Method and system for 5g PHY API procedures and messages for DMRS-BF capable o-ran based ran

Abstract

In a Radio Access Network system, a distributed unit (DU) includes a PHY layer LI, a MAC layer L2, and a PHY Application Program Interface configured to transfer DeModulation Reference Signal (DMRS) data between the PHY layer LI to the MAC L2 layer, and an RU comprising the PHY layer LI The DU is configured to provide DMRS -based beamforming (DMRS-BF) information to the RU.

Description

METHOD AND SYSTEM FOR 5G PHY API PROCEDURES AND MESSAGES FOR DMRS-BF CAPABLE O-RAN BASED RAND DESCRIPTION OF RELATED ART

[0001] The present disclosure relates to systems and methods for radioaccess networks. In particular, the present disclosure relates to the design of operation, administration and management of various network elements of 4G, 5G, and further Generations.

[0002] Mobile networks, such as 5G NR standardized by 3GPP form a RadioAccess Network (RAN) and the Core Network (a.k.a. Core). The RAN provides wireless connectivity between the user equipment (such as a mobile phone) and the Core. Traditionally, the RAN is an integrated unit developed by a single vendor. Although the traditional RAN provides enhanced capabilities with proprietary solutions, mobile network operators prefer a more diverse vendor ecosystem. Therefore, the Open RAN (O-RAN) standard is being developed to create a multi- vendor RAN solution with the promised benefits of supply chain diversity, network flexibility, lower cost, and new capabilities leading to increased competition and further innovation.

[0003] With the O-RAN standard, the RAN is disaggregated into three mainbuilding blocks: radio unit (RU), distributed unit (DU), and centralized unit (CU). The O-RAN alliance standardizes the protocols and interfaces between these units. The radio frequency (RF) signals are transcieved, amplified, and digitized in the RU. The DU and CU are the computation parts of the base station, sending the digitalized signal into the mobile network. The specific RAN functionalities that correspond to each of these three entities are determined by the split points, and the O-RAN alliance selected the split option 7-2x for the fronthaul (FH) interface design between a DU and an RU. The physical layer (PHY) functions are distributed between the RU and DU with respect to the O-RAN split option 7-2x. For example, weight-based dynamic beamforming (WDBF) Open-RUs (O-RUs) perform low PHYfunctionalities (e.g., FFT / IFFT and beamforming), whereas Open-DU (O-DUs) perform high PHY functionalities (e.g., channel estimation, equalization, and demodulation) as shown in Figure 1. SUMMARY

[0004] Described are implementations of a system, a method, and a computerprogram products for a Radio Access Network system. In an implementation, a method comprises: configuring distributed unit DU to provide DeModulation Reference Signal (DMRS)-based beamforming (DMRS-BF) information to a radio unit (RU) on a slot-by-slot basis for a user group; extracting, at the RU, DMRS data from and Physical Uplink Shared Channel (PUSCH) data from the DMRS-BF information, the extracted DMRS data comprising channel estimation and beamforming weights calculation, equalization, and demodulation; computing beamforming weights from the extracted DMRS data; and applying the beamforming weights to the PUSCH data.

[0005] The method can further comprise:receiving, at the RU, DMRS and sound reference signal (SRS) messages from User Equipment (UE); processing the DMRS data at a physical (PHY) layer L1 at the RU or the DU; transferring the processed DMRs data to a medium access control (MAC) L2 or higher layer via an PHY Application Program Interface (API); deciding, at the MAC L2 or higher layer a selected uplink (UL) beamforming method; conveying the selected UL beamforming method to the DU via the PHY API; and conveying the selected UL beamforming method to the RU so that the RU can apply the selected UL beamforming method for the UE.

[0006] The method can further comprise:configuring the RU to support advanced PHY configurations selected from the group of: a delay profile between the RU and the DU, wherein the delay profile comprises parameters including subcarrier spacing, carrier bandwidth, and beamforming method; symbol reordering for DMRS symbols; RMM measurements; post processing of SINR, wherein time and frequency of SINR are adjusted to optimize the system performance based on throughput performance, computational complexity, and FH bit rate tradeoff; or dimension reduction for reducing a dimension for processing the DMRS at the RU; or any combination thereof.

[0007] The advanced PHY configurations can include further capabilities andconfigurations for DMRS and scheduling selected from the group of: limits and capabilities on a number of the UEs, UE layers, user groups; a maximum number of ueIds, maximum number of ueIds per slot, and maximum number of user groups per slot; a maximum number user groups per UE in time; a maximum number of DMRS configurations per user group; lists of supported “DMRS Symbol Mask,” “Hopping mode,” and “Low PAPR Type” values; or support for the use of “ueId persistence” across multiple slots; or any combination thereof.

[0008] The method can further comprise: configuring the PHY API to conveya message including the advanced PHY parameter between the PHY Layer L1 to the MAC L2 or higher layer.

[0009] The method can further comprise: conveying a capability messagefrom the DU PHY to the DU MAC or conveying a configuration message from the-DU MAC to the DU PHY.

[0010] The RMM measurement can be selected from the group of: UE TimingAdvance Error (TAE), UE layer signal power, UE frequency offset, Interference plusnoise (IpN) for allocated PRBs, IpN for unallocated PRBs, or Antenna DMRS SNR, orany combination thereof. The RRM measurements can be accompanied by a symbol mask to indicate associated relevance in time-frequency resources.

[0011] The time and frequency resolution of SINR can be static whenconfigured through an M-plane or dynamic when configured through C-a plane. A compression method and block size can be adjusted with the time and frequency resolution of SINR.

[0012] In an implementation, a Radio Access Network system comprises:a distributed unit DU comprising a physical (PHY) layer L1, a medium access control (MAC) layer L2, and a PHY Application Program Interface (API) configured to transfer DeModulation Reference Signal (DMRS) data between the PHY layer L1 to the MAC L2 layer; and a radio unit RU comprising the PHY layer L1; wherein the DU is configured to provide DMRS-based beamforming (DMRS-BF) information to the RU.

[0013] The system can further comprise:DU being configured to provide the DMRS-BF information to the RU on a slot-by-slot basis for each user group; and the RU being configured to extract DMRS data from and PUSCH data from the DMRS-BF information, the extracted DMRS data comprising channel estimation and beamforming weights calculation, equalization, and demodulation; compute beamforming weights from the extracted DMRS data; andapply the beamforming weights to the PUSCH data.

[0014] The system can be configured so that:the RU is configured to receive DMRS and SRS messages from User Equipment (UE); the RU and / or DU is configured to process the DMRS data at a PHY layer L1 and transfer the processed DMRS data to the MAC L2 or higher layer via a PHY API; the MAC L2 or higher layer is configured to a select uplink (UL) beamforming method and convey the selected UL beamforming method to the DU via the PHY API; and the DU is configured to convey the selected UL beamforming method to the RU so that the RU can apply the selected UL beamforming method for the UE.

[0015] The system can be configured so that:the RU is configured to support advanced PHY configurations selected from the group of: a delay profile between the RU and the DU, wherein the delay profile comprises parameters including subcarrier spacing, carrier bandwidth, and beamforming method; symbol reordering for DMRS symbols; RMM measurement; post processing of SINR, wherein time and frequency of SINR are adjusted to optimize the system performance based on throughput performance, computational complexity, and FH bit rate tradeoff; or dimension reduction for reducing a dimension for processing the DMRS at the RU; or any combination of thereof.

[0016] The system can be configured so that advanced PHY configurationsinclude further capabilities and configurations for DMRS and scheduling selected from the group of: limits and capabilities on a number of the UEs, UE layers, user groups; a maximum number of ueIds, maximum number of ueIds per slot, and maximum number of user groups per slot; a maximum number user groups per UE in time; a maximum number of DMRS configurations per user group; lists of supported “DMRS Symbol Mask,” “Hopping mode,” and “Low PAPR Type” values; or support for the use of “ueId persistence” across multiple slots; or any combination of thereof.

[0017] The system can be configured so that the PHY API is configured toconvey a message including the advanced PHY parameter between the PHY Layer L1 to the L2 / L3 layer.

[0018] The system can be configured so that the RMM measurement isselected from the group of: UE Timing Advance Error (TAE), UE layer signal power, UE frequency offset, Interference plus noise (IpN) for allocated PRBs, IpN for unallocated PRBs, or Antenna DMRS SNR or any combination thereof. The RRM measurements can be accompanied by a symbol mask to indicate associated relevance in time-frequency resources.

[0019] The system can be configured so that the time and frequencyresolution of SINR are static when configured through an M-plane or dynamic when configured through C-a plane. A compression method and block size can be adjusted with the time and frequency resolution of SINR. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a Block Diagram of an O-RAN Uplink Massive MIMO Systemwith WDBF O-RU.

[0021] Figure 2 is a Block Diagram of an O-RAN Uplink Massive MIMO Systemwith DMRS-BF-NEQ O-RU.

[0022] Figure 3 is a Block Diagram of an O-RAN Uplink Massive MIMO Systemwith DMRS-BF-EQ O-RU.

[0023] Figure 4 is a Block Diagram of 5G FAPI Architecture.

[0024] Figure 5 is a Block Diagram of the PHY API Interactions.

[0025] Figure 6 is a Block Diagram of an implementation of a PHY API forAdvanced O-RAN.

[0026] Figure 7 is a Block Diagram of an O-RAN Architecture with DifferentPHY and MAC Vendors.

[0027] Figure 8 shows an Example Call Flow for the “UL BeamformingMethod” Selection. DETAILED DESCRIPTION

[0028] O-RAN based on 7-2x architecture enables the utilization of differentPHY / MAC algorithms along with a wide variety of parameter selections. For example, O-RAN based on 7-2x architecture allows multiple digital beamforming methods at O-RU. One of the uplink beamforming methods is weight-based dynamic beamforming (WDBF), and another method is DMRS-based beamforming (DMRS- BF).

[0029] As of the present disclosure, DMRS-BF is a new O-RAN feature.Accordingly, new PHY API procedures and messages between PHY and MAC layers are needed to enable and optimize DMRS-BF capable O-RAN-based Radio Access Networks, considering different algorithm and parameter selections. DMRS-based beamforming (DMRS-BF) has been approved by Working Group 4 (WG4) of the O- RAN Alliance with the objective of providing uplink performance improvement, especially in high mobility and / or high interference scenarios. The key differencebetween DMRS-BF and its typical predecessor, Weight-based Dynamic Beamforming (WDBF), is that some of the PHY layer functionalities on the O-DU moved to the O- RU, which include DMRS channel estimation, DMRS-based beamforming, and / or equalization functions, as shown in Figures 2 and 3. Accordingly, the O-RAN FH interface between O-DU and O-RU has been changed to enable DMRS-BF in June 2024 [1].’

[0030] For typical scenarios, WDBF provides sufficient throughputperformance while balancing FH bit rate consumption and O-RU computational complexity. However, in certain scenarios, such as high mobility and high interference, DMRS-BF can be preferable, although it increases computational complexity and O-RU energy consumption. Please note that it is only one of the examples among tremendous algorithm and parameter choices.

[0031] Also, within the RAN, there are multiple internal interfaces betweendifferent network components. These interfaces are being standardized as well. For example, the functional application platform interface (FAPI) initiative provides a common API with the goal of establishing interoperability and innovation among suppliers of platform hardware, platform software, and application software. It defines internal RAN interfaces and is agnostic to RAN architecture. For 5G NR, the FAPI suite comprises multiple specification documents for various APIs. Among these specification documents, the 5G PHY API specification defines a control interface (P5) and a user plane or data path interface (P7) for an API between the medium access control (MAC) and PHY protocol layers, as shown in Figures 4 and 5.

[0032] The context of this disclosure is related to 5G PHY API Procedures andMessages for DMRS-BF Capable O-RAN based Radio Access Networks, considering the recent and potential DMRS-BF related future advances.

[0033] [1] O-RAN.WG4.CUS.0-R003-v15.00.01, “O-RAN Working Group 4(Open Fronthaul Interfaces WG) Control, User and Synchronization Plane Specification,” June 2024.

[0034] There are four planes specified in the O-RAN specifications, namely,control plane (C-plane), user-plane (U-plane), synchronization plane (S-plane), and management plane (M-plane). The main purpose of the C-plane messages is to transmit data-associated control information required for the processing of user data (e.g., scheduling and beamforming commands). In contrast to the C-plane, the M-Plane refers to non-real-time management operations (e.g., configuration and capability commands) between the O-DU and the O-RU.

[0035] Accordingly, CUS / M-Plane specifications have been updated toaccommodate DMRS-BF [1]. In this method, briefly, the O-DU provides DMRS configuration to the O-RU on a slot-by-slot basis for each user group. Then, the O-RU uses that information to extract DMRS data from uplink PUSCH data, computes beamforming weights using the extracted DMRS data, and finally applies these weights to the associated PUSCH data. Currently, there are two types of DMRS-BF as DMRS-BF-NEQ and DMRS-BF-EQ. The beamforming weights do not include an equalization function in DMRS-BF-NEQ, whereas, they include an equalization function in DMRS-BF-EQ. Please note that this beamforming method leads to a change in the functional split between O-DU and O-RU. Also, more functionalities, such as SRS processing, can be present in the O-RUs in the future.

[0036] Figure 2 shows the high-level block diagram of a UL (m)MIMO O-RANsystem with an O-RU that has DMRS extraction and DMRS processing capabilities. This O-RU is referred to as DMRS-BF-NEQ O-RU. Basically, O-DU sends the necessary DMRS information, which is required to extract DMRS, to O-RU and O-RU extracts and processes DMRS to improve performance by replacing / improving SRS-based beamforming weights with DMRS-based beamforming weights. Moreover, O-RU can send RRM measurements, (a.k.a. non-port reduced DMRS measurements), which is not possible to calculate at O-DU with high precision, to O-DU to improve the performance further.

[0037] Figure 3 shows the high-level block diagram of another UL (m)MIMOO-RAN system with an O-RU that has DMRS extraction, DMRS processing, andequalization capabilities. This O-RU is referred to as DMRS-BF-EQ O-RU. Similar to DMRS-BF-NEQ O-RU, DMRS-BF-EQ O-RU extracts and processes DMRS to improve performance. It also performs equalization and sends the equalized signal stream(s) along with the supplementary demodulation information (e.g., post-processing SINR) to O-DU. Moreover, O-RU can send RRM measurements, (a.k.a. non-port reduced DMRS measurements), which is not possible to calculate at O-DU with high precision, to O-DU to further improve the performance.

[0038] At a high level, DMRS-BF operates as follows:1. An O-DU configures its beamforming method to DMRS-BF. Currently, it iseither DMRS-BF-NEQ or DMRS-BF-EQ due to an O-RU configuration restriction in the O-RAN CUS-plane specification. In the future, an O-DU can be configured to use both types simultaneously. 2. The O-RU declares support for delay profiles associated with specificbeamforming methods, including DMRS-BF-NEQ and DMRS-BF-EQ. 3. The O-RU declares support for DMRS-BF specific capabilities, includingthe symbol reordering method, RRM measurement capabilities as well as their parameter values, SINR resolution, ueId format, and symbolId meaning. 4. The O-RU declares support for various DMRS-BF features, including limitson the number of UEs, UE layers, user groups, and optional capabilities such as simultaneous symbol reordering and DMRS symbol transfer. 5. The O-DU configures O-RU endpoints to use a specific delay profile alongwith its associated beamforming method. Currently, only one of the two DMRS-BF methods can be used at a time in the whole O-RU due to an O- RU configuration restriction in the O-RAN CUS-plane specification. In the future, an O-RU can be configured to use both DMRS-BF types simultaneously. 6. The O-DU configures all DMRS-BF-EQ endpoints if the required capabilityof reporting port-reduced DMRS symbols is used or not.The O-DU configures the use of DMRS-BF specific capabilities, includingthe symbol reordering method, RRM measurements to be reported with relevant parameters, SINR resolution, and various other DMRS-BF related configurations.For each slot, O-DU creates C-plane messages using various sectionextensions (SEs) and section types (STs), including ST 5 describing the scheduling decisions for each user group, which can also include SE 10 (e.g., if there is more than one UE layer in the user group), SE 24 (to describe the DMRS configurations for the user group), SE 25 (if symbols are to be reordered), and use SE 26 (if frequency offset data is to be sent from O-DU to O-RU).For each slot, the O-RU executes UL DMRS-BF as follows:a. The O-RU collects DMRS symbols and can perform dimensionreduction with or without any explicit control by the O-DU. The O- DU controlled dimension reduction is enabled through SE 27. b. The O-RU computes a channel estimate for each UE layer based onthe received DMRS symbols. In general, this operation can span multiple user groups if the UE spans multiple user groups. c. The O-RU computes weights to apply to each UE layer. The specificweight computation is different for DMRS-BF-NEQ and DMRS-BF- EQ. The DMRS-BF-EQ weights include the equalization function as well. d. The O-RU applies the computed weights to each UE layer’s PUSCHdata. If the O-DU configures the O-RU to compute and send port- reduced DMRS data, then the O-RU also applies the weights to the DMRS symbols and creates “port-reduced DMRS symbols”.The O-RU sends each UE layer’s U-plane data to the O-DU. The symbolscan optionally be reordered or reassigned and configured by the O-DU.The O-RU sends RRM measurements if supported and configured by theO-DU.12. The O-RU sends SINR data if supported by the O-RU and configured by theO-DU.

[0039] In addition to the new CUS / M-plane messages between O-DU and O-RU, new PHY API procedures and messages between the physical (PHY) and medium access control (MAC) protocol layers are required to enable and optimize DMRS-BF (and possible future DMRS-BF related enhancements) for DMRS-BF capable O-RAN-based Radio Access Networks, considering different algorithm and parameter selections. Currently, the FAPI suite comprises multiple specification documents, including the 5G PHY API specification. Accordingly, the new PHY API procedures and messages for advanced O-RAN architectures can be included to the control interface (P5) and the user plane or data path interface (P7), which are defined in the 5G PHY API specification, as shown in Figure 6.

[0040] As noted above, a difference between DMRS-BF and its conventionalpredecessor, Weight-based Dynamic Beamforming (WDBF), is that some of the PHY layer functionalities on the O-DU moved to the O-RU, which include DMRS channel estimation, DMRS-based beamforming, and / or equalization functions. In addition to the new CUS / M-plane messages between O-DU and O-RU through the O-RAN FH interface, new PHY API procedures and messages between the physical (PHY) and medium access control (MAC) protocol layers are required to enable and optimize DMRS-BF (and possible future DMRS-BF related enhancements) for advanced O- RAN-based radio access networks.

[0041] This disclosure provides 5G PHY API procedures and messages toenable and optimize DMRS-BF capable O-RAN based Radio Access Networks, according to the recent and ongoing DMRS-BF related advances in the O-RAN architecture. A common API enables establishing proper interoperability among suppliers of different suppliers such as O-DU PHY (L1) and O-DU MAC (L2) is implemented. Implementations are configured to deploy different PHY / MAC algorithms along with a wide variety of parameter selections considering throughput, energy consumption, and fronthaul (FH) bit rate tradeoffs.

[0042] As shown in Figure 7, O-DU MAC (L2), O-DU PHY (L1), and O-RU PHY(L1) can be from different vendors and hence they might have different capabilities and limitations. In addition to the CUS / M-plane messages between O-DU and O-RU through the O-RAN FH interface, PHY API procedures and messages between the physical (PHY) and medium access control (MAC) protocol layers advantageously enable and optimize DMRS-BF (including ongoing DMRS-BF related enhancements) for advanced O-RAN-based radio access networks. It is especially advantageous to deploy different PHY / MAC algorithms along with a wide variety of parameter selections considering throughput, energy consumption, and fronthaul (FH) bit rate tradeoffs.

[0043] The DMRS-BF related implementations are as follows:

[0044] (a) UL Beamforming:

[0045] O-RUs with advanced physical layer (PHY) capabilities, such as DMRS-BF capable O-RUs, can have DMRS extraction, DMRS processing (such as channel estimation and beamforming weights calculation), equalization, and demodulation features. Figures 2A-2B (DMRS-BF-NEQ) and Figures 3A-3B (DMRS-BF-EQ) show the high-level functional block diagrams of (m)MIMO O-RAN systems using O-RUs with such advanced PHY capabilities.

[0046] Although implementation of PHY capabilities as described herein at O-RU can improve the performance of (m)MIMO O-RAN systems, they might not be required in all scenarios. For example, mobility and / or interference levels can be low in a network, and implementations of these PHY features at O-RU might not improve the performance significantly. Therefore, different UL beamforming methods can be selected / configured or co-exist and multiplexed (such as WDBF and DMRS-BF) over different time and frequency resources, depending on the scenario.

[0047] Currently, there are six UL beamforming methods in the O-RAN CUS-plane specification (and ongoing available options can be extended). The UL beamforming method information can be exchanged between PHY and MAC througha PHY API. This information, which is conveyed through the PHY API, can be a capability message, which is typically sent from O-DU PHY to O-DU MAC, or can be a configuration message, which is typically sent from O-DU MAC to O-DU PHY (i.e., in the opposite direction). Please note that O-DU MAC can receive capability messages from O-RU through the O-RAN FH interface as well. A sample UL beamforming method selection / configuration message, which can be appended to the PUSCH PDU, is given below: Tag Field Type Description 0x0000 beamformingType Uint8_t Configured beamforming method 000 = Predefined-beam forming (PDBF) 001 = Weight-based dynamic beamforming (WDBF) 010 = Attribute-based dynamic beamforming (ABBF) 011 = Channel-information based beamforming (CIBF) 100 = DMRS-based beamforming, no equalization (DMRS-BF-NEQ) 101 = DMRS-based beamforming with equalization (DMRS-BF-EQ) Table 1: A sample message for “UL Beamforming Method Selection”

[0048] Also, a sample call flow for the UL beamforming method selection isshown in Figures 8A-8B. The UEs can send their DMRS and SRS signals to the RAN, which are captured by O-RU. Depending on the O-RU and O-DU capabilities, these signals can be processed in O-RU, O-DU, or in both entities to evaluate the channel conditions. The PHY (L1) of the O-RU or O-DU (depending on the capability) can perform channel estimation and extract measurements from DMRS and / or SRS signals. Afterward, these can be transferred to O-DU MAC and higher layers (L2 / L3) through a PHY API, and L2 / L3 of O-DU can make a decision for the UL beamforming method selection. This UL beamforming method selection is conveyed to the O-DU PHY through the PHY API, which is followed by a transfer to the O-RU through the O-RAN FH interface. In this way, the O-RU can perform UE-specific UL beamformingfor different time and frequency resources, depending on the scenario. Please note that this just an example and the “UL beamforming method” information can be exchanged between O-DU PHY and O-DU MAC through a PHY API in different manners.

[0049] (b) O-RAN Delay Profile:

[0050] O-RUs with advanced PHY capabilities can support multiple delayprofiles between O-DU and O-RU. O-RAN delay profile refers to an M-plane configuration that includes parameter settings for different combinations of subcarrier spacing, carrier bandwidth, and beamforming method. Different delay profiles are needed for various PHY configurations (e.g., BW and subcarrier spacings), efficient implementation of beamforming (e.g., WDBF and DMRS-BF), and service (e.g., eMBB, URLLC, and mMTC) multiplexing.

[0051] The O-RAN delay profile information can be exchanged between O-DUMAC (L2) to O-DU PHY (L1) through a PHY API. This information, which is conveyed through the PHY API, can be a capability message, which is typically sent from O-DU PHY to O-DU MAC, or can be a configuration message, which is typically sent from O- DU MAC to O-DU PHY (i.e., in the opposite direction). Please note that O-DU MAC can receive capability messages from O-RU through the O-RAN FH interface as well. An example sample message for the O-RAN delay profile selection / configuration is given below: Tag Field Type Description 0x0000 delayProfileType Uint8_t Configured delay profile 00 = DP-1 (e.g., for WDBF, eMBB, 30 kHz SCS, 100 MHz BW) 01 = DP-2 (e.g., for WDBF URLLC, 60 kHz SCS, 20 MHz BW) 10 = DP-3 (e.g., for DMRS-BF, eMBB, 30 kHz SCS, 100 MHz BW) Section K: DP-4 (e.g., for DMRS-BF, URLLC, 60 kHz SCS,20 MHz BW) Table 2: A sample message for “O-RAN Delay Profile Selection”

[0052] As will be appreciated, this message is an example, and the list can beextended further to different O-RAN delay profiles.

[0053] (c) Symbol Reordering

[0054] O-RUs with advanced PHY capabilities can support symbol reorderingto decrease latency. For example, considering DMRS-BF-NEQ, DMRS-based processing is performed both at O-RU and O-DU. Prioritizing the DMRS symbols in the O-RAN FH decreases the latency for O-DU DMRS processing. Another example is that if there are prioritized UEs in the network, these UEs’ PUSCH data can be sent earlier than others in the O-RAN FH interface to achieve lower latencies for the selected UEs.

[0055] Symbol reordering can be achieved in multiple ways, such as per-window symbol reordering and per-section tx-window reassignment. The selected symbol reordering information (such as the symbol reordering method and its associated parameters, including reordering / reassignment patterns) can be conveyed from L2 to L1 through a PHY API. This information, which is conveyed through the PHY API, can be a capability message, which is typically sent from O-DU PHY to O-DU MAC, or can be a configuration message, which is typically sent from O- DU MAC to O-DU PHY (i.e., in the opposite direction). Please note that O-DU MAC can receive capability messages from O-RU through the O-RAN FH interface as well. Also, similar to the other DMRS-BF features, various capabilities for symbol reordering, such as supported symbol reordering methods or the maximum number of different symbol reordering and reassignment patterns, can be exchanged through the PHY API. Tag Field Type Description 0x0000 symreorderingType Uint8_t Configured symbol reordering method 0 = per-window symbol reordering Section A: = per-section tx-windowreassignment Table 3: A sample message for “Symbol Reordering Method Selection”

[0056] If there is no such message, the symbols can be transmitted in “naturalorder” through the O-RAN FH interface. Along with the symbol reordering method selection, the specific symbol reordering or reassignment pattern can be exchanged through the PHY API. Note that these are example messages, and the list can be extended further with other “symbol reordering” related information.

[0057] (d) RRM Measurements

[0058] O-RUs with advanced PHY capabilities can support RRMmeasurements. The RRM measurements, which are also known as “non-port reduced DMRS measurements” can be utilized in higher layers (e.g., MAC) to improve the performance. These measurements can be listed, along with their brief descriptions from the O-RAN CUS plane specification, as follows: i. UE Timing Advance Error: UE Timing Advance Error (TAE) is measured onPUSCH DMRS and is reported to have one TAE value per UE per slot. The TAE is defined at the antenna reference point and represents the average TAE over all array elements and all used PRBs. ii. UE Layer Signal Power: UE layer signal power measurement is measured onPUSCH DMRS and is reported with one power value per DMRS port (associated with a UE layer) per slot. The layer signal power is defined at the antenna reference point and represents the linear average of signal power over all PRBs received per DMRS port of a UE, i.e., the received signal power per layer of a UE summed over all array elements and averages all used PRBs. iii. UE Frequency Offset: UE frequency offset is measured on PUSCH DMRS foreach UE and is reported with one value per UE per slot. The UE frequency offset is defined at the antenna reference point and represents the average frequency offset over all array elements and all used PRBs.iv. Interference Plus Noise for Allocated PRBs: Interference plus noise (IpN) forallocated PRBs is measured on PUSCH DMRS and is reported with one value per PRB, either for one symbol, or as an average over multiple symbols. It is defined at the antenna reference point and represents the per-PRB linear sum of inter-cell interference and noise power over all array elements. Please note that the IpN measurement for allocated PRBs is useful for the scheduler in the O-DU to optimize scheduling decisions.v. Interference Plus Noise for Unallocated PRBs: Interference plus Noise forunallocated PRBs is measured on uplink PRBs where no UE is scheduled and is reported with one value per PRB, either for one symbol or as an average over multiple symbols. It is defined at the antenna reference point and represents the per-PRB linear sum of inter-cell interference and noise power over all array elements. The measurement report can include a symbolMask field indicating over which DMRS symbols the IpN was averaged.vi. Measurement of DMRS SNR Per Antenna: Antenna DMRS SNR is measured onPUSCH DMRS and is reported to have one value per layer (assigned DMRS- port) per antenna per slot. Here, an “antenna” can either refer to an array element or an element after any O-RU internal dimension reduction. The measurement is defined at the antenna reference point. It represents, for each O-RU antenna (or O-RU dimension reduced antenna), the ratio between DMRS signal power per layer and the power of inter-cell interference plus noise. The reported value is a linear average over all DMRS REs of the assigned DMRS port in the PRB range(s) where the layer is scheduled. Also, it should be pointed out that compared to WDBF O-RAN or traditional RAN, DMRS SNR measurement is a non-port reduced measurement (i.e., before the UL beamforming operation).

[0059] These measurements are mandatory to report for DMRS-BF-EQexcept (vi). On the other hand, these measurements are optional to report for DMRS-BF-NEQ unless the O-RU leads to a change in time / frequency / power properties of the received signal. When enabled, these measurements are sent every slot except (v) and (vi). The measurements (v) and (vi) are c-plane triggered and do not need to be sent to every slot.

[0060] The RRM measurement information can be exchanged between PHYand MAC through a PHY API. This information, which is conveyed through the PHY API, can be a capability or measurement data message, which is typically sent from O-DU PHY to O-DU MAC, or can be a configuration message, which is typically sent from O-DU MAC to O-DU PHY (i.e., in the opposite direction). Please note that O-DU MAC can receive capability messages from O-RU through the O-RAN FH interface as well.

[0061] As an example of a PHY API message, the RRM measurements can beconfigured (such as enabled / disabled or set time / frequency granularities) in O-DU MAC (L2), and then these measurements can be conveyed from O-DU PHY (L1) to O- DU MAC (L2) through a PHY API. The RRM measurements can be accompanied by a symbol mask to indicate their associated relevance in the time-frequency resources. Similar to the other DMRS-BF features, capabilities for RRM measurements, such as supported RRM measurements and their associated features (e.g., time / frequency granularities and averaging properties), can be exchanged through the PHY API.

[0062] Compared to the current RRM measurements listed in the CUS planespecification, ongoing RRM measurements can include additional information to improve performance further. For example, messages can include spatial information in IpN measurements (i.e., (iv) and (v)), or to exploit long-term statistics for better Timing Offset (i.e., (i)) and Frequency Offset (i.e., (iii)) measurements. Also, the RRM measurements are exemplified in this section using the DMRS symbols. However, RRM measurements that are derived from SRS can be utilized and sent through a PHY API as well. An example includes the “usercorrelation” information, which is used for user pairing. This RRM measurement can be extracted from SRS and sent from PHY (L1) to MAC (L2) for pairing decisions.

[0063] (e)Post-processing SINR and Its Format:

[0064] O-RUs with advanced PHY capabilities can perform the equalizationoperation. After the equalization, the equalized data streams are transferred to O- DU along with the post-processing SINR. The post-processing SINR is needed for LLR calculation.

[0065] The time and frequency resolution of SINR has an impact on thethroughput performance. Hence, the time and frequency resolution of SINR can be adjusted to optimize the system performance considering throughput performance, computational complexity, and FH bit rate tradeoffs. The time and frequency resolution of SINR can be static (e.g., when configured through the M-plane) or dynamic (e.g., when configured through C-plane). Also, the SINR compression method and block size can be adjusted as a part of the SINR format (along with the time / frequency granularities) to optimize the system performance.

[0066] The post processing SINR and its format information can beexchanged between PHY and MAC through a PHY API. This information, which is conveyed through the PHY API, can be a capability or data message, which is typically sent from O-DU PHY to O-DU MAC, or can be a configuration message, which is typically sent from O-DU MAC to O-DU PHY (i.e., in the opposite direction). Please note that O-DU MAC can receive capability messages from O-RU through the O-RAN FH interface as well. As an example PHY API message, the SINR can be configured (such as enabled / disabled or set time / frequency granularities) in O-DU MAC (L2), and then it can be conveyed from O-DU PHY (L1) to O-MAC (L2) (if needed) through a PHY API.

[0067] (f) Dimension Reduction:

[0068] O-RUs with advanced PHY capabilities can optimize the RANperformance considering computational complexity tradeoffs. A way of reducing the computational complexity is reducing the dimension for DMRS processing at O-RU. The dimension reduction can be enabled on the O-RU side agnostic to O-DU. Alternatively, O-DU can instruct O-RU for DMRS dimension reduction.

[0069] In the case of O-DU-controlled dimension reduction, the dimensionreduction information can be exchanged between PHY and MAC through a PHY API. This information, which is conveyed through the PHY API, can be a capability message, which is typically sent from O-DU PHY to O-DU MAC, or can be a configuration message (such as beam indicators), which is typically sent from O-DU MAC to O-DU PHY (i.e., in the opposite direction). Please note that O-DU MAC can receive capability messages from O-RU through the O-RAN FH interface as well.

[0070] (g) Other Capabilities and Configurations Related to DMRS andScheduling

[0071] O-RUs with advanced PHY capabilities can perform DMRS-BF, but O-DU (PHY and MAC) and O-RU (PHY) capabilities and limitations can be different. Hence, these components can act according to each others’ capabilities and limitations for algorithm and parameter selection. For example, the O-DU MAC scheduler can be configured to know the supported DMRS configurations by O-DU PHY (along with the O-RU capabilities).

[0072] The capabilities and configurations related to DMRS and schedulinginformation can be exchanged between PHY and MAC through a PHY API. This information, which is conveyed through the PHY API, can be a capability message, which is typically sent from O-DU PHY to O-DU MAC, or can be a configuration message, which is typically sent from O-DU MAC to O-DU PHY (i.e., in the opposite direction). Please note that O-DU MAC can receive capability messages from O-RU through the O-RAN FH interface as well. Sample capabilities and configuration messages related to DMRS and scheduling are listed below:Limits and capabilities on the number of UEs, UE layers, user groups. Maximum number of ueIds, maximum number of ueIds per slot, and maximum number of user groups per slot. Maximum user groups per UE in time. Maximum number of DMRS configurations per user group. Lists of supported “DMRS Symbol Mask,” “Hopping mode,” and “Low PAPR Type” values. Support for the use of “ueId persistence” across multiple slots. Support for UEs having partial overlap in frequency and time while establishing user groups.

[0073] This list can be extended further with the SRS capabilities andconfigurations as well. The PHY and MAC can exchange information regarding SRS information through PHY API similar to DMRS information.

[0074] This disclosure provides 5G PHY API procedures and messages toenable and optimize DMRS-BF capable O-RAN based Radio Access Networks, according to the recent and ongoing developments in DMRS-BF related advances in the O-RAN architecture.

[0075] This disclosure describes implementation to exchange DMRS-BFrelated information, including capabilities, parameters, configurations, and data, between O-DU PHY and O-DU MAC. The sample messages can be extended for other DMRS-BF related features. Moreover, the same concept can be extended with the O- RAN features, such as SRS processing at O-RU. Similar messages can be exchanged for 6G networks as well.

[0076] ACRONYMSo 3GPP: Third generation partnership projecto BS: Base Stationo CA: Carrier Aggregationo CAPEX: Capital Expenditureo CBRS: Citizens Broadband Radio Serviceso CC: Component carriero COTS: Commercial off-the-shelfo CP: cyclic prefixo C-plane: Control planeo C-RAN: cloud radio access networko CU: Central unito DCI: downlink control indicatoro DL: downlinko DMRS DeModulation Reference Signalo DU: Distribution / Distributed unito eAxC ID: Extended Antenna-Carrier identifier: a data flow for a singleantenna (or spatial stream) for a single carrier in a single sector.o eNB: Evolved Node B (applies to LTE)o FDD: Frequency-division duplexo FEC: forward error correctiono FH: Fronthaulo FFT: Fast Fourier Transformo gNB: g NodeB (applies to NR)o iFFT: inverse Fast Fourier Transformo HARQ: hybrid automatic repeat requesto LTE: long term evolutiono LTE-A: LTE Advancedo M-plane: Management planeo MCS: modulation and coding schemeo MIMO: multiple input, multiple outputo MMSE-IRC: Minimum mean square error - interference rejectioncombiningo MMSE-MRC: Minimum mean square error - maximum-ratiocombiningo mmWave: millimeter waveo MNO: Mobile network operatoro NR: New radioo OAM: Operation and managemento O-DU: O-RAN Distributed Unito O-RU: O-RAN Radio Unito O-RAN: Open RAN (Basic O-RAN specifications are prepared by the O-RAN alliance)o OPEX: Operating Expenseo PBCH: Physical Broadcast Channelo PCFICH: Physical Control Format Indicator Channelo PDCCH: Physical downlink Control Channelo PDCP: Packet Data Convergence Protocolo PDSCH: physical downlink shared channelo PDU: Protocol data unito PHICH: Physical Hybrid ARQ Indicator Channelo PHY: physical layero LPHY: lower physical layero UPHY: upper physical layero PUCCH: Physical Uplink Control Channelo PUSCH: Physical Uplink Shared Channelo QAM: quadrature amplitude modulationo QPSK: Quadrature Phase Shift Keyingo RACH: random access channelo PRACH: physical random access channelo RF: radio frequency interfaceo RLC: Radio Link Controlo RRC: Radio Resource Controlo RRM: Radio resource managemento RRU: Remote radio unito RU: Radio Unito RS: reference signalo RSSI: received signal strength indicatoro RPC: Remote procedure callo SMO: Service Management and Orchestrationo S-plane: Synchronization planeo SCell: Secondary cello SIMO: single input, multiple outputo SINR: signal-to-interference-plus-noise ratioo SRS: Sounding reference signalo SSS: Secondary Synchronization Signalo TB: transport blocko TTI: Transmission Time Intervalo TDD: Time division duplexo U-plane: User planeo UCI: Uplink Control Informationo UE: user equipmento UL: uplinko UL DMRS: uplink demodulation reference signalo ULSCH: Uplink Shared Channelo vBBU: Virtualized baseband unito VNF: Virtual Network Function

[0077] DEFINITIONSo Channel: the contiguous frequency range between lower and upperfrequency limits. oC-plane: Control Plane: refers specifically to real-time control betweenO-DU and O-RU, and should not be confused with the UE’s control plane. oDL: DownLink: data flow towards the radiating antenna (generally onthe LLS interface). oLLS: Lower Layer Split: logical interface between O-DU and O-RUwhen using a lower layer (intra-PHY based) functional split.o M-Plane: Management Plane: refers to non-real-time managementoperations between the O-DU and the O-RU.o O-CU: O-RAN Control Unit – a logical node hosting PDCP, RRC, SDAPand other control functions.o O-DU: O-RAN Distributed Unit: a logical node hosting RLC / MAC / High-PHY layers based on a lower layer functional split.o O-RU: O-RAN Radio Unit: a logical node hosting Low-PHY layer and RFprocessing based on a lower layer functional split. This is similar to 3GPP’s “TRP” or “RRH” but more specific in including the Low-PHY layer (FFT / iFFT, PRACH extraction).o OTA: Over the Airo S-Plane: Synchronization Plane: refers to traffic between the O-RU orO-DU to a synchronization controller which is generally an IEEE 1588 Grand Master (however, Grand Master functionality can be embedded in the O-DU).o U-Plane: User Plane: refers to IQ sample data transferred between O-DU and O-RU.o UL: UpLink: data flow away from the radiating antenna (generally onthe LLS.

Claims

CLAIMS 1. A method for a Radio Access Network system comprising: configuring distributed unit DU to provide DeModulation Reference Signal (DMRS)-based beamforming (DMRS-BF) information to a radio unit (RU) on a slot-by-slot basis for each user group; extracting, at the RU, DMRS data from and Physical Uplink Shared Channel (PUSCH) data from the DMRS-BF information, the extracted DMRS data comprising channel estimation and beamforming weights calculation, equalization, and demodulation; computing beamforming weights from the extracted DMRS data; and applying the beamforming weights to the PUSCH data.

2. The method of claim 1, further comprising: receiving, at the RU, DMRS and sound reference signal (SRS) messages from User Equipment (UE); processing the DMRS data at a physical (PHY) layer L1 at the RU or the DU; transferring the processed DMRs data to a medium access control (MAC) L2 or higher layer via an PHY Application Program Interface (API); deciding, at the MAC L2 or higher layer, a selected uplink (UL) beamforming method; conveying the selected UL beamforming method to the DU via the PHY API; and conveying the selected UL beamforming method to the RU so that the RU can apply the selected UL beamforming method for the UE.

3. The method of claim 1, further comprising: configuring the RU to support advanced PHY configurations selected from the group of: a delay profile between the RU and the DU, wherein the delay profile comprises parameters including subcarrier spacing, carrier bandwidth, and beamforming method;symbol reordering for DMRS symbols; RMM measurements; post processing of SINR, wherein time and frequency of SINR are adjusted to optimize the system performance based on throughput performance, computational complexity, and FH bit rate tradeoff; dimension reduction for reducing a dimension for processing the DMRS at the RU; and any combination thereof.

4. The method of claim 3, wherein the advanced PHY configurations include further capabilities and configurations for DMRS and scheduling selected from the group of: limits and capabilities on a number of the UEs, UE layers, user groups; a maximum number of ueIds, maximum number of ueIds per slot; maximum number of user groups per slot; a maximum number user groups per UE in time; a maximum number of DMRS configurations per user group; lists of supported “DMRS Symbol Mask,” “Hopping mode,” and “Low PAPR Type” values; and support for the use of “ueId persistence” across multiple slots; and any combination thereof.

5. The method of claim 4, further comprising: configuring the PHY API to convey a message including the advanced PHY parameter between the PHY Layer L1 to the MAC L2 or higher layer.

6. The method of claim 3, further comprising: conveying a capability message from the DU PHY to the DU MAC; or conveying a configuration message from the-DU MAC to the DU PHY.

7. The method of claim 3, wherein the RMM measurement is selected from the group of: UE Timing Advance Error (TAE), UE layer signal power, UE frequency offset, Interference plus noise (IpN) for allocated PRBs, IpN for unallocated PRBs, Antenna DMRS SNR, or any combination thereof.

8. The method of claim 7, wherein the RRM measurements are accompanied by a symbol mask to indicate associated relevance in time-frequency resources.

9. The method of claim 3, wherein the time and frequency resolution of SINR are static when configured through an M-plane or dynamic when configured through C-a plane.

10. The method of claim 3, further comprising a compression method and block size that are adjusted with the time and frequency resolution of SINR.

11. A Radio Access Network system comprising: a distributed unit DU comprising a physical (PHY) layer L1, a medium access control (MAC) layer L2, and a PHY Application Program Interface (API) configured to transfer DeModulation Reference Signal (DMRS) data between the PHY layer L1 to the MAC L2 layer; and a radio unit RU comprising the PHY layer L1; wherein the DU is configured to provide DMRS-based beamforming (DMRS-BF) information to the RU.

12. The system of claim 11, further comprising: the DU being configured to provide the DMRS-BF information to the RU on a slot-by-slot basis for each user group; the RU being configured to extract DMRS data from and PUSCH data from the DMRS-BF information, the extracted DMRS data comprising channel estimation and beamforming weights calculation, equalization, and demodulation;compute beamforming weights from the extracted DMRS data; and apply the beamforming weights to the PUSCH data.

13. The system of claim 1, wherein: the RU is configured to receive DMRS and SRS messages from User Equipment (UE); the RU and / or DU is configured to process the DMRS data at a PHY layer L1 and transfer the processed DMRS data to the MAC L2 or higher layer via a PHY API; the MAC L2 or higher layer is configured to a select uplink (UL) beamforming method and convey the selected UL beamforming method to the DU via the PHY API; and the DU is configured to convey the selected UL beamforming method to the RU so that the RU can apply the selected UL beamforming method for the UE.

14. The system of claim 11, wherein: the RU is configured to support advanced PHY configurations selected from the group of: a delay profile between the RU and the DU, wherein the delay profile comprises parameters including subcarrier spacing, carrier bandwidth, and beamforming method; symbol reordering for DMRS symbols; RMM measurement; post processing of SINR, wherein time and frequency of SINR are adjusted to optimize the system performance based on throughput performance, computational complexity, and FH bit rate tradeoff; or dimension reduction for reducing a dimension for processing the DMRS at the RU; and any combination of thereof.

15. The system of claim 14, wherein advanced PHY configurations include further capabilities and configurations for DMRS and scheduling selected from thegroup of: limits and capabilities on a number of the UEs, UE layers, user groups; a maximum number of ueIds, maximum number of ueIds per slot, and maximum number of user groups per slot; a maximum number user groups per UE in time; a maximum number of DMRS configurations per user group; lists of supported “DMRS Symbol Mask,” “Hopping mode,” and “Low PAPR Type” values; support for the use of “ueId persistence” across multiple slots; and any combination of thereof.

16. The system of claim 14, wherein: the PHY API is configured to convey a message including the advanced PHY parameter between the PHY Layer L1 to the L2 / L3 layer.

17. The system of claim 14, wherein the RMM measurement is selected from the group of: UE Timing Advance Error (TAE), UE layer signal power, UE frequency offset, Interference plus noise (IpN) for allocated PRBs, IpN for unallocated PRBs, Antenna DMRS SNR, and any combination thereof.

18. The system of claim 14, wherein the RRM measurements are accompanied by a symbol mask to indicate associated relevance in time-frequency resources.

19. The system of claim 14, wherein the time and frequency resolution of SINR are static when configured through an M-plane or dynamic when configured through C-a plane.

20. The system of claim 14, further comprising a compression method and block size that are adjusted with the time and frequency resolution of SINR.

Images