WO2018207002A1

WO2018207002A1 - Method and apparatus for probabilistic dtx detection in a communication network

Info

Publication number: WO2018207002A1
Application number: PCT/IB2017/052744
Authority: WO
Inventors: Kevin Luo; Akram Bin Sediq; Jianguo Long; Yongquan Qiang
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2017-05-10
Filing date: 2017-05-10
Publication date: 2018-11-15

Abstract

Systems and methods for probabilistic Discontinuous Transmission (DTX) detection are provided. In some embodiments, a method of operation of a radio access node includes transmitting a control signal to a wireless device scheduling a data signal and transmitting the scheduled data signal. The method also includes detecting an ambiguous state of reception of the control signal by the wireless device based on a feedback of reception of the data signal. The method also includes determining whether a negative acknowledgment (NACK) or DTX is more likely to have occurred based on the feedback of reception and previous feedbacks from the wireless device and updating a determination of whether a NACK or DTX is more likely to have occurred for the previous feedbacks. In this way, more accurate link adaptation and/or retransmission decisions can be made. This could increase the spectral efficiency and data rate to the wireless device.

Description

METHOD AND APPARATUS FOR PROBABILISTIC DTX DETECTION IN A

COMMUNICATION NETWORK

Technical Field

[0001] The present disclosure relates to a control signal outer-loop adjustment for the control signal link adaptation (LA) in a cellular communications network.

Background

[0002] 3^rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) and LTE-Advanced standards have been developed to meet growing capacity demands due to rapid expansion of wireless data services. One challenging aspect of these standards is the optimal usage of limited radio resources shared by multiple wireless devices. Specifically, the Physical Downlink Shared Channel (PDSCH) is designed to carry downlink data, and the Physical Uplink Shared Channel (PUSCH) is designed to carry uplink data, while the Physical Downlink Control Channel (PDCCH) is designed to carry Downlink Control Information (DCI) in each subframe to provide wireless devices with necessary scheduling information in terms of resource allocation, modulation and coding scheme information, and/or power level information for proper downlink data reception and/or uplink data transmission. The terms DCI and PDCCH are sometimes used interchangeably when there is no confusion.

[0003] In current LTE standards, there are several DCI formats including DCI formats 0, 3, and 4 for Uplink (UL) and DCI formats 1 and 2 for Downlink (DL). A DCI carrying DL scheduling information is also called a DL assignment and a DCI carrying UL scheduling information a UL grant. The information of each DCI is rate matched and scrambled with a cell-specific and slot-specific scrambling sequence. One wireless device could have one or more DCIs in the same subframe. Each DCI is carried on one or more control channel elements (CCEs) depending on DCI length and DL radio channel condition. The number of CCEs used is often called the CCE aggregation level, which can be 1 , 2, 4, or 8. An aggregation level larger than 1 means that the DCI payload is encoded over more than one CCE, resulting in a low code rate, which is often needed for wireless devices in poor radio channel conditions.

[0004] PDCCH Link Adaptation (LA) intends to choose an optimal CCE aggregation level and power for each DCI and for each wireless device based on the DL channel condition of the wireless device. If the channel condition is good, a small number of CCEs (a low CCE aggregation level) and/or a low transmit power may be used. Otherwise, a large number of CCEs and a high transmit power may be used. The number of control symbols available to be used for PDCCH is limited. As such, the number of available CCEs for each subframe, which are shared by all the wireless devices serviced by a network node, is also limited. That means the performance of PDCCH LA may greatly impact the LTE network performance by affecting factors such as capacity and the number of wireless devices served by a network node.

[0005] As an example, in the case of Voice over Internet Protocol (VoIP), which demands a large number of DCIs, PDCCH capacity may be a key limiting factor for VoIP capacity. If PDCCH LA is too aggressive by using a small number of CCEs for each wireless device and/or a low transmit power for each wireless device in order to support as many wireless devices as possible within each subframe, wireless devices may have more PDCCH decoding failures, meaning some wireless devices may fail to locate the related DL data sent through the PDSCH or may miss UL grants for PUSCH transmission. This may result in significant throughput reduction and/or reduced user satisfaction. On the other hand, if PDCCH LA is too conservative by using a large number of CCEs or a high transmit power for each wireless device, the number of wireless devices that can be accommodated within each subframe will be smaller, which may lead to a low VoIP capacity, which is especially unacceptable in VoIP applications. As such, good PDCCH LA design is important.

[0006] The DL channel condition used in the PDCCH LA for a wireless device is based on the Channel Quality Indicator (CQI), which is determined by the wireless device and reported to the network node through UL channels such as PUSCH or Physical Uplink Control Channel (PUCCH). The network node will use CQI reports to estimate Signal-to-lnterference-plus-Noise ratio (SINR), which, together with a target PDCCH Block Error Rate (BLER), is used to determine PDCCH LA. This is referred to as pure CQI report based PDCCH LA. Accurate and timely CQI reports will help the network node adjust the CCE aggregation level and transmit power. Unfortunately, accurate and timely CQI reports may be difficult to obtain due to these limitations: (a) CQI reporting cannot be too frequent, as its reporting interval is limited by signaling overhead; (b) CQI reporting accuracy may vary from one wireless device to another depending on wireless device specific implementation; (c) often, each wireless device derives its CQI by checking cell-specific reference signals, which may not necessarily take into account the interference on PDCCH regions or PDSCH resource blocks. As such, there is a strong need for an additional adjustment on the CQI reported from the wireless device. This additional adjustment is referred to as an outer-loop adjustment. The outer-loop adjustment done for control signal link adaptation, e.g., for the PDCCH LA, is referred to as control signal outer-loop adjustment. Similarly, there is also an outer-loop adjustment done for the data signal link adaptation, e.g., for the PDSCH LA, and that is referred to as data signal outer-loop adjustment. Summary

[0007] Systems and methods for probabilistic Discontinuous Transmission (DTX) detection are provided. In some embodiments, a method of operation of a radio access node in a cellular communications network includes transmitting a control signal to a wireless device scheduling a data signal and transmitting the scheduled data signal to the wireless device according to the transmitted control signal. The method also includes detecting an ambiguous state of reception of the control signal by the wireless device based on a feedback of reception of the data signal. The method also includes determining whether a negative acknowledgment (NACK) or DTX is more likely to have occurred based on the feedback of reception of the data signal and one or more previous feedbacks from the wireless device and updating a determination of whether a NACK or DTX is more likely to have occurred for the previous feedbacks. In this way, more accurate link adaptation and/or retransmission decisions can be made. This could increase the spectral efficiency and data rate to the wireless device.

[0008] In some embodiments, the method also includes, in response to determining whether a NACK or a DTX is more likely to have occurred, updating a Link Adaptation (LA) parameter used to choose a coding scheme and/or power level for transmission of information in the control signal.

[0009] In some embodiments, a NACK is more likely to have occurred and updating the LA parameter used to choose the coding scheme and/or power level includes incrementing the LA parameter by a predetermined amount.

[0010] In some embodiments, a DTX is more likely to have occurred and updating the LA parameter used to choose the coding scheme and/or power level includes decrementing the LA parameter by a predetermined amount.

[0011] In some embodiments, determining whether a NACK or a DTX is more likely to have occurred includes determining whether a NACK or a DTX is more likely to have occurred using maximum a posteriori probability (MAP) detection.

[0012] In some embodiments, a radio access node includes a processor and a memory coupled to the processor. The memory contains instructions executable by the processor whereby the radio access node is operative to transmit a control signal to a wireless device scheduling a data signal and transmit the scheduled data signal to the wireless device according to the transmitted control signal. The memory also contains instructions executable by the processor whereby the radio access node is operative to detect an ambiguous state of reception of the control signal by the wireless device based on a feedback of reception of the data signal. The memory also contains instructions executable by the processor whereby the radio access node is operative to determine whether a NACK or a DTX is more likely to have occurred based on the feedback of reception of the data signal and one or more previous feedbacks from the wireless device and update a determination of whether a NACK or DTX is more likely to have occurred for the previous feedbacks. [0013] In some embodiments, a radio access node includes a transmit module operative to transmit a control signal to a wireless device scheduling a data signal and transmit the scheduled data signal to the wireless device according to the transmitted control signal; a state detection module operative to detect an ambiguous state of reception of the control signal by the wireless device based on a feedback of reception of the data signal; and a determination module operative to determine whether a NACK or a DTX is more likely to have occurred based on the feedback of reception of the data signal and one or more previous feedbacks from the wireless device and update a determination of whether a NACK or DTX is more likely to have occurred for the previous feedbacks.

[0014] Those skilled in the art will appreciate the scope of the present disclosure and realize additional aspects thereof after reading the following detailed description of the embodiments in association with the accompanying drawing figures.

Brief Description of the Drawings

[0015] The accompanying drawing figures incorporated in and forming a part of this specification illustrate several aspects of the disclosure, and together with the description serve to explain the principles of the disclosure.

[0016] Figure 1 A is a diagram depicting an exemplary cellular

communications network for communications between cells and wireless devices; Figure 1 B is a diagram of a radio access node comprising a primary cell (pCell) and a secondary cell (sCell) for a wireless device;

[0017] Figure 2A is a diagram depicting an exemplary radio access node functioning as a pCell for a wireless device with a control signal outer-loop adjustment for link adaptation; Figure 2B is a diagram depicting an exemplary network node functioning as a sCell for a wireless device with a control signal outer-loop adjustment for link adaptation;

[0018] Figure 3 illustrates a procedure of performing a control-signal outer- loop adjustment for a wireless device at a radio access node; [0019] Figures 4A and 4B illustrate the operation of a radio access node for performing a control-signal outer-loop adjustment for link adaptation when detecting an ambiguous state of reception of a control signal by a wireless device according to some embodiments of the present disclosure;

[0020] Figure 5 illustrates an exemplary trellis of potential reception states with a most probable path through the states according to some embodiments of the present disclosure;

[0021] Figure 6 illustrates a procedure of performing a control-signal outer- loop adjustment for a wireless device at a radio access node;

[0022] Figure 7 is a block diagram of a radio access node according to some embodiments of the present disclosure;

[0023] Figure 8 is a block diagram of a wireless device according to some embodiments of the present disclosure; and

[0024] Figure 9 is a block diagram of a radio access node including a transmit module, a state detection module, and a link adaptation update module according to some embodiments of the present disclosure.

Detailed Description

[0025] The embodiments set forth below represent information to enable those skilled in the art to practice the embodiments and illustrate the best mode of practicing the embodiments. Upon reading the following description in light of the accompanying drawing figures, those skilled in the art will understand the concepts of the disclosure and will recognize applications of these concepts not particularly addressed herein. It should be understood that these concepts and applications fall within the scope of the disclosure and the accompanying claims.

[0026] Before discussing the embodiments of the current disclosure, an exemplary cellular communications network 10 for communications between a radio access node 12 and wireless devices 14-1 and 14-2 (referred to herein as wireless device 14 or wireless devices 14) is discussed. Radio access node 12 may support more than one cell. Here, a cell is a geographical area covered by a base station transceiver (or radio access node 12) such that all wireless devices 14 in the geographical area have wireless connections with the base station for communications. Sometimes, a cell also refers to a radio access node 12 serving all wireless devices 14 in the geographical area. The terms "network node" and "radio access node" and "cell" are sometimes used interchangeably when there is no confusion. In Figure 1 A, the cellular communications network 10 is a 3^rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) cellular communications network. As such, 3GPP terminology is oftentimes used herein. However, while the embodiments described herein focus on 3GPP LTE, the embodiments and concepts disclosed herein may be used in any suitable type of cellular communications network (e.g., WiMAX). Cellular communications network 10 is illustrated in Figure 1 A and includes a radio access node 12 comprised of cell 16-1 and cell 16-2 (referred to herein as cell 16 or cells 16), a wireless device 14-1 which communicates with cell 16-1 using a downlink radio channel 18-1 and an uplink radio channel 20-1 , a wireless device 14-2 which communicates with cell 16-2 using a downlink radio channel 18-2 and an uplink radio channel 20-2. The uplink signal sent from a wireless device 14 to a cell 16 may include at least one of downlink Channel Quality Indicator (CQI) report, Hybrid Automatic Repeat Request (HARQ) Acknowledge (ACK) and/or Negative Acknowledge (NACK) feedback, and uplink data. The CQI and HARQ

ACK/NACK are for the corresponding cell 16 which the wireless device 14 is in. In Figure 1 A, radio access node 12 includes two cells 16. Each cell is an evolved (or enhanced) Node B (eNB) responsible for wirelessly transmitting data to and wirelessly receiving data from a wireless device 14 in the cellular

communications network 10. While only two wireless devices 14 are shown in Figure 1 A, the system is not limited thereto. Note that while the radio access node 12 has two cells 16 in this embodiment, radio access node 12 may alternatively be a low power or small base station (e.g., pico, micro, or home eNB) in a heterogeneous deployment. Also, wireless devices 14 may be User Equipments (UE).

[0027] The radio access node 12 in Figure 1 A may configure wireless devices 14 to receive downlink signals from more than one cell simultaneously, where each cell 16 may use a different carrier frequency to carry a downlink signal. This is called carrier-aggregation, in which a cell 16 is termed as a carrier. A carrier-aggregation capable wireless device 14 may be configured to receive downlink signals from more than one carrier (or cell 16) but send uplink signal back to one cell 16 only. A cell 16 with which a wireless device 14 initially established a communication and to which the wireless device 14 is sending an uplink signal is called the primary cell (pCell) for the wireless device 14, while a cell sending additional downlink signal to the wireless device 14 is called a secondary cell (sCell) for the wireless device 14. This is illustrated in Figure 1 B, where wireless device 14-3 is receiving downlink signals from cell 16-1 and cell 16-2 but is sending an uplink signal to cell 16-1 . In this case, cell 16-1 is the pCell for wireless device 14-3 while cell 16-2 is a sCell for wireless device 14-3. While only one sCell is shown in Figure 1 B, the system is not limited thereto. A wireless device 14 may be serviced by one pCell and more than one sCell. Also, the concept of pCell and sCell are device specific, meaning a pCell for one wireless device 14 may be an sCell for another wireless device 14 and an sCell for one wireless device 14 may be a pCell for another wireless device 14. When a wireless device 14 is configured to receive data from more than one carrier, the uplink signal sent from the wireless device 14 to the pCell may include at least one of downlink CQI report, and HARQ ACK and/or NACK feedback for the pCell and for at least one sCell.

[0028] To address the limitations in the pure CQI report based Physical Downlink Control Channel (PDCCH) LA, each cell 16 may include an outer-loop adjustment block to determine a control-signal outer-loop adjustment for a wireless device 14. This is illustrated in Figure 2A, where radio access node 12 is a pCell of a wireless device 14, and an outer-loop adjustment block 22 is used to determine a control-signal outer-loop adjustment (outerLoopAdj) for a given wireless device 14. The outer-loop adjustment outerLoopAdj is added to a PDCCH Signal-to-lnterference-plus-Noise Ratio (SINR) estimate based on CQI reports for the pCell. The resultant SINR estimate is used by the link adaptation 24 to determine the required Control Channel Element (CCE) aggregation level and transmit power, as discussed above. Though not explicitly shown, the outer- loop adjustment block 22 also determines a data-signal outer-loop adjustment for Physical Downlink Shared Channel (PDSCH) link adaptation. The outputs of both control signal and data-signal link adaptation are used to control a transmit processing block 26. The received uplink signal from the wireless device 14 contains at least one of CQI reports, HARQ ACK/NACK feedback, and uplink data for the pCell and perhaps an sCell as shown in Figure 2A. Receive processing block 28 extracts CQI and/or a HARQ feedback signal for at least one cell 16. The CQI report for pCell is fed to link adaptation block 24, where the CQI is used to estimate SINRs for PDCCH and PDSCH. The feedback signal for pCell is fed to ACK/NACK/Discontinuous Transmission (DTX) detection block 30, where an ACK, NACK, or DTX state of reception refers to the state of reception of the data signal by the wireless device 14. In some cases, an ambiguous state of reception of the control signal by the wireless device 14 is detected. Here, DTX represents Discontinuous Transmission, and it is used to indicate a state where an uplink signal from a wireless device 14 is expected but not detected by the radio access node 12. The results from ACK/NACK/DTX detection block 30 are fed to outer-loop adjustment block 22 to update the value of control-signal outer-loop adjustment for control-signal link adaptation as well as to update the value of data-signal outer-loop adjustment for data-signal link adaptation. The CQI and HARQ feedback for at least an sCell are fed to an outer-loop adjustment determination block to determine a control-signal outer-loop adjustment for the wireless device 14 in a secondary cell.

[0029] The outer-loop adjustment block to determine a control-signal outer- loop adjustment for the wireless device 14 in an sCell is illustrated in Figure 2B. This is similar to the one in the pCell of the wireless device 14 in Figure 2A except that the CQI report and HARQ feedback are not extracted from an uplink signal as the wireless device 14 does not send any uplink signal to its sCell. Rather the CQI report and HARQ feedback are from the pCell. As shown in Figure 2B, an outer-loop adjustment block 22 is used to determine a control- signal outer-loop adjustment (outerLoopAdj) for a given wireless device 14. The outer-loop adjustment outerLoopAdj is added to a PDCCH SINR estimate based on CQI reports. The resultant SINR estimate is used by the link adaptation 24 to determine the required CCE aggregation level and transmit power, as discussed above. Though not explicitly shown, the outer-loop adjustment block 22 also determines a data-signal outer-loop adjustment for PDSCH link adaptation. The outputs of both control signal and data-signal link adaptation are used to control a transmit processing block 26. Receive processing block 28 extracts CQI and HARQ feedback signal for the sCell. The feedback signal is fed to

ACK/NACK/DTX detection block 30, where ACK, NACK, and/or DTX state of reception of the data signal by the wireless device 14 is detected. In some cases, an ambiguous state of reception of the control signal is detected. The results from ACK/NACK/DTX detection block 30 are fed to outer-loop adjustment block 22 to update the value of control-signal outer-loop adjustment for control- signal link adaptation as well as to update the value of data-signal outer-loop adjustment for data-signal link adaptation. The link adaptations here are for the wireless device 14 in the sCell.

[0030] For more details regarding data-signal outer-loop adjustment, the interested reader is directed to U.S. Patent Application Serial No. 14/071 ,829, entitled GENERALIZED OUTER LOOP LINK ADAPTATION, and is hereby incorporated herein by reference for its teachings on data-signal outer-loop adjustment.

[0031] Figure 3 illustrates the procedure of a radio access node 12 for performing a control-signal outer-loop adjustment, denoted by outerLoopAdj, for the control-signal link adaptation for a wireless device 14 in a given cell 16. The given cell 16 can be a pCell or an sCell. The radio access node 12 first initializes the outerLoopAdj value for the wireless device 14 (step 100). In this example, the outerLoopAdj value is set to zero, but the current disclosure is not limited thereto. The outerLoopAdj value is potentially updated for each UL subframe and the radio access node 12 waits for the next UL subframe (step 102). If the given cell is an sCell (step 104), the outerLoopAdj value for the control-signal is not calculated independently from the outer-loop adjustment calculation for the data-signal, but is instead set to the outerLoopAdjpDscH, a data-signal outer-loop adjustment value calculated for a Physical Downlink Shared Channel (PDSCH), plus an offset (step 106). It is more important for the wireless device 14 to decode the PDCCH signal than to decode the PDSCH signal. As such, this offset intends for the control-signal link adaptation to choose a conservative coding and power setting for the control signal such that the PDCCH signal will be at least as easily decodable as the PDSCH signal. The reason for that arrangement is due to the fact that for sCells most UL subframes will include some ambiguity about whether the PDCCH signal was received by the wireless device 14. What is meant by ambiguity will be discussed in more detail below.

[0032] Returning to Figure 3, if the given cell is not a sCell, a further check on the presence of Physical Uplink Shared Channel (PUSCH) will be done if a PUSCH corresponding to a UL grant is expected (step 108). If a PUSCH is expected, a determination is made as to whether a PUSCH is detected in the subframe (step 1 10). If so, outerLoopAdj is adjusted up by upStep (step 1 12) which is the amount by which the control-signal LA parameter is incremented when reception of the control signal is confirmed. This indicates that the UL grant on the PDCCH was received by the wireless device 14, and the radio access node 12 can increase the control-signal outerLoopAdj to send to link adaptation 24 for the wireless device 14. If PUSCH was expected but is not detected, outerLoopAdj is adjusted down by downStep (step 1 14). This indicates that the PDCCH was not received by the wireless device 14, and the radio access node 12 should decrease the control-signal outerLoopAdj to send to link adaptation for the wireless device 14. When the wireless device 14 fails to send a transmission, this is referred to herein as a Discontinuous Transmission (DTX). The radio access node 12 can detect the presence of PUSCH by detecting insufficiently received energy on the expected demodulation reference signals (DMRS).

[0033] If PUSCH is not expected, a determination is made as to whether HARQ feedback corresponding to a DL assignment is expected in the subframe (step 1 16). If HARQ feedback is not expected, the process returns to step 102 to wait for the next UL subframe. If HARQ feedback is expected, the radio access node 12 checks for either an acknowledgment (ACK) or a negative ACK (NACK) (step 1 18). When an ACK or a NACK can be determined without any ambiguity, outerLoopAdj is adjusted up by upStep (step 120) and the process then returns to step 102 to wait for the next subframe. If an ACK or a NACK is detected without any ambiguity, this indicates that the PDCCH was received by the wireless device 14 and the radio access node 12 can increase the control-signal outerLoopAdj to send to link adaptation 24. What is meant by determining without any ambiguity will be discussed in more detail below.

[0034] If an ACK or a NACK is not detected without ambiguity, the radio access node 12 checks for a DTX with no ambiguity (step 122). A DTX with no ambiguity means that an ACK or NACK should be present in the subframe but its energy is not detected. In that case, outerLoopAdj is adjusted down by downStep (step 124). This indicates that the PDCCH was not received by the wireless device 14 and the radio access node 12 should decrease the

outerLoopAdj to send to link adaptation 24. In this manner, outerLoopAdj is updated only if there is no ambiguity regarding the reception of an ACK, the reception of a NACK, or a DTX. If there is ambiguity, then outerLoopAdj is not changed. As a result, as discussed below in detail, the link adaptation is less than ideal.

[0035] More specifically, the afore-described outer-loop adjustment procedure works well whenever the HARQ feedback result can be determined to be one of these three cases:

(a) PDSCH ACK - wireless device 14 successfully decodes both PDCCH and PDSCH of a cell and sends ACK of the PDSCH of the cell back to a radio access node 12;

(b) PDSCH NACK - wireless device 14 successfully decodes PDCCH of a cell but fails to decode PDSCH of the cell, and reports back NACK of the PDSCH of the cell to a radio access node 12; and

(c) DTX - wireless device 14 fails to decode PDCCH of a cell and does not send any feedback for the cell to a radio access node 12. [0036] However, as alluded to above, there are cases where there is an ambiguity between NACK and DTX. Some of these cases include when HARQ- NACK and a Scheduling Request (SR) need to be transmitted on the same resource. When the wireless device 14 transmits the SR, the HARQ feedback will appear to be a NACK. However, it is possible that the wireless device 14 did not receive PDCCH, but the presence of the SR bit may appear as a NACK. Since the two possibilities result in opposite actions (NACK implies PDCCH was received; DTX implies PDCCH was not received), the radio access node 12 does nothing in response to this uplink subframe.

[0037] In the case of carrier aggregation, a wireless device 14, wireless device 14-3 in Figure 1 B as an example, is connected to a pCell and one or more sCells. The wireless device 14 needs to send HARQ feedback for pCell and each of the configured sCells regardless of the PDCCH decoding result. This also results in an ambiguity. Specifically, HARQ feedback may be carried by PUSCH to the radio access node 12 where HARQ feedback for pCell and each configured sCell may be concatenated. As such, if the wireless device 14 cannot detect any PDCCH in one configured sCell, the wireless device 14 needs to send NACK, which actually means a DTX whenever there is something scheduled on the sCell. In this situation, it is not possible for the radio access node 12 to distinguish between a NACK and a DTX.

[0038] Another possibility is that HARQ feedback may be carried by the PUCCH to the radio access node 12 where HARQ feedback bits for pCell and configured sCells are jointly coded. In some cases of jointly coded HARQ feedback, there is no distinction between NACK and DTX by design. Table 1 below is an example which is copied from Table 10.1 .2.2.1 -3 for PUCCH format 1 b in Technical Specification (TS) 36.213, with two PUCCH resources. The cases labeled NACK/DTX are cases with ambiguity as the radio access node 12 is not able to distinguish between a NACK and a DTX and therefore does nothing in response to this uplink subframe. Table 1 : Transmission of Format 1 b HARQ-ACK channel selection

[0039] These cases result in an ambiguous state of reception of the control signal by the wireless device 14. When the radio access node 12 detects this ambiguous state of reception, the PDCCH outer-loop algorithm described in Figure 3 chooses not to do any outer-loop adjustment. In other words, the PDCCH outer-loop algorithm described in Figure 3 does outer-loop adjustment only in scenarios when there is no ambiguity between DTX and NACK. In this way, many opportunities to adjust the PDCCH LA may be missed, leading to less efficient use of PDCCH resources.

[0040] When multiple transmissions of the same data signal may be necessary, the Redundancy Version (RV) is encoded in each DL grant to indicate the start position in the circular buffer for wireless device 14 to perform soft combining. It is up to the vendor to select a particular RV order in the HARQ transmissions. To realize the gain of Incremental Redundancy (IR) over Chase Combining (CC), RV=0 is commonly selected in the initial transmission (first HARQ transmission) given its superior performance over other options. The RV of the rest of the HARQ transmissions (retransmissions) is often non-zero to harness the benefit of IR. For instance, if the same user data has to be transmitted four times, the radio access node 12 may choose RV order to be 0, 2, 1 , and 3, i.e., RV=0 for first (initial) transmission, RV=2 for second

transmission (first re-transmission), RV=1 for third transmission (second retransmission), and RV=3 for fourth transmission (third re-transmission).

Similarly, the radio access node 12 may choose RV order to be 0, 1 , 2, and 3, etc. [0041] It is known that missing RV=0 severely affects the HARQ performance. Hence in practice, if the radio access node 12 detects DTX in the HARQ feedback of the initial transmission, it is desirable to select RV=0 in the second HARQ transmission to improve the probability of successful decoding. On the other hand, if the radio access node 12 detects NACK in the HARQ feedback of the initial transmission, then the radio access node 12 should pick RV=1 , 2, or 3, to improve the probability of successful decoding.

[0042] The success of data transmission depends on the reliability of both PDCCH and PDSCH, which are maintained by PDCCH outer-loop control and PDSCH outer-loop control, respectively, as discussed above. Upon the reception of the HARQ feedback, i.e., ACK or NACK, the radio access node 12 can make a retransmission decision and adjust the PDSCH outer-loop. If the wireless device 14 is aware of a DTX and encodes DTX in the HARQ feedback, the radio access node 12 can adjust PDCCH outer-loop so that PDCCH resource allocation e.g., CCE aggregation level or transmit power, can be efficient and robust.

[0043] However, explicitly packing DTX in HARQ feedback is not always feasible. In fact, using LTE HARQ feedback procedures such as multiplexing, format 1 b with channel selection, or format 3 when a wireless device 14 detects a NACK or DTX, the standards often only allow the wireless device 14 to pack one HARQ bit, which can carry either ACK or NACK/DTX. This causes ambiguity to the radio access node 12 which cannot distinguish whether the wireless device 14 has encountered a NACK or DTX. This is referred to herein as an ambiguous state of reception. If PDCCH is unreliable and the radio access node 12 does not make adjustment in PDCCH outer-loop due to the absence of explicit DTX feedback, DTX may continue to occur and cause data transmissions to fail and DL throughput will be significantly affected. On the other hand, if PDCCH outer- loop makes conservative decisions, the cost of PDCCH resource will be unnecessarily high.

[0044] In the next generation of the wireless communication systems, radio access nodes 12 will be capable of transmitting a larger number of codewords to the wireless device 14 by using few DL grants as more orthogonal channels are available and each channel is interference-free in the presence of massive Multiple Input Multiple Output (MIMO) and beamforming. More bits to send ACK/NACK feedback seem inevitable, which puts higher pressure on the capacity of the HARQ feedback. This implies that encoding explicit DTX in the feedback is even more unlikely since it can be very expensive, resource-wise. However, an actual DTX will severely affect the communication due to the fact that the loss of a DL grant may cause transmission failure of a large number of codewords. Embodiments described herein provide solutions to mitigate the impact of the absence or lack of explicit DTX feedback.

[0045] To address the limitations in the method of Figure 3 where the radio access node 12 could not determine whether the transmission was a NACK or a DTX, an exemplary operation of a radio access node 12 for performing a control- signal outer-loop adjustment for the PDCCH LA in a cell when detecting an ambiguous state of reception of a control signal by a wireless device 14 is provided.

[0046] In some embodiments described herein, these problems are addressed by the radio access node 12 using feedback (e.g., HARQ feedback) for all the transmissions jointly to compute the probability of the decoding outcome: DTX or NACK. This enables control channel outer-loop to make decisions based on the combination of feedbacks {ACK, NACK, DTX,

NACK/DTX} of each DL transmission. In some embodiments, this approach follows the concept of Maximum a posteriori Probability (MAP) detection. In some embodiments, the radio access node 12 also adjusts the modulation level as well as the RV to improve DTX detection.

[0047] This is illustrated in Figures 4A and 4B. According to some

embodiments of the present disclosure, the radio access node 12 in a cell transmits a control signal to a wireless device 14 scheduling a data signal (step 200). The radio access node also transmits the scheduled data signal to the wireless device 14 according to the transmitted control signal (step 202). The radio access node 12 in the cell then detects an ambiguous state of reception of the control signal by the wireless device 14 based on a feedback of reception of the data signal (step 204). The radio access node 12 then determines whether a NACK or a DTX is more likely to have occurred based on the feedback and one or more previous feedbacks from the wireless device 14 (step 206). Also, the radio access node 12 may update a determination of whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device 14 (step 208). More details for this updating step are provided below.

[0048] Figure 4B illustrates some ways that step 204 may detect an ambiguous state of reception of the control signal in a cell 16 by a wireless device 14. As an example, the radio access node 12 may detect a condition that is indicative of either a UL SR from the wireless device 14 or a NACK from the wireless device 14, but the radio access node 12 is unable to distinguish whether the NACK or a DTX has occurred (step 204A). As another example, the radio access node 12 may detect a condition that is indicative of HARQ ACKs and/or NACKs for a pCell and at least one sCell from the wireless device 14 which are concatenated, but where the radio access node 12 is unable to distinguish whether a NACK or a DTX has occurred for one of the pCell and at least one sCell (step 204B). Also, the radio access node 12 may detect a condition that is indicative of HARQ ACKs and/or NACKs for a pCell and at least one sCell from the wireless device 14 which are jointly coded, but where the radio access node 12 is unable to distinguish whether a NACK or a DTX has occurred for one of the pCell and the at least one sCell (step 204C).

[0049] Returning to Figure 4A, the radio access node 12 may optionally update a LA parameter used to choose a coding scheme and/or power level for transmission of information in the control signal (step 210).

[0050] In some embodiments, the radio access node 12 selects RV=0 if it determines that the highest probability of the decoding outcome at the wireless device 14 for an RV=0 transmission is DTX. For example, the radio access node 12 may have selected RV=0 and 2 for the i^th and (i+1 )^th transmission,

respectively. If after receiving the feedback for these two transmissions, the radio access node 12 computes that the DTX probability of the i transmission is greater than all other scenarios, it will select RV=0 in the next HARQ

transmission.

[0051 ] If the wireless device 14 encounters a DTX for RV=0, the probability of NACK is much higher than DTX in the following retransmissions. This makes DTX detection for the retransmissions extremely difficult. To alleviate this difficulty, in the present disclosure, when the high probable events contain a DTX for RV=0, the radio access node 12 uses RV=0 again in the next retransmission. This significantly reduces the probability of the next NACK event in the MAP detection. Hence, in spite of no explicit DTX feedback, these embodiments may improve DTX detection if, in fact, the wireless device 14 encounters a DTX for RV=0.

[0052] It is noted that selection of RV=0 may also provide improvement on PDSCH as the probability of successful PDSCH decoding increases in comparison to other options if the previous RV=0 is lost.

[0053] In some embodiments, if the MAP calculation produces a second NACK, the radio access node 12 reduces a Modulation and Coding Scheme (MCS) level if possible while it maintains the same transport block size in the next HARQ transmission. This approach considers IR gain and provides further improvement to the PDSCH decoding. As a result, the NACK probability is reduced. If the HARQ retransmission fails (i.e., the radio access node 12 receives NACK/DTX) even after MCS is reduced, then the probability of DTX becomes higher and DTX becomes "easier" to detect.

[0054] Upon the reception of each HARQ feedback, eNB or the radio access node 1 2 calculates Maximum a posteriori Probability of a sequence of ACK, NACK, DTX using a state trellis in the following expression:

where s_t e {ACK, NACK, DTX} is the decoding outcome at the wireless device 14 in the i^th HARQ transmission of the j^th branch on the trellis, r _j 6 {ACK, NACK, DTX, NACK/DTX} is the received HARQ feedback in the i^th HARQ transmission of the j^th branch on the trellis, I is the number of HARQ transmissions corresponding to the current feedback, I≤ L. L is the maximum number of HARQ transmissions, υ^₌₁ U[₌₁ s_{i ;} is the union of all decoding outcomes until the Z^th HARQ transmission and N branches, and υ^₌₁ U[₌₁ r_{i ;} is the union of all decoding outcomes until the Z^th HARQ transmission and N branches. In some embodiments, the most probable states are determined using the Viterbi algorithm or some variant of it. Larger values of N give a more accurate estimation of the decoding outcomes but require additional processing and a longer delay for outer loop adjustment.

[0055] This method considers the possible decoding outcome (decoding events), {ACK, NACK, DTX}, in the past jointly. The proposed PDCCH outer- loop algorithm is based on all the information the radio access node 12 has collected regarding the transmission of each transport block (on the basis of each carrier). This is in contrast with the PDSCH outer-loop algorithm, which only considers the initial HARQ transmission.

[0056] In some embodiments, certain probabilities are defined as the following:

• P_D: probability of DTX.

· P^: probability of NACK for all i HARQ transmissions of a transport block, i = 1, 2,

• Let SINRpuccH be stimated SINR by the PDCCH outer-loop. Let S_UP and 5_D0WN be step sizes for increasing and decreasing SLNR_POCCU, respectively.

[0057] In some embodiments, to estimate the probability of DTX and NACK, the radio access node 12 maintains four counters. In particular:

• m_TX: number of HARQ transmissions (total transmissions and retransmissions)

• m_n ^l _T : number of DTXs at the i^th HARQ transmissions. • m _x: number of non-DTX'd HARQ transmissions from the previous non-DTX'd RV = 0 transmission to the most recent non-DTX'd transmission at the i^th HARQ transmissions.

• ^mNACK ^: number of detected or received NACK from the previous non-DTX'd RV = 0 transmission to the most recent non-DTX'd transmission at the i^th HARQ transmissions.

[0058] In some embodiments, these counters can be used to estimate probability P_D by the following expression:

[0059] Similarly, probability is estimated by the following expression:

•p(i) _ ^mNACK

m_TX

[0060] In some embodiments, before sufficient statistics have been collected at the radio access node 12, a warm-up period is used to characterize the probability of NACK and DTX. During this period, the radio access node 12 calculates P_D and P^, but does not use them to identify DTX in the PDCCH outer-loop.

[0061 ] The characterization consists of two phases: estimation of P^ and estimation of P_D. In the ¾^i} estimation phase; a conservative PDCCH resource management strategy is used, normal HARQ transmission procedure is used, and the radio access node 12 counts every NACK/DTX feedback as NACK.

Similarly, in the P_D estimation phase, a conservative HARQ transmission strategy is used, normal PDCCH procedure is used, and the radio access node 12 counts every NACK/DTX feedback as DTX. The radio access node 12 can start the characterization with either phase.

[0062] Figure 5 shows a trellis illustrating the decoding outcomes (HARQ events) at the wireless device 14 over four DL HARQ transmissions. The probability of each branch on the trellis can be completely described by P_D and P_j^. The diagram illustrates the branches and paths in the MAP decision in PDCCH outer-loop. In a trellis, each node corresponds to a hidden state

(decoding outcome) at a given time. Each arrow represents a transition to a new state at the next instant of time. Some transitions are not possible. As shown in the trellis of Figure 5, any instant of time that is an "ACK" decoding outcome does not have any further states to transition to because the radio access node 12 will not need to retransmit the data. Every path through the trellis corresponds to a sequence of states (decoding outcomes). Many such paths exist, and the probability that each one caused the observed states is calculated. The Viterbi algorithm is one way to efficiently compute the most likely path through the trellis, such paths are referred to as survivor paths. If the path with the maximum probability contains a DTX event, the radio access node 12 determines that a DTX has occurred in the HARQ transmissions. The path in bold is an example of a possible path of DTX-NACK-DTX-NACK in the trellis of Figure 5.

[0063] Upon receiving each HARQ feedback, the radio access node 12 selects the path that satisfies the MAP equation above by calculating the probability of each path on the trellis using P_D and If the outcome indicates that additional DTX has likely occurred comparing with the path selected last time, PDCCH outer loop lowers STNR_POCCH by 5_D0WN. Meanwhile, PDCCH outer loop increases STNR_{PO H} by nS_UP if n non-DTX transmissions are detected. Depending on the implementation, the radio access node 12 also updates the statistics of NACK and DTX after each MAP decision is made.

[0064] Figure 6 illustrates the MAP-based DTX detection for the PDCCH outer-loop control according to some embodiments. The radio access node 12 first receives a HARQ feedback from the wireless device 14 (step 300). The radio access node 12 determines if there is an ambiguous state of NACK and DTX (step 302). If there is no ambiguity, the process proceeds to adjust the PDCCH outer loop based on the HARQ feedback (step 304).

[0065] If there is ambiguity, the radio access node 12 performs a MAP detection of the most likely states as discussed above (step 306). Notably, this requires the use of various statistics 31 discussed above. As before, the radio access node adjusts the PDCCH outer loop based on the most likely states of the HARQ feedbacks in step 304. The process may also include determining to retransmit using RV=0 as discussed above.

[0066] The radio access node 12 then updates the various counters 33 that are used in calculations (step 308). Based on these updated counters, the radio access node 12 updates the various statistics 31 (step 310). This process can continue as long as needed to properly transmit the data signal to the wireless device 14.

[0067] It is noted that as opposed to deterministic mapping, in some embodiments, the radio access node 12 selects RV=0 solely depending on the probability calculation for each path on the trellis.

[0068] For a given code rate of the DL grant Quadrature Phase Shift Keying (QPSK) and estimated SINR, an estimated Block Error Rate (BLER) value,

P_D can be obtained using any standard technique of BLER estimation. Using P_D, the value of P_D can be estimated as:

P_D = (1 - a)P_D + aP_D,

where a is a filtering factor.

[0069] Additionally, the PDCCH outer-loop may trigger power adjustment or rate adjustment for the transmission of the DL grant. P_D is updated for each of these PDCCH outer-loop actions. P_D is mapped to an SINR. Power adjustment requires the update of this SINR, which is mapped to a BLER. The BLER is used to update P_D. Using the adjusted code rate, a BLER value is determined and used to update P_D . In some embodiments, to maintain the base of the statistics, m_TX remains but∑i₌₁ m_m ^l _x is adjusted accordingly when P_D is updated. The new P_D will be used on the trellis on next branch until it is updated next time.

[0070] Figure 7 is a block diagram of a radio access node 12 according to some embodiments of the present disclosure. As illustrated, the radio access node 12 includes a baseband unit 32 with a processor 34, memory 36, and a network interface 38. As illustrated, the radio access node 12 also includes a radio unit 40 with a transceiver 42 and one or more antennas 44. In some embodiments, the radio access node 12, or the functionality of the radio access node 12 described with respect to any one of the embodiments described herein, is implemented in software that is stored in, e.g., the memory 36 and executed by the processor 34. The network interface 38 may include one or more

components (e.g., network interface card(s)) that connect the radio access node 12 to other systems.

[0071] In some embodiments, a computer program including instructions which, when executed by at least one processor, causes the at least one processor to carry out the functionality of the radio access node 12 according to any one of the embodiments described herein is provided. In some

embodiments, a carrier containing the aforementioned computer program product is provided. The carrier is one of an electronic signal, an optical signal, a radio signal, or a computer readable storage medium (e.g., a non-transitory computer readable medium such as the memory 36).

[0072] Figure 8 is a block diagram of a wireless device 14 according to some embodiments of the present disclosure. As illustrated, the wireless device 14 includes a processor 46, memory 48, a transceiver 50, and at least one antenna 52. In some embodiments, wireless device 14, or the functionality of the wireless device 14 described with respect to any one of the embodiments described herein, is implemented in software that is stored in, e.g., the memory 48 and executed by the processor 46. The transceiver 50 uses the at least one antenna 52 to transmit and receive signals and may include one or more components that connect the wireless device 14 to other systems.

[0073] In some embodiments, a computer program including instructions which, when executed by at least one processor, causes the at least one processor to carry out the functionality of the wireless device 14 according to any one of the embodiments described herein is provided. In some embodiments, a carrier containing the aforementioned computer program product is provided. The carrier is one of an electronic signal, an optical signal, a radio signal, or a computer readable storage medium (e.g., a non-transitory computer readable medium such as the memory 48). [0074] Figure 9 is a block diagram of radio access node 12 according to some embodiments of the present disclosure. As illustrated, the radio access node 12 includes a transmit module 54, a state detection module 56, and a determination module 58 that are each implemented in software that, when executed by a processor of the radio access node 12, causes the radio access node 12 to operate according to one of the embodiments described herein. The transmit module 54 operates to transmit a control signal to a wireless device 14, as described above with respect to the transmitting step 200. The state detection module 56 operates to detect an ambiguous state of reception of the control signal by the wireless device 14 as discussed above with respect to, for example, steps 204-204C. The determination module 58 operates to determine if a NACK or a DTX is more likely to have occurred as discussed above with respect to, for example, step 206, and update a determination of whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14) as discussed above with respect to, for example, step 208.

[0075] The following acronyms are used throughout this disclosure.

• 3GPP Third Generation Partnership Project

• ACK Acknowledgement

• BLER Block Error Rate

• CC Chase Combining

• CCE Control Channel Elements

• CQI Channel Quality Indicator

• DCI Downlink Control Information

• DL Downlink

• DTX Discontinuous Transmission

• eNB Evolved Node B

• HARQ Hybrid Automatic Repeat Request

• IR Incremental Redundancy

• LA Link Adaptation

• LTE Long Term Evolution

• NACK Negative Acknowledgement MIMO Multiple Input Multiple Output

outerLoopAdj Outer-Loop Adjustment

pCell Primary Cell

PDCCH Physical Downlink Control Channel

PDSCH Physical Downlink Shared Channel

PUCCH Physical Uplink Control Channel

PUSCH Physical Uplink Shared Channel

sCell Secondary Cell

SINR Signal-to-lnterference-and-Noise Ratio

SR Scheduling Request

UE User Equipment

UL Uplink

WiMAX Worldwide Interoperability for Microwave

[0076] Those skilled in the art will recognize improvements and modifications to the embodiments of the present disclosure. All such improvements and modifications are considered within the scope of the concepts disclosed herein.

Claims

What is claimed is:

1 . A method of operation of a radio access node (12) in a cellular

communications network (10) comprising:

transmitting (200) a control signal to a wireless device (14) scheduling a data signal;

transmitting (202) the scheduled data signal to the wireless device (14) according to the transmitted control signal;

detecting (204) an ambiguous state of reception of the control signal by the wireless device based on a feedback of reception of the data signal;

determining (206) whether a negative acknowledgment, NACK, or discontinuous transmission, DTX, is more likely to have occurred based on the feedback of reception of the data signal and one or more previous feedbacks from the wireless device (14); and

updating (208) a determination of whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14). 2. The method of claim 1 further comprising:

in response to determining whether a NACK or a DTX is more likely to have occurred, updating (210) a link adaptation, LA, parameter used to choose a coding scheme and/or power level for transmission of information in the control signal.

3. The method of claim 2 wherein:

a NACK is more likely to have occurred; and

updating the LA parameter used to choose the coding scheme and/or power level comprises incrementing the LA parameter by a predetermined amount.

4. The method of any of claims 1 through 2 wherein:

a DTX is more likely to have occurred; and

updating the LA parameter used to choose the coding scheme and/or power level comprises decrementing the LA parameter by a predetermined amount.

5. The method of claims 1 through 4 further comprising:

in response to determining that a DTX is more likely to have occurred, retransmitting the scheduled data signal to the wireless device (14) using a redundancy version, RV, of 0.

6. The method of any of claims 1 through 5 wherein determining whether a NACK or a DTX is more likely to have occurred comprises determining whether a NACK or a DTX is more likely to have occurred using maximum a posteriori probability, MAP, detection.

7. The method of any of claims 1 through 6 wherein updating the

determination of whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14) comprises using a Viterbi algorithm.

8. The method of claim 7 further comprising:

in response to determining whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14), updating the LA parameter used to choose a coding scheme and/or power level for transmission of information in the control signal.

9. The method of any of claims 7 through 8 further comprising:

in response to determining that a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14) for a transmission with RV = 0, retransmitting the scheduled data signal to the wireless device (14) using an RV = 0.

10. The method of any of claims 7 through 9 wherein determining whether a NACK or a DTX is more likely to have occurred using MAP detection comprises computing arg max_s Ρ(υ^₌₁ U[₌₁ s_{i ;} | υ^₌₁ U^n,), where

Sj ₇ 6 {ACK, NACK, DTX} is a decoding outcome at the wireless device (14) in the i^th transmission of the j^th branch on a trellis, 6 {ACK, NACK, DTX, NACK/DTX} is the feedback from the wireless device (14) in the i^th transmission of the j^th branch on the trellis, and I is the number of transmissions corresponding to the current feedback, N is the length of a survivor path.

1 1. The method of any of claims 7 through 10 wherein determining whether a NACK or a DTX is more likely to have occurred using MAP detection comprises maintaining counters for m_TX, m_QTX, m^x, and m^_ACK where:

m_TX is a total number of transmissions;

mmx is the number of DTX at the i^th HARQ transmission;

m _x is a number of non-DTX transmissions that have occurred from when a previous non-DTX transmission of the scheduled data signal to the wireless device (14) using an RV of 0 until a most recent non-DTX transmission at the i^th HARQ transmission; and

™NACK is ^a number of NACK transmissions that have occurred from when a previous non-DTX transmission of the scheduled data signal to the wireless device (14) using an RV of 0 until a most recent non-DTX transmission at the i^th HARQ transmission.

The method of claim 1 1 wherein determining whether a NACK or a DTX likely to have occurred using MAP detection further comprises:

_ yL_ _mi

estimating a probability of a DTX as P_D = ^{1-1 DTX}; and estimating a probability of NACK for all transmissions as (0 - _ ¹⁷¹ NACK

"'τχ

13. The method of claim 12 wherein determining whether a NACK or a DTX is more likely to have occurred using MAP detection further comprises:

updating the probability of a DTX based on an estimated Block Error Rate, BLER, value, P_Das P_D = (1 - a)P_D + aP_D, where a is a filtering factor between 0 and 1 .

14. The method of any of claims 7 through 13 wherein determining whether a NACK or a DTX is more likely to have occurred using MAP detection comprises finding a most probable length N path through the trellis representing decoding outcomes at the wireless device (14) of the transmissions by using a Viterbi algorithm.

15. The method of any of the preceding claims wherein the cellular

communications network is a Long Term Evolution, LTE, network.

16. The method of any of the preceding claims wherein the control signal is transmitted on a Physical Downlink Control Channel, PDCCH.

17. A radio access node (12) in a cellular communications network (10), comprising:

a processor; and

a memory coupled to the processor, the memory containing instructions executable by the processor whereby the radio access node (12) is operative to:

transmit a control signal to a wireless device (14) scheduling a data signal;

transmit the scheduled data signal to the wireless device (14) according to the transmitted control signal;

detect an ambiguous state of reception of the control signal by the wireless device based on a feedback of reception of the data signal;

determine whether a negative acknowledgment, NACK, or a discontinuous transmission, DTX, is more likely to have occurred based on the feedback of reception of the data signal and one or more previous feedbacks from the wireless device (14); and

update a determination of whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14).

18. The radio access node (12) of claim 17 further operable to:

in response to determining whether a NACK or a DTX is more likely to have occurred, update a link adaptation, LA, parameter used to choose a coding scheme and/or power level for transmission of information in the control signal.

19. The radio access node (12) of claim 18 wherein:

determining whether a NACK or a DTX is more likely to have occurred comprises being operable to determine that a NACK is more likely to have occurred; and

updating the LA parameter used to choose the coding scheme and/or power level comprises being operable to increment the LA parameter by a predetermined amount. 20. The radio access node (12) of any of claims 17 through 18 wherein:

determining whether a NACK or a DTX is more likely to have occurred comprises being operable to determine that a DTX is more likely to have occurred; and

updating the LA parameter used to choose the coding scheme and/or power level comprises being operable to decrement the LA parameter by a predetermined amount.

21. The radio access node (12) of claims 17 through 20 further operable to: in response to determining that a DTX is more likely to have occurred, retransmit the scheduled data signal to the wireless device (14) using a redundancy version, RV, of 0.

22. The radio access node (12) of any of claims 17 through 21 wherein determining whether a NACK or a DTX is more likely to have occurred comprises being operable to determine whether a NACK or a DTX is more likely to have occurred using maximum a posteriori probability, MAP, detection.

23. The radio access node (12) of any of claims 17 through 22 wherein updating a determination of whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14) comprises being operable to use a Viterbi algorithm.

24. The radio access node (12) of claim 23 further operable to:

in response to determining whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14), update the LA parameter used to choose a coding scheme and/or power level for transmission of information in the control signal.

25. The radio access node (12) of any of claims 23 through 24 further operable to:

in response to determining that a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14) DTX for a transmission with RV =0, retransmit the scheduled data signal to the wireless device (14) using an RV of 0. 26. The radio access node (12) of any of claims 23 through 25 wherein determining whether a NACK or a DTX is more likely to have occurred using MAP detection comprises being operable to compute

arg max_s .

s_ii7- 1 Uy u ,;), where s_i:j e {ACK, NACK, DTX} is a decoding outcome at the wireless device (14) in the i^th HARQ transmission on the y^thbranch on a trellis, 6 {ACK, NACK, DTX, NACK/DTX} is the feedback from the wireless device (14) in the i^th transmission on the j^th branch on the trellis, and I is the number of HARQ transmissions corresponding to the current feedback, N is the length of a survivor path.

27. The radio access node (12) of any of claims 23 through 26 wherein determining whether a NACK or a DTX is more likely to have occurred using MAP detection comprises being operable to maintain counters for m_TX , m_m ^l _x, m _x, and m^_ACK where:

m_TX is a total number of transmissions;

m_m ^l is the number of DTX at the i^th HARQ transmission;

m _x is a number of non-DTX transmissions that have occurred from when a previous non-DTX transmission of the scheduled data signal to the wireless device (14) using an RV of 0 until a most recent non-DTX transmission after at i^th HARQ transmission; and

mNACK '^{s a} number of NACK transmissions that have occurred from when a previous non-DTX transmission of the scheduled data signal to the wireless device (14) using an RV of 0 until a most recent non-DTX transmission at i^th HARQ transmission.

28. The radio access node (12) of claim 27 wherein determining whether a NACK or a DTX is more likely to have occurred using MAP detection further comprises being operable to:

_ yL_ _mi

estimate a probability of a DTX as P_D = ^1- τ¹η_τχ ^DTX; and estimate a probability of NACK for all transmissions as = ^mNf^CK.

29. The radio access node (12) of claim 28 wherein determining whether a NACK or a DTX is more likely to have occurred using MAP detection further comprises being operable to: update the probability of a DTX based on an estimated Block Error Rate, BLER, value, P_Das P_D = (1 - a)P_D + aP_D, where a is a filtering factor between 0 and 1 . 30. The radio access node (12) of any of claims 23 through 29 wherein determining whether a NACK or a DTX is more likely to have occurred using MAP detection comprises being operable to find a most probable path through the trellis representing decoding outcomes at the wireless device (14) of the transmissions by using a Viterbi algorithm.

31. The radio access node (12) of any of claims 17 through 30 wherein the cellular communications network (10) is a Long Term Evolution, LTE, network.

32. The radio access node (12) of any of claims 17 through 31 wherein the control signal is transmitted on a Physical Downlink Control Channel, PDCCH.

33. A radio access node (12) adapted to:

transmit a control signal to a wireless device (14) scheduling a data signal; transmit the scheduled data signal to the wireless device (14) according to the transmitted control signal;

detect an ambiguous state of reception of the control signal by the wireless device (14) based on a feedback of reception of the data signal;

update a determination of whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14). 34. The radio access node (12) of claim 33 adapted to perform the method of any of claims 2 through 16.

35. A computer program comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out the method according to any one of claims 1 through 16.

36. A carrier containing the computer program of claim 35, wherein the carrier is one of an electronic signal, an optical signal, a radio signal, or a computer readable storage medium. 37. A radio access node (12) comprising:

a transmit module (54) operative to transmit a control signal to a wireless device (14) scheduling a data signal and transmit the scheduled data signal to the wireless device (14) according to the transmitted control signal;

a state detection module (56) operative to detect an ambiguous state of reception of the control signal by the wireless device (14) based on a feedback of reception of the data signal; and

a determination module (58) operative to, determine whether a negative acknowledgment, NACK, or a discontinuous transmission, DTX, is more likely to have occurred based on the feedback of reception of the data signal and one or more previous feedbacks from the wireless device (14) and update a

determination of whether a NACK or a DTX is more likely to have occurred for the one or more previous feedbacks from the wireless device (14).