codebook granularity, and this is the preferred embodiment for a 3 bit DAI.

Fourth Embodiment

[0176] The fourth embodiment is like the first embodiment except that an extra signalling bit indicates whether the ACK/NACK codebook size is a multiple of 2 or 4. Although the additional bit increases the overhead, there may be a benefit in terms of increased robustness.

TABLE-US-00011 TABLE 10 Improved 3 bit DAI for LTE CA with an extra DAI bit indicating whether the codebook size is a multiple of 2 or 4 Main error case Number Sequence of ACK/NACK (leading Probability of 3 bit DAI codebook to wrong of the successive Sequence of 3 bit DAI values values size (no of codebook main error DCIs (binary) (decimal) DCIs) size) case 1 100 4 2 The DCI 0.01 is missed 2 100, 101 4, 5 2 Both 10.sup.4 DCI's are missed 3 000, 001, 010 0, 1, 2 4 All three 10.sup.6 DCIs are missed 4 000, 001, 010, 011 0, 1, 2, 3 4 All four 10.sup.8 DCIs are missed 5 100, 101, 110, 111, 100 4, 5, 6, 7, 4 6 The last 10.sup.6 three DCIs are missed 6 100, 101, 110, 111, 100, 101 4, 5, 6, 7, 4, 5 6 At least ~10.sup.8 four DCIs are missed 7 000, 001, 010, 011, 000, 001, 010 0, 1, 2, 3, 0, 1, 2 8 At least ~10.sup.8 four DCIs are missed

[0177] In a variation of the fourth embodiment the extra signalling bit indicates whether the ACK/NACK codebook size is a multiple of 4 or 8. There is a further benefit in terms of increased robustness, at the cost of coarser granularity in the codebook size.

TABLE-US-00012 TABLE 11 Improved 3 bit DAI for LTE CA with an extra DAI bit indicating whether the codebook size is a multiple of 4 or 8 Main error case Number Sequence of ACK/NACK (leading Probability of 3 bit DAI codebook to wrong of the successive Sequence of 3 bit DAI values values size (no of codebook main error DCIs (binary) decimal DCIs) size) case 1 000 0 4 The DCI 0.01 is missed 2 000, 001 0, 1 4 Both 10.sup.4 DCI's are missed 3 000, 001, 010 0, 1, 2 4 All three 10.sup.6 DCIs are missed 4 000, 001, 010, 011 0, 1, 2, 3 4 All the 10.sup.8 DCIs are missed 5 100, 101, 110, 111, 100 4, 5, 6, 7, 4 8 All the .sup.10.sup.10 DCIs are missed 6 100, 101, 110, 111, 100, 101 4, 5, 6, 7, 4, 5 8 All the .sup.10.sup.12 DCIs are missed 7 100, 101, 110, 111, 100, 101, 110 4, 5, 6, 7, 4, 5, 6 8 All the .sup.10.sup.14 DCIs are missed

[0178] FIG. 6 is a block diagram illustrating an example of a UE 1 to which the present invention may be applied. The UE 1 may include any type of device which may be used in a wireless communication system described above and may include cellular (or cell) phones (including smartphones), personal digital assistants (PDAs) with mobile communication capabilities, laptops or computer systems with mobile communication components, and/or any device that is operable to communicate wirelessly. The UE 1 includes transmitter/receiver unit(s) 804 connected to at least one antenna 802 (together defining a communication unit) and a controller 806 having access to memory in the form of a storage medium 808. The controller 806 may be, for example, a microprocessor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other logic circuitry programmed or otherwise configured to perform the various functions described above, for example to determine the ACK/NACK codebook size and detect and identify missing DAI values in the manner shown in FIG. 5. For example, the various functions described above may be embodied in the form of a computer program stored in the storage medium 808 and executed by the controller 806. The transmission/reception unit 804 is arranged, under control of the controller 806, to receive control messages from the cells and send ACK/NACK and so forth as discussed previously.

[0179] FIG. 7 is a block diagram illustrating an example of equipment suitable for use as an eNB 11 and at least one of the base stations 21, 22, . . . 2n referred to above (and connected by high-speed backhaul to any other of the base stations). It includes transmitter/receiver unit(s) 904 connected to at least one antenna 902 (together defining a communication unit) and a controller 906. The controller may be, for example, a microprocessor, DSP, ASIC, FPGA, or other logic circuitry programmed or otherwise configured to perform the various functions described above, including scheduling a DL assignment for each UE and constructing DCI accordingly, including forming sequence numbers such as the DAI values shown in Tables 2 to 4. For example, the various functions described above may be embodied in the form of a computer program stored in the storage medium 908 and executed by the controller 906. The transmission/reception unit 904 is responsible for transmission of the control messages and so on under control of the controller 906. The controller 906 not only manages any integrated base station such as 21 in FIG. 1 but also manages any separate base stations not directly controlled by the base station equipment.

[0180] Thus, to summarise, an embodiment of the present invention may provide that in a system such as LTE, a sequence number or downlink assignment indicator (DAI) can be attached to downlink control information (DCI) messages relating to packets transmitted on some or all of successive carrier frequencies in a given subframe. This can be used to determine the number of ACK/NACK bits (ACK/NACK codebook size) needed to acknowledge reception of the packets. The receiver determines the ACK/NACK codebook size to within a certain granularity or subject to other constraints, or in other words as one of a set of fixed sizes, depending on the received sequence of DAI values. The granularity or constraint(s) may be pre-determined, configured or indicated by the sequence.

[0181] The received sequence of DAI values can be used to detect at least some cases of false detection of DCI messages (e.g. when a DCI message is detected by the UE where none was transmitted). For example if a DCI is received with a DAI that does not fit in the sequence it can be rejected.

[0182] Various modifications are possible within the scope of the present invention.

[0183] As is clear from the fourth embodiment, the invention is not limited to DAI with 2 bits, and can be applied with more than two bits. Similarly the invention is not limited to the example cases or sequence lengths shown in Tables 2 to 4.

[0184] Whilst embodiments have been described with respect to DCI, this is not the only possibility. More generally the present invention can be applied to control messages of any particular type, in other words which share one or more criteria such as: [0185] transmitted within a defined time window; [0186] received within a defined time window; [0187] for downlink scheduling; [0188] transmitted with a particular message format; [0189] transmitted via a particular channel.

[0190] The invention as described is applied to control messages and incrementing a DAI value to count successive control messages. However, it can also be applied where a DAI value counts data codewords. This would be particularly relevant where the number of data codewords scheduled by a single DCI for a particular carrier can vary (e.g. either 1 or 2 codewords may be transmitted in the downlink, with a corresponding 1 or 2 ACK/NACK bits to be transmitted in the uplink).

[0191] In a system like LTE, when a large number of CCs are configured, there is a significant probability of false reception of a control channel message by a UE. Depending on the DAI value derived from a falsely detected control channel message scheduling a downlink data transmission, this could lead to an incorrect ACK/NACK codebook size being used by the UE. In order to reduce the probability of this occurring, some mitigation measures are desirable. For example, for a falsely detected DCI, the DAI value is unlikely to fit the DAI sequence progression expected by the UE. This would therefore be very likely to be interpreted as indicating the loss of several DCIs, leading to an assumed ACK/NACK codebook size which is too great, and a corresponding transmission of an incorrect number of ACK/NACK bits. This might lead to transmission of an incorrect ACK/NACK payload to the eNodeB, use of the wrong message format, or transmission in the wrong resources, causing interference. In order to reduce the probability of such events, the ACK/NACK transmission by the UE could be dropped if too large a number of missing DCIs is assumed or deduced by the UE (e.g. more than 2). The threshold on the number of missed DCIs before the ACK/NACK transmission is dropped could be: [0192] Predetermined [0193] Configured by signalling from the eNodeB [0194] Implicitly determined by the number of configured or activated CCs (e.g. a higher threshold for a large number of CCs) [0195] Implicitly determined by the number or proportion of ACK/NACK bits indicating ACK (e.g. if only a few bits would indicate ACK, then the threshold could be lower) [0196] Implicitly determined by the number of DCIs detected (e.g. a higher threshold for a larger number of detected DCIs)

[0197] A combination of the above could also be used to determine whether the ACK/NACK transmission is dropped.

[0198] Any of the embodiments and variations mentioned above may be combined in the same system. Whilst the above description has been made with respect to LTE and LTE-A, the present invention may have application to other kinds of wireless communication system also. Accordingly, references in the claims to terminal are intended to cover any kind of subscriber station, mobile device, MTC device and the like and are not restricted to the UE of LTE.

[0199] In any of the aspects or embodiments of the invention described above, the various features may be implemented in hardware, or as software modules running on one or more processors. Features of one aspect may be applied to any of the other aspects.

[0200] The invention also provides a computer program or a computer program product for carrying out any of the methods described herein, and a computer readable medium having stored thereon a program for carrying out any of the methods described herein.

[0201] A computer program embodying the invention may be stored on a computer-readable medium, or it may, for example, be in the form of a signal such as a downloadable data signal provided from an Internet website, or it may be in any other form.

[0202] It is to be understood that various changes and/or modifications may be made to the particular embodiments just described without departing from the scope of the claims.

INDUSTRIAL APPLICABILITY

[0203] In a wireless communication system where sequences of control messages such as DCI are sent between a transmitter and receiver, embodiments of the present invention allow with a 2-bit DAI for example, detection of the loss of any three successive DCI (not including the last DCI). A specific improvement over prior art is that the loss of up to three of the last DCIs in the sequence can be detected and corrected without adding any more bits. This allows a more accurate and reliable common understanding between the transmitter and receiver concerning the number of DCI messages transmitted and the corresponding ACK/NACK information. This is useful, for example, if the format of transmission of ACK/NACK information relating to the DCI messages depends on the number of messages transmitted, since the most likely cases of failed detection of a DCI message can be identified at the receiver.