METHODS AND COMPUTING DEVICE FOR CONFIGURING RESOURCE SCHEDULING
TECHNICAL FIELD
The present disclosure is related generally to wireless networks and, more particularly, to methods and a computing device for configuring resource scheduling.
BACKGROUND
In a typical multi-slot scheduling scheme, the scheduling grant (e.g., from a wireless carrier’s network to a mobile device) provides scheduling parameters that stay the same for all of the scheduled slots. This minimizes the control signaling overhead but sacrifices flexibility. Specifically, if the scheduling scenario, changes from slot to slot, e.g., from single user ( “SU” ) to multiple user ( “MU” ) or vice versa) , performance will be degraded if all slots adopt the same scheduling parameters. In addition, higher priority traffic, such as ultra-Reliable and Low Latency Communications ( “uRLLC” ) traffic, may preempt some resources of some slots, and the preemption may be different for each slot. Therefore, some scheduling parameters should be updated for each slot in a multi-slot scheduling scheme.
DRAWINGS
While the appended claims set forth the features of the present techniques with particularity, these techniques, together with their objects and advantages, may be best understood from the following detailed description taken in conjunction with the accompanying drawings of which:
FIG. 1 is a diagram of a system in which various embodiments of the disclosure are implemented.
FIG. 2 shows an example hardware architecture, according to an embodiment.
FIG. 3 shows a slot configuration in which the demodulation reference signal (“DMRS” ) is at the beginning of each transmission region.
FIG. 4 shows a slot configuration similar to that of FIG. 3, but with an additional DMRS.
FIG. 5 illustrates how one control channel in the first slot can be used to configure scheduling parameters for data transmission in all of the slots, according to an embodiment.
FIG. 6 illustrates how the variations in the power of the DMRS among various slots can be communicated, according to an embodiment.
FIG. 7 illustrates how resources may overlap for multiple devices, according to an embodiment.
FIG. 8 depicts how the power offset or power ratio is transmitted via a dedicated DCI in slot, according to an embodiment.
FIG. 9 depicts how dedicated downlink control information ( “DCI” ) may be used to transmit changed parameters, according to an embodiment.
FIG. 10 depicts how a communication node communicates the “changed” parameters to a second communication node via DCI after the multi-slot scheduling, according to an embodiment.
FIG. 11 depicts how urgent traffic preempts normal traffic, according to an embodiment.
FIG. 12 depicts how preemption information may be transmitted after the preempted resources, according to an embodiment.
FIG. 13 depicts multi-slot scheduling, according to an embodiment.
FIG. 14 depicts single slot scheduling, according to an embodiment.
FIG. 15 depicts how rate matching parameters can be communicated in the data region of a last slot in a multi-slot scheduling scenario, according to an embodiment.
DESCRIPTION
The present disclosure is generally directed to a method for configuring resource scheduling in a wireless network. According to various embodiments, a first communication
node (e.g., a user equipment ( “UE” ) or node of a carrier network, such a next generation Node B ( “gNB” ) ) , schedules a second communication node for communication on multiple scheduling units (e.g., multiple slots or subframes, with each scheduling unit being set, for example at one slot per scheduling unit, two slots per scheduling unit, etc. ) . The first communication node transmits a first message and a second message to the second communication node. The first message includes scheduling parameters that apply to all of the multiple scheduling units, and the second message reconfigures at least one of the scheduling parameters for a subset of the multiple scheduling units.
In an embodiment, the second message reconfigures at least one of the plurality of different scheduling parameters by indicating that the allocated set of resources has been preempted, in which case no preemption indication is included in the first message.
In various embodiments, the second communication node buffers incoming transmissions, receives configuration parameters, and applies those parameters to the buffered transmissions in order to, for example, properly decode the buffered transmission.
Embodiments described herein provide flexibility for multi-slot scheduling and the ability to dynamically configure or reconfigure the parameters by communicating only the changed parameters (without the need to provide the entire set of parameters) . This results in increased flexibility without excessive overhead.
FIG. 1 depicts a wireless communication system 100 in which the various embodiments may be deployed. The communication system 100 includes several communication nodes. The communication nodes depicted are gNB 102, UE #0 and UE #1. It is to be understood that there may be many other communication nodes and that the ones represented in FIG. 1 are meant only for the sake of example. In an embodiment, the wireless communication system 100 has many components that are not depicted in FIG. 1, including other gNBs, other UEs, wireless infrastructure, wired infrastructure, and other devices commonly found in LTE networks. An example implementation of the gNB 102 is a new radio ( “NR” ) base station. Example implementations of the UE #0 and UE #1 include any device capable of wireless communication, such as a smartphone, tablet, laptop computer, and non-traditional devices (e.g., household appliances or other parts of the “Internet of Things” ) .
FIG. 2 illustrates a basic (computing device) hardware architecture found in both the gNB 102 and in the UEs, according to an embodiment. The gNB 102 and the UEs have other components as well, some of which are common to both and others that are not. The hardware architecture depicted in FIG. 2 includes logic circuitry 202, memory 204, transceiver 206, and one more antennas represented by antenna 208. The memory 204 may be or include a buffer that, for example, holds incoming transmissions until the logic circuitry is able to process the transmission. Each of these elements is communicatively linked to one another via one or more data pathways 210. Examples of data pathways include wires, conductive pathways on a microchip, and wireless connections.
The term “logic circuitry” as used herein means a circuit (atype of electronic hardware) designed to perform complex functions defined in terms of mathematical logic. Examples of logic circuitry include a microprocessor, a controller, or an application-specific integrated circuit. When the present disclosure refers to a device carrying out an action, it is to be understood that this can also mean that logic circuitry integrated with the device is, in fact, carrying out the action.
Possible implementations of the memory 204 include: volatile data storage; nonvolatile data storage; electrical memory; magnetic memory; optical memory; random access memory ( “RAM” ) ; cache memory; and hard drives.
In currently-proposed implementations of NR networks, flexible design is preferred in order to meet requirements for different scenarios. One possible way to implement a DMRS in NR networks is to place it at the beginning of each transmission region, as shown in FIG. 3. In this case, the communication node receiving the DMRS (e.g., UE or gNB) can demodulate the DMRS, obtain channel estimation results, and be able to use the results for data demodulation after symbol #2. Thus, the communication node can transmit an acknowledgement/negative acknowledgement ( “ACK/NACK” ) back to the sender of the DMRS (e.g., gNB or UE) in the same slot as the corresponding data transmission, as shown in FIG. 3. Compared with Long Term Evolution ( “LTE” ) networks, the time gap between the receipt of data and the corresponding ACK/NACK report in NR networks is much shorter.
This front loaded DMRS scheme is very useful for communication nodes with low latency traffic. However, for other types of traffic that are not as sensitive to latency, additional DMRS can be used based on the front loaded DMRS for better channel estimation, especially for high Doppler scenarios. FIG. 4 shows an example of this scheme, where an additional DMRS of one symbol is included. The receiving communication node can carry out channel interpolation using the two DMRS symbols.
Some NR network proposals have also introduced multiple slot ( “multi-slot” ) scheduling in order to reduce the control channel overhead. What this means is that one control channel can schedule multiple slots for one communication node (e.g., a gNB can schedule multiple slots for one UE) , such that these slots share the same control signals and (typically) have the same scheduling parameters. As shown in FIG. 5, one control channel in slot 0 can be used to configure scheduling parameters for data transmission in slot 0, 1, 2, and 3. Usually, scheduling parameters include resource allocation, modulation and coding scheme ( “MCS” ) , DMRS related parameters, hybrid automatic repeat request ( “HARQ” ) , etc., which are similar to their LTE counterparts.
Section 1
Referring now to FIG. 6, according to an embodiment, the communication node that transmits the DMRS (e.g., a gNB) can inform the communication node receiving the DMRS (e.g., a UE) regarding a power offset between different DMRS (e.g., between a first DRMS and a second DMRS) . For multi-slot scheduling, if all of the scheduling parameters are same, there may be a tradeoff between flexibility and saving on control signaling overhead. For example, if the gNB 102 schedules UE #0 with multiple slots using one control channel, such as the physical downlink control channel ( “PDCCH” ) , it may use one PDCCH in slot 0 to configure all of UE #0’s slots and do so using the same resource allocation. Similarly, the gNB 102 can use the same DMRS pattern for each of the slots, and the power ratio between the DMRS and the data among different slots will be the same. This is the case in FIG. 6. This method is reasonable if the traffic load is low.
However, if the traffic load is high, and multiple users are going to be scheduled in one or more slots, then some scheduling parameters will not be the same for every slot. For
example, as shown in FIG. 7, UE #0 is scheduled using multi-slot scheduling and UE #1 is scheduled using single slot scheduling (in slot 1) . In this example, the resources allocated by the gNB 102 for UE #0 and UE #1 partially overlap in slot 1. The transmission power of the gNB 102 will be split between UE #0 and UE #1 in slot 1. Therefore, the power of the signal received by UE #0 in slot 1 may be lower than in other slots. If the power is split equally between UE #0 and UE #1 in the overlapping resources (i.e., in overlapping resource elements “REs” ) , and since the DMRS for UE #0 and UE #1 uses frequency-division multiplexing ( “FDM” ) but data for UE #0 and UE #1 map to same RE, then the power ratio between the DMRS and data for UE #0 in slot 1 is double the DMRS/UE #0 power ratio of other slots. In other words, there will be a 3 dB power boost in slot 1.
As can be seen, because of per-slot dynamic scheduling, a DMRS power imbalance may exist among different slots for a given instance of multi-slot scheduling. If there is no signaling to inform the UE of this variation when it occurs, the UE’s demodulation accuracy will be degraded significantly.
Section 1.1
According to an embodiment, a communication node transmitting the DMRS informs the communication node receiving the DMRS of these variations (e.g., when the first communication node schedules the second communication node using multi-slot scheduling) . There are a number of different ways in which this may be implemented.
In an embodiment, the first communication node (e.g., a gNB) informs the second communication node (e.g., a UE) regarding the variation among the different slots of the DMRS power/data power in the form of a power offset between the first slot and other slots. In this embodiment, the first communication node informs the second communication node via the control channel of the first slot.
In another embodiment, the first communication node communicates, via the control channel of the first slot, the DMRS power/data power for each of the multiple slots scheduled for the second communication node. This may require a large DCI overhead.
Section 1.2
According to an embodiment, the first communication node communicates the power offset or power ratio within the last slot (of the multiple scheduled slots) and this offset or power ratio applies to all slots except for the first slot (of the multiple scheduled slots) . For example, assuming the gNB schedules a UE for slots 0, 1, 2, and 3, the gNB transmits the power ratio or offset information for slots 1, 2, and 3 to the UE in the control channel of slot 3, but transmits the power ratio for slot 0 in the control channel of slot 0. In this embodiment, a dedicated DCI in the last slot can be provided for the first communication node to transmit these parameters. As shown in FIG. 8, the power offset or power ratio for slots 1, 2, and 3 is transmitted via a dedicated DCI in slot 3.
It should be noted that the dedicated DCI can be transmitted when needed, i.e., when a power difference between the first slot and the other slots exist. The UE would use blind detection to determine whether the dedicated DCI exists or not. Therefore, if all parameters are the same for all of the slots, there is no additional overhead needed for a dedicated DCI.
In various embodiments, a communication node can send other types of parameters besides power offset and power ratio using one or more of the techniques described herein. For example, the quasi-co-location ( “QCL” ) indication parameter may be different among slot 0 and other slots. Specifically, the information regarding the receive beam corresponding to the QCL information may change for different slots. Therefore, a first communication node (e.g., a gNB) can use the dedicated DCI to provide the QCL information for slot 1, slot 2, and slot 3.
Another example is reference signal parameters, e.g., CSI-RS or SRS trigger. In particular, the state of aperiodic CSI-RS/SRS triggering may be different for different slots. For example, assume that in slot 0 a gNB does not trigger the transmission of aperiodic CSI-RS/SRS in DCI., The gNB can use the dedicated DCI to trigger CSI-RS/SRS in slot 1, 2, or 3 because of a sudden requirement. Of course, the gNB can trigger different RS resources in slot 1 and slot 2, e.g., a different RS density between slot 1 and slot 2.
Section 1.3
In an embodiment, the first communication node transmits the changed parameters (if they change) for each slot within that slot using a dedicated DCI (which would be in the
dedicated PDCCH of each slot shown in FIG. 9) . Although this may result in multiple slots having a dedicated DCI, the impact on overhead may not be so significant, since the dedicated DCI would only contains a small number of parameters, and would therefore be the smaller than that of a normal DCI. It should be noted that the dedicated DCI can be transmitted when needed, i.e., when a power difference between the first slot and other slots exists. In such a scheme, the UE uses blind detection to determine whether the dedicated DCI exists or not. Also, if all parameters are same for the different slots, there is no additional overhead needed for a dedicated DCI.
Section 1.4
Turning to FIG. 10, according to an embodiment, the first communication node communicates the “changed” parameters to the second communication node via DCI after the slots of the multi-slot scheduling (shown as Slot n in FIG. 10) .
It is noted that power offset or power ratio information can be informed implicitly by other signals, e.g., different DMRS patterns may correspond to different power offset values. In such a case, the gNB can indicate different power offset values by using different DMRS patterns.
It is noted that the dedicated DCI can be transmitted when needed, i.e. power difference between first slot and other slots exist. Then UE need to blindly detect whether the dedicated DCI exists or not. Therefore, if all parameters are same for different slots (e.g., the MCS for data is the same among the different slots and the start position of the data in the different slots is the same) ) , there is no additional overhead for dedicated DCI.
According to various embodiments, a communication node can signal changes in other parameters besides the power offset and power ratio. Examples include one or more of following parameters: DMRS port index, DMRS density in the frequency domain, waveform parameters, rate matching parameters, reference signal parameters, modulation scheme parameters, code scheme parameters, modulation and coding scheme parameters, start position of data transmission parameters, start position of data transmission parameters, end position of data transmission parameters, duration of data transmission parameters, physical resource block
bundling size, scrambling sequence, power control parameters, quasi-co-location indication parameters, hybrid automatic repeat request process identifier parameters, and acknowledgement/not acknowledgement feedback timing parameters. It should be noted that resource allocation would remain the same among different slots for multi-slot scheduling. In other words, frequency resource bandwidth and positions do not change with time variation for multi-slot scheduling.
Section 2.0
There are situations in NR networks in which one type of traffic might need to preempt another type of traffic (either for different communication nodes or for the same communication node) . For example, the gNB 102 may allocate a set of resources to UE #0 (via a first message) , but then need to quickly reallocate those resources to UE #1 (i.e., preempt UE #0’s resources) because UE #1 is higher priority traffic in a particular slot. If so, the gNB 102 will need to inform UE #0 of this preemption (via a second message) , and do so after the preemption has already occurred. The allocated set of resources may be, for example, resources of a transport channel, such as code blocks, code block groups, or transport blocks.
For example, there are many different traffic types, and these different types can be for the same UE or different UEs, e.g., enhance Mobile Broadband ( “eMBB” ) traffic or uRLLC traffic. Compared to eMBB traffic, uRLLC needs much lower latency but incurs less traffic load. In order to reduce the transmission time of uRLLC communications, the corresponding slot format is usually much shorter than that of eMBB. For example, one slot of eMBB (which is the minimal scheduling unit for eMBB) is made up of 14 symbols or 7 symbols, but one slot of uRLLC (which is the minimal scheduling unit for uRLLC) is made up of only 1 or 2 symbols.
Since latency sensitivity for uRLLC traffic is higher than that of eMBB, some of the eMBB transmission resources may be preempted by uRLLC traffic. As shown in FIG. 11, when there is urgent uRLLC traffic within the eMBB transmission region, it will impact the eMBB transmission. Because the uRLLC traffic has higher priority, the resources preempted by the uRLLC traffic will not be used for eMBB transmission. The gNB informs the UE having the eMBB transmission that the resource locations or corresponding code blocks, or code block groups, or transmission blocks are preempted by other UEs. In order to reduce the complexity of
this scheme for the UE, the gNB may transmit an indication of the preemption after the preempted resources, e.g., the gNB informs the UE having the eMBB traffic in the PDCCH of the next slot.
Referring to FIG. 12, for example, UE #0 with eMBB traffic is scheduled in slot 0, where 14 symbols and a total of 5 code block groups ( “CBGs” ) are transmitted in the data region and where CBG 0-4 correspond to the resources of orthogonal frequency division multiplexing ( “OFDM” ) symbols (3, 4) , (5, 6) , (7, 8) , (10, 11) , and (12, 13) respectively. Meanwhile, UE #1 with uRLLC traffic is scheduled in symbol (5, 6) , which corresponds to the resources of CBG 1 of UE #0. Therefore, the CBG 1 will not be received correctly by UE #0. Thus, the gNB 102 uses the PDCCH in slot 1 to inform UE #0 that CBG 3 is preempted (or that symbols (5, 6) are preempted) in slot 0. Moreover, instead of waiting for ACK/NACK feedback of the slot 0 transmission, the gNB 102 retransmits CBG 1 in slot 1.
In an embodiment, for multi-slot scheduling, a communication node transmits an indication of the preemption for all of the multiple slots in one control signaling region after the multiple slots, e.g., before ACK/NACK feedback corresponding to the transmission in these multiple slots is due to be received. Moreover, all CBGs corresponding to the preempted traffic in these multiple slots can be re-scheduled via the control signaling region. As shown in FIG. 13, the gNB transmits the grant for the multi-slot scheduling (including slots 0, 1, 2, and 3) via the PDCCH in slot 0. Note that five CBGs are still assumed in each slot, for a total 20 CBGs in four slots, so the preemption indication that the gNB transmits may need to be 20 bits. In an embodiment, the preemption indications is transmitted in the PDCCH of slot 3+n, and each bit corresponds to each CBG, where 0 means the corresponding CBG is preempted by other traffic, and 1 means the corresponding CBG is not preempted by other traffic. In FIG. 13, two CBGs are preempted and the gNB reschedules these two CBGs in slot 3+n, where n is a positive integer. For simplicity, n can be equal to 1.
It should be noted that since the purpose of the preemption indication is to inform the UE which transmission part is not transmitted or which part is punctured, the preemption indication is actually a rate matching indication, and will sometimes be referred to as such in this disclosure.
If an equal number of CBGs is assumed in each slot, the payload requirements of the preemption indications are different for single slot scheduling versus multiple slot scheduling. Therefore, the preemption indication size is related to the number of scheduling slots. In other words, the payload of required for the control signal to indicate preemption at least depends on the number of scheduling slots. Since the number of slots aggregated (i.e., scheduled according to multi-slot scheduling) may be dynamic, the preemption indication size is dynamically related to the number of scheduling slots. In other words, the payload of the control signal needed to indicate preemption at least dynamically depends on the number of scheduling slots. In an embodiment, the preemption indication also informs the UE which time/frequency domain resources are impacted or which CBGs are informed.
Even for single slot scheduling, the number of CBGs may not always be same for different slots. Like the preemption indication, the payload of the control signaling used to indicate these parameters at least depends on some of following: slot format, start position of the data transmission, duration (or end position) of data transmission, allocated resource bandwidth, MCS, and the number of CBGs.
Regarding the slot format, one slot usually is composed of some symbols for the downlink ( “DL” ) control channel, some symbols for DL data transmission, some symbols for guard period ( “GP” ) , and some symbols for UL data or the control channel. The relative mix of these components is based on the slot format. Since different slot formats have different numbers of symbols for data transmission, related frequency/time domain resources or the number of CBGs may be different and further impact the payload size of the indication.
Regarding the start position of data transmission, for one slot scheduling, the data region may not be adjacent to the PDCCH, and the gNB may inform the UE of the start position of data transmission in PDCCH. Also, different data transmission start positions may lead to different numbers of CBGs or different numbers of time/frequency resources, which impact the payload size of the indication.
Section 2.1
Turning to FIG. 15, according to an embodiment, a communication node (e.g., the gNB) can inform another communication node (e.g., the UE) of rate matching parameters in the data region of last slot of the multiple slots.
In an embodiment, a communication node (e.g., the gNB) can inform or update some configurations for each slot for multi-slot scheduling another communication node.
Section 3
As discussed above in Section 1.1, the DMRS/data power ratio may be different among multiple slots and a communication node (e.g., the gNB) may make dynamic adjustments to accommodate this. In addition to the DMRS/data power ratio, some of following parameters may also be reconfigured by the communication node in later communication slots, according to various embodiments:
(1) DMRS parameters including at least one of: DMRS port number, total number of DMRS ports, DMRS density in frequency domain, scrambling ID, and DMRS pattern. For DMRS port number or DMRS port index, the port number can change from one slot to another slot, e.g., in slot 0, DMRS ports 0 and 1 are allocated to UE #0, and ports 2 and 3 are allocated to UE #1; but in slot 1, DMRS ports 2 and 3 are allocated to UE #0 and ports 0 and 1 are allocated to UE #1. In embodiment, the port index can be changed by a predefined mechanism, making an additional control signal unnecessary. This method is also suitable for semi-persistent scheduling ( “SPS” ) scheduling.
According to various embodiments, the port index can be reconfigured using any of the methods discussed in Sections 1.1, 1.2, 1.3, 1.4, and 2.1.
According to various embodiments, different total numbers of DMRS ports can be reconfigured using any of the methods discussed in Sections 1.1, 1.2, 1.3, and 1.4.
According to various embodiments, the DMRS density in the frequency domain can be reconfigured using any of the methods discussed in Sections 1.1, 1.2, 1.3, 1.4, and 2.1.
(2) Scrambling ID, e.g., the n number of the scramble identity ( “nSCID” ) domain for DMRS sequence generation. According to various embodiments, the scrambling ID can be
reconfigured using any of the methods discussed in Sections 1.1, 1.2, 1.3, 1.4, and 2.1, or by a predefined mechanism. As example of a predefined mechanism is as follows: If there are 2 scrambling IDs, 0 and 1, then 0 and 1 can be used for even slots and odd slots respectively.
According to various embodiments, the DMRS pattern can be reconfigured using any of the methods discussed in Sections 1.1, 1.2, 1.3, 1.4, and 2.1. For example, in slot 0, only the front loaded DMRS is configured, as shown in FIG. 3, but both the front loaded DMRS and the additional DMRS is configured in slot 1, as shown in FIG. 4.
(3) Waveform. Usually, two waveforms can be used in uplink transmission: cyclic prefix OFDM ( “CP-OFDM” ) and Discrete Fourier transform spread OFDM ( “DFT-S-OFDM” ) . To achieve flexibility, different waveforms can be reconfigured for different slots using any of the methods discussed in Sections 1.1, 1.2, 1.3, 1.4, and 2.1
(4) Parameters on rate matching (e.g., preemption indication )
(5) Zero power channel state information reference signal ( “CSI-RS” ) configuration can be reconfigured for different slots using any of the methods discussed in Sections 1.1, 1.2, 1.3, and 1.4.
(6) Other reference signal triggers, including aperiodic CSI-RS, phase tracking reference signal ( “PTRS” ) , aperiodic SRS, etc.
(7) MCS.
(8) Modulation scheme, e.g., binary phase-shift key ( “BPSK” ) , quadrature phase shift keying ( “QPSK” ) .
(9) Code scheme mean code rate.
(10) Start position of data transmission.
(11) End position of data transmission.
(12) Duration of data transmission.
(13) Physical resource block ( “PRB” ) bundling size.
(14) Scrambling sequence.
(15) Power control parameters.
(16) Sounding Reference Signal ( “SRS” ) parameters.
(17) Quasi-co-location ( “QCL” ) indication.
(18) HARQ process ID.
(19) ACK/NACK feedback timing.
Section 4
If should be noted that, even in the case of single slot scheduling (e.g., FIG. 14) , multiple parameters can be configured. Referring again to FIG. 4 and the accompanying description (in which a front loaded DMRS transmitted in symbol 2 and an additional DMRS is transmitted in symbol 10) , any of the methods discussed in Sections 1.1, 1.2, 1.3, and 1.4 can be used to reconfigure the power of the additional DMRS.
It should be understood that the exemplary embodiments described herein should be considered in a descriptive sense only and not for purposes of limitation. Descriptions of features or aspects within each embodiment should typically be considered as available for other similar features or aspects in other embodiments. It will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from their spirit and scope of as defined by the following claims. For example, the steps of the various methods can be reordered in ways that will be apparent to those of skill in the art.