First node, and method for frequency synchronization

Abstract

Embodiments herein relate to a method performed by a first node (213) to handle frequency synchronization between two or more second nodes (211, 212) The method comprises receiving, from at least two second nodes (211, 212), reports on updates of a frequency indicating a value for adjustment of a reference clock frequency, and receiving, from at least one of the two second nodes (212), a phase Δφ or frequency Δƒ measurement. The method further comprises estimating a frequency offset between the at least two second nodes (211, 212) related to frequency synchronization taking the reports and the phase Δφ or frequency Δƒ measurement during an observation interval, Δtobs into account.

Description

[0001] FIRST NODE, AND METHOD FOR FREQUENCY SYNCHRONIZATION TECHNICAL FIELD

[0002] Embodiments herein relate to a first node, and a method performed therein regarding wireless communication. Furthermore, a computer program product and a computer-readable storage medium are also provided herein. In some aspects, they relate to handling frequency synchronization between two or more second nodes.

[0003] BACKGROUND

[0004] In a typical wireless communication network, wireless devices, also known as wireless communication devices, mobile stations, stations (STA) and / or User Equipment (UE), communicate via a Wide Area Network or a Local Area Network such as a Wi-Fi network or a cellular network comprising a Radio Access Network (RAN) part and a Core Network (CN) part. The RAN covers a geographical area which is divided into service areas or cell areas, which may also be referred to as a beam or a beam group, with each service area or cell area being served by a radio network node such as a radio access node e.g., a Wi-Fi access point, a Base Station (BS) or a radio base station (RBS), which in some networks may also be denoted, for example, a Base Station (BS), a NodeB, eNodeB (eNB), or gNodeB (gNB) as denoted in Fifth Generation (5G) telecommunications. A service area or cell area is a geographical area where radio coverage is provided by the radio network node. The radio network node communicates over an air interface operating on a radio frequency with the wireless devices within the range of the radio network node.

[0005] 3rd Generation Partnership Project (3GPP) is the standardization body for specifying the standards for the cellular system evolution, e.g., including 3G, 4G, 5G and the future evolutions. Specifications for Evolved Universal Terrestrial Radio Access (E-UTRA) and Evolved Packet System (EPS) have been completed within the 3GPP. In 4G also called a Fourth Generation (4G) network, EPS is core network and E-UTRA is radio access network. In 5G, 5G Core (5GC) is core network, NR is radio access network. As a continued network evolution, the new release of 3GPP specifies a 5G network also referred to as 5G New Radio (NR) and 5GC.

[0006] Frequency bands for 5G NR are being separated into two different frequency ranges, Frequency Range 1 (FR1) and Frequency Range 2 (FR2). FR1 comprises sub-6 GHz frequency bands. Some of these bands are bands traditionally used by legacy standards but have been extended to cover potential new spectrum offerings from 410 MHz to 7125 MHz. FR2 comprises frequency bands from 24.25 GHz to 52.6 GHz. Bands in this millimeter wave range have shorter range but higher available bandwidth than bands in the FR1.

[0007] Multi-antenna techniques may significantly increase the data rates and reliability of a wireless communication system. For a wireless connection between a single user, such as UE, and a base station (BS), the performance is in particular improved if both the transmitter and the receiver are equipped with multiple antennas, which results in a Multiple-Input Multiple-Output (MIMO) communication channel. This may be referred to as Single-User (SU)-MIMO. In the scenario where MIMO techniques is used for the wireless connection between multiple users and the base station, MIMO enables the users to communicate with the base station simultaneously using the same time-frequency resources by spatially separating the users, which increases further the cell capacity. This may be referred to as Multi-User (MU)-MIMO. Note that MU-MIMO may benefit when each UE only has one antenna. The cell capacity can be increased linearly with respect to the number of antennas at the BS side. Due to that, more and more antennas are employed in BS. Such systems and / or related techniques are commonly referred to as massive MIMO.

[0008] Integrating sensing capability to a communication system can be a very cost-efficient way to achieve new functionalities. Some Integrated Sensing and Communication (ISAC) features depend on very stable radio frequency (RF) phase between Transmission and Reception Points (TRPs), which is equivalent to very small frequency offset between TRPs. This is also true for positioning services such as Carrier Phase Positioning (CPP) and bi- / multi-static radar functionality. The gap is very large between current system performance designed for regular communication features and required performance for mentioned ISAC features.

[0009] Another feature requiring very small frequency offset between TRPs is Coherent Joint Transmission (C-JT).

[0010] One way to achieve better synchronization is to use over the air synchronization hereafter called Radio Interface Based Synchronization (RIBS). RIBS today is generally developed to handle Time Alignment Errors between TRPs but can be extended to include characterization of frequency offset between TRP nodes. This updated RIBS function, here called RIB-Radio Phase Stability (RIB-RPS) is needed in order to match the requirements from the mentioned ISAC functions and positioning features. RIB-RPS uses signals sent from one TRP to one or more receiving TRPs. In the receiving TRPs, the frequency offset between transmitting and receiving TRP is estimated. This estimated frequency offset can then be used for monitoring purposes (RIB-RPS-M) or for actual frequency control purposes (RIB-RPS-S). RIB-RPS-M means that the frequency offset is estimated, and the result is used for observability purposes, and also for frequency correction measures in the ISAC features or positioning features. For example, the ISAC features can calculate a phase drift vs time based on the frequency offset information, and correct the calculated phase difference with some granularity in time. It means that the actual RF frequency for transmitter, TX, and receiver, RX, is not actually corrected, only the effects of it is compensated.

[0011] RIB-RPS-S means that the frequency offset is estimated, and the result is used for actually controlling the RF frequency for TX and RX, either at one TRP node, or at both nodes.

[0012] Todays and future Radio Units (RUs) that are to be used as TRPs in multi-TRP systems are often enhanced Common Public Radio Interface (eCPRI) based, meaning a packet-based fronthaul is used for signaling to / from distributed unit (DU) to / from radio unit (RU). The IEEE standardized Precision Time Protocol (PTP) is used to distribute timing packets over eCPRI.

[0013] The frequency synchronization accuracy between TRPs depends on factors, such as synchronization source(s), characteristics of fronthaul synchronization distribution (like PTP), TRPs reference oscillator designs, temperature variations, etc. Any frequency offset between TRPs will cause phase drift over time. Specifically, a synchronization control algorithm generates a reference clock for the TRPSs, but relative phase and frequency offset between the TRPs may vary because of periodical updates of the reference clock in each TRP and / or accuracy of the reference clock.

[0014] One way for a synchronization algorithm to control the time of the TRP is to adjust the reference clock frequency keeping track of time, here with a Frequency Control Word (FCW), this to either advance or delay the time. Such frequency control, even if it leads to smooth changes in time, leads to abrupt changes in the reference clock frequency and thereby also to the RF frequency for each TRP. Since the FCW updates are independent per TRP it will cause changes of the frequency offset between TRPs over time. Typical behaviour is shown in Figure 1 illustrating phase and frequency offset vs time between two TRPs with eCPRI. Every abrupt change in frequency is due to FCW update at one or both of the TRPs.

[0015] The RIB-RPS observation scheme requires observation of RF phase change and / or time change between multiple received symbols to estimate a frequency offset. Using the observed RF phase change and the knowledge of time between the received symbols the frequency offset can be decided from the relationship between change in phase, frequency and time: A<p = 2TT A / tobs(1)

[0016] A<p is phase change in radians, Af is frequency offset in Hz, tobsis observation time (time between start and end of the observation) in seconds. It can be understood that for small Af, there needs to be a considerable time tobsin order to confidently detect a phase change A<p, because the observation will suffer from inaccuracies due to phase noise, receiver noise and interference. Therefore, a relatively small frequency offset requires relatively long time to evolve into a phase offset that can be reliably detected.

[0017] SUMMARY

[0018] As part of developing embodiments herein, one or more problems have been identified that will be described.

[0019] As an example, if it is wanted to detect and resolve an inter TRP offset to better than 0,1 ppb, e.g. corresponding to a fraction of the doppler caused by a slow moving object in a bi-static ISAC scenario, this corresponds to 0,2Hz for a 2GHz RF carrier. Within a slot of 0,5ms the drift will be as small as 6e-4 radians (0.036 degrees), which is too small to be accurately detected and measured.

[0020] If the measurement interval is increased to 500ms this will lead to a phase offset of 36 degrees and hence more reliability to detect and measure small frequency offsets between TRPs.

[0021] However, when extending the observation window between measurements, the measurements will also suffer from regular frequency updates as shown in Figure 1 and it will be difficult to blindly characterize a frequency offset between the TRPs.

[0022] It should also be noted that the phase ambiguity due to phase wrap around would have to be handled as part of the solution when increasing observation time.

[0023] Based on the challenges in detecting a very accurate frequency offset with short observation time intervals, and the abruptly varying frequency offset due to irregular FCW updates, it can be understood that it is a problem to maintain a small frequency offset between TRPs with a blind RIB-RPS function. Blind in the sense that no knowledge of the FCWs are utilized in the frequency offset detection algorithm.

[0024] An object of embodiments herein is to improve frequency synchronization between two or more TRPs.

[0025] According to an aspect of embodiments herein, the object is achieved by a method performed by a first node to handle frequency synchronization between two or more second nodes. It is disclosed that the method comprises receiving, from at least two second nodes, reports on updates of a frequency indicating a value for adjustment of a reference clock frequency.

[0026] The receiving also include receiving from at least one of the two second nodes, a phase Δφ̂ or frequency Δf̂ measurement.

[0027] It is further disclosed that the method comprises the performance of estimating a frequency offset between the at least two second nodes related to frequency synchronization taking the reports and the phase Δφ̂ or frequency Δf̂ measurement during an observation interval, Δtobs, into account.

[0028] It is disclosed that the receiving, from at least two second nodes, reports on updates of a frequency indicating a value for adjustment of a reference clock frequency, further comprises receiving respective timestamps, indicating a time of adjustment of the reference clock frequency, at the respective second node.

[0029] Embodiments herein teach that estimating a frequency offset between the at least two second nodes related to frequency synchronization taking the reports and the phase A<p or frequency A measurement during an observation interval, Atobs, into account may further comprise estimating a frequency offset based on a known time of adjustment of the reference clock frequency.

[0030] It is also disclosed that estimating a frequency offset between the at least two second nodes may comprise:

[0031] determining the estimated frequency offset as un-valid if an update of the frequency was made during the observation interval, Δtobs

[0032] determining the estimated frequency offset as valid if an update of the frequency was not made during the observation interval, Δtobs.

[0033] Embodiments herein teach that estimating a frequency offset between the at least two second nodes may comprise compensating a measurement for an accumulated frequency update effect during the observation interval, Δtobs, where the observation interval, Δtobsmay stretch over one or several frequency updates.

[0034] Disclosed embodiments show that the method may comprise determining an instantaneous frequency offset at a point in time between the at least two second nodes based on the reports on updates of a frequency and phase Δφ̂ or frequency Δf̂ measurement.

[0035] Disclosed embodiments teach that the method may comprise estimating a maximum frequency offset Δfmax estbased on at least one of: information from a synchronisation system of the at least two second nodes, on estimated clock frequency error, historical frequency updates and their statistics; information about the synchronization system of the at least two second nodes such as synchronization source(s), characteristics of fronthaul synchronization distribution, reference oscillator designs of the at least two second nodes, temperature variations etc; and

[0036] an existing standardized maximum value of the maximum frequency offset.

[0037] It is also disclosed that the observation interval, Δtobs, may be determined based on the maximum frequency offset Δfmax est.

[0038] Embodiments herein teach that the method may comprise determining the interval, Δtobs, by starting from a fixed observation interval, Δtobs, value and increasing the observation interval, Δtobs, value.

[0039] Embodiments herein teach that the method may comprise the updating of the estimated frequency offset using received reports on updates of a frequency in-between determined frequency offsets.

[0040] It is disclosed that the method may comprise estimating or detecting hardware (HW) related changes in frequency, and determining a prediction model for compensation of the changes in frequency offset Δferrordue to HW related effects based on a time interval between at least two subsequent measurements of the instantaneous frequency offset and estimated frequency offsets.

[0041] It is also disclosed that the method may comprise using the prediction model to determine the observation interval, Δtobs.

[0042] Disclosed embodiments teach that the method may comprise using the prediction model as basis to derive a frequency offset estimate quality index.

[0043] Aspects of embodiments herein discloses that the method may comprise transmitting the compensation obtained from the prediction model to at least one of the two second nodes.

[0044] It should be understood that the first node may be acting as a RIB-RPS entity between a number of second nodes, where the second nodes may be acting as Transmission and Reception points, TRPs, in a multi-TRP system.

[0045] According to another aspect of embodiments herein, the object is achieved by a first node adapted to handle frequency synchronization between two or more second nodes.

[0046] It is disclosed that the first node is adapted to receive, from at least two second nodes, reports on updates of a frequency indicating a value for adjustment of a reference clock frequency. The first node is also adapted to receive, from at least one of the two second nodes, a phase Δφ̂ or frequency Δf̂ measurement, and to estimate a frequency offset between the at least two second nodes related to frequency synchronization taking the reports and the phase Δφ̂ or frequency Δf̂ measurement during an observation interval, Δtobs, into account.

[0047] It is disclosed that the first node may be adapted to, in the reception, from at least two second nodes, of reports on updates of a frequency indicating a value for adjustment of a reference clock frequency, receive respective timestamps, indicating a time of adjustment of the reference clock frequency, at the respective second node.

[0048] The first node may be further adapted to, in the estimation of a frequency offset between the at least two second nodes related to frequency synchronization taking the reports and the phase A<p or frequency A measurement during an observation interval, Atobs, into account, estimate a frequency offset based on a known time of adjustment of the reference clock frequency.

[0049] The first node may be further adapted to, in the estimation of a frequency offset between the at least two second nodes,

[0050] determine the estimated frequency offset as un-valid if an update of the frequency was made during the observation interval, Atobs,

[0051] determine the estimated frequency offset as valid if an update of the frequency was not made during the observation interval, Δtobs.

[0052] Embodiments discloses that the first node may be adapted to, in the estimation of a frequency offset between the at least two second nodes, compensate a measurement for an accumulated frequency update effect during the observation interval, Δtobs, where the observation interval, Δtobsmay stretch over one or several frequency updates.

[0053] The first node may be adapted to determine an instantaneous frequency offset at a point in time between the at least two second nodes based on the reports on updates of a frequency and phase A<p or frequency A measurement.

[0054] It is also disclosed that the first node may be adapted to estimate a maximum frequency offset Δfmax estbased on at least one of:

[0055] information from a synchronisation system of the at least two second nodes, on estimated clock frequency error, historical frequency updates and their statistics; information about the synchronization system of the at least two second nodes such as synchronization source(s), characteristics of fronthaul synchronization distribution, reference oscillator designs of the at least two second nodes, temperature variations etc; and an existing standardized maximum value of the maximum frequency update. It is disclosed that the first node may be adapted to determine the observation interval, Δtobs, based on the maximum frequency offset Δfmax est.

[0056] It is also disclosed that the first node may be adapted to determine the observation interval, Δtobs, by starting from a fixed observation interval, Δtobs, value and increase the observation interval, Δtobs, value.

[0057] Aspects of embodiments teach that the first node may be adapted to update the estimated frequency offset using received reports on updates of a frequency in-between determined frequency offsets.

[0058] The first node may also be adapted to estimate or detect hardware, HW, related changes in frequency, and determine a prediction model for compensation of the changes in frequency offset Δferrordue to HW related effects based on a time interval between at least two subsequent measurements of the instantaneous frequency offset and estimated frequency offsets.

[0059] It is disclosed that the first node may be adapted to use the prediction model to determine the observation interval, Δtobs.

[0060] The prediction model may also be used by the first node as basis to derive a frequency offset estimate quality index.

[0061] It is disclosed that the first node may also be adapted to transmit the compensation obtained from the prediction model to at least one of the at least two second nodes.

[0062] Embodiments herein teach that the first node may be adapted to act as a Radio Interface Based Radio Phase Stability, RIB-RPS, entity between a number of second nodes, where the second nodes may be adapted to act as Transmission and Reception points, TRPs, in a multi-TRP system.

[0063] According to another aspect of embodiments herein, the object is achieved by a computer program comprising instructions, which when executed by a processor, causes the processor to perform steps according to the method above.

[0064] A carrier comprising the computer program of claim 35, wherein the carrier is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer-readable storage medium.

[0065] Embodiments herein may provide one or more of the following advantages:

[0066] The first node, such as a node acting as a RIB-RPS entity, is enabled to maintain a very accurate estimate of frequency offset over time with a very low overhead in over the air (OTA) signalling. It uses a Af progression model that considers frequency updates, such as FCW updates, with time stamps to update and keep track of inter TRP frequency offsets. By utilizing a Af progression model, a RIB-RPS entity can use a longer observation time for frequency offset measurements between TRPs and thereby achieve better accuracy needed to determine small frequency offsets. This by not being limited and restricted by phase ambiguity introduced by actual frequency offsets, FCW updates and applied frequency corrections for actual frequency control purposes, i.e. RIB-RPS-S. The disclosed embodiments enable the decoupling of the RIB-RPS measurements and the effects of FCW updates, which eliminates the need to coordinate RIB-RPS measurement events and FCW updates.

[0067] Disclosed embodiments eliminate the need to align and coordinate FCW updates to occur at the same time at multiple TRPs.

[0068] It is also possible to use an optional predictor based on regular received phase p measurement, alternatively frequency offset A measurement, from the second node to provide better frequency offset estimates i.e. including drift characteristics, not included in FCW, for optimal tuning of measurement periodicity and as a quality indicator of measurement accuracy.

[0069] When no or limited knowledge of actual frequency offset is available, only initial measurements need to be performed with sufficient short periodicity to avoid phase ambiguity thereafter measurements periodicity for ambiguity can be extended. This since periodicity is determined by estimated measurement frequency offset error uncertainty, allowing reducing measurement overhead once the estimation has converged.

[0070] With the invention disclosed herein, the effects of PTP fronthaul sync can be reduced and allows for phase coherent operation between TRPs even in very challenging fronthaul configurations, without the need for creating special fronthaul configurations. These special configurations may need switches or other network equipment with better characteristics than what is common, e.g. Class C, or having functionality to control or compensate synchronization accuracies, implying higher cost. Thus, the disclosed RIB-RPS solution allows for a very cost competitive system still able to maintain accurate frequency synchronization or to perform compensation from determined frequency offset.

[0071] A solution according to disclosed embodiments provides a decreased requirement of stable and power-hungry oscillators in radios, which will achieve an improvement in energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Examples of embodiments herein are described in more detail with reference to attached drawings in which:

[0073] Figure 1 is a graph illustrating amplitude, phase and frequency offset vs time between two TRPs.

[0074] Figure 2 is a schematic block diagram illustrating embodiments of a communications network.

[0075] Figure 3 is an overview of a RIB-RPS function.

[0076] Figure 4 is a flowchart illustrating an example of a RIB-RPS entity.

[0077] Figure 5 is a graph illustrating the accumulation of FCW.

[0078] Figures 6a and 6b are graphs illustrating the effect of different starting times.

[0079] Figures 7a and 7b is a sequence diagram and a graph illustrating first frequency offset at initialization.

[0080] Figure 8 is a schematic block diagram illustrating embodiments of FCW reporting to a RIB-RPS entity.

[0081] Figure 9 is a graph illustrating examples of regular RIB-RPS measurements.

[0082] Figure 10 schematically illustrates RIB-RPS-S to control and align the frequency of a sync receiver towards a sync provider.

[0083] Figure 11 is a graph illustrating predicted offset versus measured offset.

[0084] Figure 12 is a generalized block diagram illustrating one embodiment of UL CPP.

[0085] Figure 13 is a generalized block diagram illustrating one embodiment of DL CPP.

[0086] Figure 14 is a generalized block diagram illustrating one embodiment of DL CPP.

[0087] Figure 15 is a generalized block diagram illustrating one embodiment of DL CPP.

[0088] Figure 16 is a generalized block diagram illustrating one embodiment related to radar sensing with compensation in sensing feature.

[0089] Figure 17 is a generalized block diagram illustrating one embodiment related to radar sensing with TRP compensation.

[0090] Figure 18 is a generalized block diagram illustrating one embodiment related to radar sensing with synchronizing TRPs.

[0091] Figure 19 is a generalized block diagram illustrating one embodiment related to interaction with DL Coherent Joint Transmission (C-JT).

[0092] Figure 20 is a flowchart summary of an exemplifying embodiment of RIB-RPS.

[0093] Figure 21 is a generalized block diagram exemplifying embodiment of a first node.

[0094] Figure 22 is a generalized block diagram illustrating a first node. DETAILED DESCRIPTION

[0095] Figure 2 is a schematic overview depicting a wireless communications network 200 wherein embodiments herein may be implemented. The wireless communications network 200 comprises one or more RANs, and one or more CNs. The communications network 200 may use 5G NR but may further use a number of other different technologies, such as, 6G, Wi-Fi, Long Term Evolution (LTE), LTE-Advanced, Wideband Code Division Multiple Access (WCDMA), Global System for Mobile communications / enhanced Data rate for GSM Evolution (GSM / EDGE), Worldwide Interoperability for Microwave Access (WiMax), or Ultra Mobile Broadband (UMB), just to mention a few possible implementations.

[0096] A first node 213 may be acting to handle frequency synchronization between two or more second nodes 211, 212, where two such second nodes are exemplified as base stations in the figure, such as a first base station 211 and a second base station 212, operating in the RAN the communications network 200. The base stations 211, 212, may each be a transmission and reception point (TRP) e.g. a radio access network node such as a base station, e.g. a radio base station such as a NodeB, an evolved Node B (eNB, eNode B), an NR Node B (gNB), a base transceiver station, a radio remote unit, an Access Point Base Station, a base station router, a transmission arrangement of a radio base station, a stand-alone access point, a Wireless Local Area Network (WLAN) access point or an Access Point Station (AP STA), an access controller, or any other network unit capable of communicating with UEs, such as a UE 221, within a cell, served by the respective base station 211, 212. The respective base station 211, 212 may be referred to as a serving radio network node and may communicate with the UE 221 with Downlink (DL) transmissions to the UE 221 and Uplink (UL) transmissions from the UE 221.

[0097] It should be understood by the skilled in the art that “UE” is a non-limiting term which means any terminal, client, mobile client, wireless communication terminal, user equipment, Device to Device (D2D) terminal, or node e.g. smart phone, laptop, mobile phone, sensor, relay, mobile tablets or even a car or any small base station communicating within a cell.

[0098] As previously described, clock frequency synchronization between nodes is wanted and this can be done in different ways. Embodiments herein will exemplify this using one way to achieve better synchronization through over the air synchronization hereafter called Radio Interface Based Synchronization (RIBS).

[0099] A RIB-RPS entity continuously receives reports on FCW updates, with timestamp, from TRPs 211, 212. The following embodiments are disclosed.

[0100] The RIB-RPS entity may be the first node 213 and may compensate each new RPS measurements for accumulated FCW effect during the measurement / observation time. The RIB-RPS entity 213 may estimate instantaneous frequency offset between TRPs 211, 212 at a particular time instance based on RPS measurements, i.e. OTA measurements, and received FCWs.

[0101] Initial RIB-RPS measurements may be performed sufficiently often to guarantee no phase ambiguity due to phase wrap around effects caused by large initial unknown frequency offsets, later measurements avoid phase ambiguity by estimated instantaneous frequency offset and regular FCW information. Only remaining frequency offset estimate errors could result in phase ambiguity due to wrap around effects. An initial step is disclosed to set correct interval and avoid wrap around.

[0102] In-between repetitive RIB-RPS measurements and its instantaneous frequency offset at a point in time, FCW information may be used for updated frequency offset estimates. In this way, the frequency offset characterization is maintained overtime without very frequent RIB-RPS OTA signalling.

[0103] Repetitive RIB-RPS estimates may be used to detect hardware (HW) related changes in frequency due to oscillator, temp changes etc. i.e. frequency drift effects not caused by FCW, to improve performance and may be used in a prediction model for more advanced compensation of for such effects.

[0104] A predictor may also be used to adaptive determine RIB-RPS intervals both to avoid wrap around effects and to optimize it periodicity for a specific target accuracy, where a too large difference between predicted and measured instantaneous frequency offset at a point in time could trigger more frequent measurements.

[0105] A predictor may also be used as basis to derive a frequency offset estimate quality index reported to applications as a quality indicator to enable an application to assess against required accuracy levels.

[0106] The RIB-RPS entity may report one or more of instantaneous frequency offset with time stamp, FCW with time stamp, predicted drift and quality index to an end application, e.g. a positioning service or ISAC service.

[0107] Alternatively, the RIB-RPS entity may report the slow varying detected HW related changes in frequency for a slow loop compensation in the TRP frequency generation themselves.

[0108] Embodiments herein disclose a new way of taking a FCW into consideration in synchronization between nodes. It is disclosed that this can be done as follow.

[0109] By monitoring the instantaneous FCW updates the RIB RPS function may be enabled to maintain an accurate frequency offset estimate, momentarily, at each instant in time. A phase integrator may compensate phase drift due to instantaneous FCW, which allows for accurate frequency offset estimation using phase drift within an observation time window, by enabling a longer observation time window.

[0110] Using a measurement adaption function based on a model fit error estimate e to tune observation interval, improve frequency offset estimates and to determine a quality indication for frequency offset estimates.

[0111] Providing frequency offset information including quality estimates to features, i.e. ISAC and Positioning, that are dependent of accurate knowledge about frequency offset.

[0112] Embodiments herein such as the embodiments mentioned above will now be further described and exemplified. The text below is applicable to and may be combined with any suitable embodiment described above.

[0113] The objective of disclosed embodiments is to enable a first node to provide frequency synchronization with two or more second nodes 211, 212. In the following the first node 213 is exemplified by a RIB-RPS entity and the second nodes 211, 212 are exemplified by two TRPs, TRP1 and TRP2.

[0114] According to disclosed embodiments, the objective of the RIB-RPS function is to determine the instantaneous frequency offset between two TRPs to a very accurate level and maintain this knowledge over time. To achieve this, a model of the frequency offset is introduced and used as a model for determining an accurate frequency offset at a particular time, and to maintaining knowledge of frequency offset variation over time.

[0115] A key component of this model is that it utilizes knowledge of frequency updates, or updates of a frequency control word, FCW, at each TRP. With this it is possible to create and maintain an accurate frequency offset knowledge by OTA observations, i.e. through TRP-TRP measurements, at a very low repetition rate.

[0116] An overview of the function is shown in Figure 3. The first node, or RIB-RPS, initially determines frequency offset Af, by using phase observations within observation time Atobs, where Atobs= tb- tawill be used in the below descriptions.

[0117] RIB-RPS maintains knowledge about the Af progression by using knowledge of applied FCW at each TRP. This Af progression model is used to determine a Aip progression model. After an appropriate time period tperiod, a new observation of Af is initiated, which is used to correct the modeled Af progression. The correction is done for unknown changes in Af, for example temperature drift in each TRP affecting the reference oscillator, or other unknown factors as described more in detail below.

[0118] The observation time Atobsis a function of how well the offset is known. Known wrap around can be taken care of, unknown wrap around requires shorter time interval, as indicated in the figure where the first Atobslis shorter than the second Atobs2due to lack or limited lack of initial frequency offset, when wrap around is known from the modeled Af progression a longer time interval, as the second Atobs2in the figure, can be used.

[0119] A long time period, tperiod, between observation times is preferred to not create overhead traffic, which is a trade for precision since long periods between observations will give lower precision.

[0120] Figure 4 shows an exemplifying embodiment of a disclosed RIB-RPS entity, or first node, 213 in the system. FCW updates are reported 4a, 4b from the at least two second nodes 211 and 212 to the first node 213, or RIB-RPS entity. Regular phase Δφ̂ measurement is processed at at least one of the at least two second nodes 212 and fed 4c to the RIB-RPS entity 213. Alternatively regular frequency offset measurements Δf̂ is processed at at least one of the at least two second nodes 212 and fed 4d to the RIB-RPS entity 213, such frequency offset measurement reports could potentially include indications of whether the measurements include compensations for potential frequency updates occurring during measurements unless known by first node through e.g. a configuration parameter. This Af report should be accompanied by an indication of during which time the measurement has been performed, so that the RIB-RPS entity can know if there has been frequency updates during the measurement time. The RIB-RPS maintains Af progression over time and outputs an estimated Af at any instance in time. Additionally, RIB-RPS keeps a quality indication which is based on the Af measurement processing.

[0121] Modelling and background

[0122] It is disclosed that the resulting frequency offset between two TRPs 211, 212 may be modelled as a function of a HW related drift and from accumulated FCW updates. This models the frequency drift behavior relative to time t0as follows:

[0123] Af t) = Af(fo) +AfHW(t) + AFCWaccum( )

[0124] f(to) is the frequency offset at a particular time t0, it is assumed that t > t0, AfHWis a hardware, HW, related frequency drift, i.e. from oscillator drift, temperature drift etc, and AFCWaccumis the accumulated frequency drift from FCW updates from synchronization algorithms in the TRPs, e.g. to align TRP time to a PTP input reference by adjusting its internal frequency reference: & FCWaccum(t) = VfAFCW(t)

[0125]

[0126] Z_it=to

[0127] where AFCW is a positive or negative change in the frequency offset between the TRPs. Note that ΔFCWaccum(t0) = 0. One way of describing the accumulated frequency drift from FCW updates is FCWaccum(t) = AFCW(t) - AFCW(t0). An example of the FCWaccumbehavior is shown in Figure 5.

[0128] There is a relationship between frequency offset between transmitter and receiver, and the observed RF phase drift over an observation interval tobs. For constant frequency offset Af, the RF phase drift of a received signal can be found from the known relationship

[0129] A<p = 2nAf Atobs

[0130] In embodiments herein Af is not constant and the observed phase drift over the observation interval Atobs= tb- tais the integral frequency offset over the observation interval

[0131] A<p = (p(tb) - <p(ta) = J?62TT Af(t) dt (1)

[0132]

[0133] La

[0134] We set the reference time for Af to taso that

[0135] Af t) = Af(ta) + AfHW(t) + AFCWaccum(t) for t > ta(2)

[0136] For a system based on PTP synchronization, the variation in AFCWaccumis order of magnitudes larger compared to the HW related frequency drift for limited observation intervals. It means that by choosing an observation interval tb- tacorrectly it can be assumed that the frequency variation is dominated by AFCWaccum. Also, part of HW related frequency drift within a TRP will be compensated by regular frequency updates, FCW.

[0137] Following this, assuming HW related frequency drift is small and neglectable during the observation interval i.e. AfHW( ) =0 over the interval tb- ta, i.e. that the observation time Atobsis much shorter than relevant temperature time constants for oscillator drift, the phase drift over the observation interval is

[0138] A<p = J?L27T (Af(ta) + AFCWaccum(f)) dt = J? aL27T (Af(ta)) dt + C aL b2n AFCWaccum(f)dt = a 2n (Af(ta))(tb- ta) + 2n J?6AFCWaccum(t)dt (3a)

[0139]

[0140] La From this we can extract the frequency offset at time ta

[0141] - - J?6& FCWaccum(t)dt

[0142] A (3b

[0143]

[0144] f(ta) = )

[0145] (tb ^a)

[0146] where the second part in the numerator can be seen as a FCW compensation factor.

[0147] Figures 6a and 6b show example of A<p variation over time, the absolute phase variation due to frequency offset, according to equation (2) above, for different start time ta. The figures show <p vs time including instantaneous FCW update effects. Figure 6a shows the case where the phase drift at tais ~ 0, and figure 6b shows the case where phase drift at tais 0.

[0148] The objective of embodiments herein is to use the model in equation (3b) to get an estimated frequency offset for a measured phase drift observation A<p = < Pb ~ < Pa. Using A<p = A<p in equation (3b)

[0149] — 2]% - fb& FCWaccum(t)dt

[0150] A / (t«) = (4a)

[0151]

[0152] (tb ta)

[0153] Note that it is straightforward to replace the measured phase drift observation A<p over the observation interval (tb- ta) with a measured frequency offset observation A over the same interval in equation by using the known relationship A<p = 2nAf Atobs, where Atobs= (tb~ ta)- In this case, equation (4a) evaluates to:

[0154] AFCl4'accum(t)dt

[0155] Af(ta) = Af — ''O, (4b)

[0156]

[0157] (tb ta)

[0158] Note that expression (4a) above does not include effects of phase wrap-around of the phase drift observation A<p. Phase wrap-around is treated further below.

[0159] Af(ta) in the above expression (4a) is an estimate of the instantaneous frequency offset at a point in time at the start of the observation interval i.e. time ta. With this estimate it is possible to maintain a prediction of future instantaneous frequency offset by accumulating AFCW changes from ta.

[0160] A

[0161]

[0162] f(t) = Af(ta) + ^t=taAFCW(t) + AfHW(f) for t > ta(5) Phase wrap-around

[0163] The measured phase drift Ay can have phase wrap-around if Af is large and / or if the observation time tb- tais large. Large Af can be due to either large Af(ta) or large AFCW. The phase wrap-around will cause a difference in the measured phase drift Arp and modeled phase drift Ap, because the modeled phase contains no wrap-around effects:

[0164] Arp = Ay — N2n

[0165] Where N is an integer > 0, i.e. if N=0 no phase wraparound has occurred. If there is sufficient accurate measured phase drift and modeled phase drift from (3a), the integer N can be calculated from the relation

[0166] A, > Ay - Arp

[0167] 2n

[0168] Any error in estimation model Ay and the measured Arp will cause an error in the above expression meaning that N is not an integer. By rounding to the nearest integer the result is exact as long as the phase error e is less than TT:

[0169] N

[0170]

[0171] = round2^±e) = if \e\ < n (6)

[0172] The phase error e is a combination of estimation error in equation (3a), which can be due to incorrect knowledge of the initial frequency offset Af(ta) and hardware related drift ΔfHW(t) assumed as AfHWt) =0 in equation (3a), and due to measurement uncertainties of the phase drift Arp. However, the requirement for |e| < it is rather relaxed and is possible to fulfill even considering model and measurement errors.

[0173] By estimating the error parameter, it is possible to get an indication of the quality of the estimation. The estimated error component is found from the remainder of (6)

[0174] |e| = |7V2TT — (Ay — Arp)| (7)

[0175] The derived wraparound and N from equation (6) can then be used to compensate Arp in earlier equation (4a).

[0176] _ jtj,& FCW(t)dt — - fb& FCWaccum(t)dt + N Aftta) = Af(ta) =

[0177]

[0178] (tb ^a) (tb ^a) The above considerations are well known from other functions like for example doppler estimation algorithms. The established way of ensuring robust estimation is to ensure that the time interval tobsis small enough so that no wrap around occurs.

[0179] For the RIB-RPS function, it is of interest to use as long tobsas possible, since every RIB-RPS measurement uses resources from the DL traffic, replacing DL traffic symbols with RIB reference symbols. It is therefore an objective for RIB-RPS to be as efficient as possible in this aspect.

[0180] For finding a correct estimate of f(ta) it is possible to distinguish between two cases:

[0181] initial estimate, where there is no initial knowledge of Af,

[0182] recurring estimate, where there is some initial knowledge of Af,

[0183] as will be described below.

[0184] Determine Initial frequency offset estimate - RIB-RPS initialization

[0185] During initialization when Af (t) = Af(ta) at the start of the first observation interval is unknown the estimated phase drift A<p from (3a) is also unknown and hence N from equation (6) needed for phase wrap around compensation cannot be determined.

[0186] Therefore, as illustrated in Figures 7a and 7b, fora wrap around error free estimation of f(ta), tb- tashall:

[0187] be small enough to ensure no wraparound, 71, or

[0188] include multiple sub measurement periods (Atsub) performed within the total measurement period {tb- ta) to detect presence of wrap arounds, 72, allowing potential wrap around compensation for A<p measurement. Time between such sub measurements to avoid wrap arounds for an estimated max Af = Afmax estshould fulfill below equation:

[0189]

[0190] 2TT Afmaxest Atsub <

[0191] As an example, using a RF carrier frequency of 2GHz and a measurement period of 50ms will then corresponds to a maximum allowed frequency offset of 10ppb. Similar for a RF carrier frequency of 28GHz with an assumed maximum frequency offset of 10ppb this initial measurement needs to be performed with a maximum measurement period of ~3,6ms.

[0192] Even if Figure 7b shows a measurement period without FCW corrections as an example, measurements could be performed even if such corrections occurs since the RIB- RPS function is aware of such corrections and can perform phase compensation to its measurements.

[0193] Estimated dimensioning max frequency offset ( Afmaxest) used as input to determination of RIB-RPS entity maximum measurement period for its initial frequency offset measurement could be determined based one or more of:

[0194] 73, Information from TRP synchronization system of estimated frequency error, historical FCW updates and their statistics (that in a well synchronized system over time should have zero mean characteristics) i.e., occurrence of large FCW would indicate relative higher probability of larger frequency offsets. A smaller frequency offset could be achieved by selection of initial frequency measurement to be coordinated and occur during periods where each TRP has small deviation to each of its determined mean FCW references.

[0195] 74, Information about TRP synchronization system such as synchronization source(s), characteristics of fronthaul synchronization distribution (like PTP), TRPs reference oscillator designs, temperature variations etc

[0196] Based on an existing standardized max value e.g. from 3GPP defined frequency error in TS 38.104.

[0197] Alternative the RIB-RPS function could perform initial RIB-RPS measurements by testing with various periodicity e.g., starting from short periods and gradually increasing making sure no phase wrap around occurs or detecting its occurrence.

[0198] If combined with inter TRP Time Alignment Error (TAE) measurements, a change in TAE overtime could be used as an estimate for Afmax estor determine if an estimated RIB-RPS frequency offset includes bias error caused by measurement wrap around effects since a measured TAE drift would then be larger than expected from estimated RIB-RPS frequency offset.

[0199] Determine Recurring frequency offset estimates

[0200] After measurements at initialization, a first estimated value of the instantaneous frequency offset, Af (t), at the time tafrom equation (5) will be available at start of new observation intervals for recurring observations.

[0201] This means that for such recurring measurements, wrap around effects due to frequency offsets at the beginning of the new measurement interval Af(ta) and FCW updates occurring during observation interval will be known with acceptable accuracy. This in its turn means the phase impact, including wrap-round effects, can be known and compensated for in measured A<p, thereby allowing for longer observation periods {tb- ta) without more frequent sub measurement intervals when deriving a new updated frequency offset estimate.

[0202] Observation periods after initialization still needs to be restricted in time due to two reasons.

[0203] First, the deterioration of modeled Af calls for a new estimate that can correct the model with regular intervals, to maintain an updated accurate Af estimate as shown in Figure 3.

[0204] Second, remaining unknown uncertainties at the start of a new observation period due to errors in earlier estimates i.e. f(ta) #= Af(ta), additional errors from a potential hardware frequency drift, AfHW( ), component i.e. a drift which is not included and corrected by regular frequency updates, FCW( ), and from measurement uncertainties for the phase drift A$. Hence the largest measurement period for recurring estimates to avoid unknown phase wrap arounds causing frequency offset estimate errors would be decided by the error sources contributing to the phase error e defined in equation (6), where |e| < n As an example, if the remaining frequency offset uncertainty is 0,1 ppb from all above error components, with a carrier frequency of 2GHz, this will correspond to measurement period to be within 2,5 s for |e| < n.

[0205] It should be mentioned that uncertainty in frequency offset at the beginning of a new measurement interval Af(ta) also includes hardware related drift AfHW(t) from previous measurements and hence also periods between measurements need to consider this avoiding too large errors in Af(ta) and thereby risk of unknown wrap around effects.

[0206] Automated measurement period

[0207] Initial measurement could be used to determine a first estimate of e and serve as a first indication of max observation period Atobsand time between recurring observations tperiod for the later steps. An estimate of e was given in equation (7). The magnitude of e indicates how well the model fits the measured phase drift, where smaller value means a better fit. Updated estimates of e could be derived on regular basis. Updates of e could be used to detect if a change in measurement period should be performed.

[0208] It is disclosed that the RIB-RPS entity 213 receives information 8a, 8b about regular FCW updates, with associated time stamps, from the TRPs, 211, 212 according to Figure 8 and can thereby use it to correct and compensate instantaneous frequency changes occurring during a RIB-RPS frequency measurement period to prevent such otherwise unknown updates causing measurements errors i.e. at the end of an observation interval tba new updated value of the instantaneous frequency offset is determined according to equation (5).

[0209] By this the RIB-RPS measurements will not need to be coordinated and restricted to occur between FCW and can thereby also be extended in time for better accuracy. Figure 9 illustrates phase variation including FCW update effects. Here it can be seen that short measurement intervals can be decided with different length, exemplified in the figure by measurement intervals 91a, 91b,..., 91x, restricted by and aligned with FCW periods, long measurement intervals 92a, 92b not aligned or restricted by FCW periods.

[0210] After a RIB-RPS measurement 91a, 91b,..., 91x, 92a, 92b an instantaneous frequency offset is determined also taking potential FCW updates t1-t5, 93 occurring during a measurement into consideration. Between measurement periods the RIB-RPS entity receives regular FCW updates 94 and can thereby use it to maintain an instantaneous frequency offset between regular measurements 92a, 92b.

[0211] Even if a longer total measurement period is used 92a, multiple measurements could be performed 91a, 91b,..., 91x within such period to average out random noise components in measurements.

[0212] Measurement periods could be performed in various ways in addition to how illustrated in Figure 9, measurements could be sequential without gaps in-between or even overlapping.

[0213] With reference to Figure 10, an alternative to only monitor the frequency offset by RIB-RPS will be illustrated. Here it can be seen that the results from monitoring could be used to frequency synchronize a sync receiver, 212 in the figure, from a RIBS sync provider, 211 in the figure, i.e. not only for monitoring relative offset, thereby illustrating an example of RIB-RPS-S to control and align the frequency of a sync receiver 212 towards a sync provider 211.

[0214] In such scenario the sync receiver 212 may use sync provider 211 for both time and frequency synchronization and the RIBS sync provider 211 could be synchronized by:

[0215] A. An external sync source which could e.g., be a remote or local GNSS, received by 1 pulse per second (1PPS) or PTP, with a time relation to a known time reference 102 like GPS time.

[0216] B. Another sync provider

[0217] C. A local time reference derived by the sync provider itself and could be without a connection to a known time reference like GPS, i.e. free running.

[0218] For A and B the sync provider could adjust its time by change of its frequency, such as through a FCW, as in previous cases and hence this needs to be reported 10a to the RIB-RPS entity 213. Based on frequency offset measurements the RIB-RPS entity 213 signals frequency control information 10b to the sync receiver 212 who applies this to correct for a RIB RPS measured frequency offset. In return the RIB-RPS entity 213 needs to know when in time frequency control was applied by the sync receiver 212. Between such frequency control, regular FCW information from sync source is distributed 10c towards the sync receiver 212 allowing it to maintain frequency synchronization also inbetween frequency control from the RIB-RPS entity 213.

[0219] Predictor and drift characteristics

[0220] By regular RIB-RPS measurements, a predicted and estimated offset can be compared towards a related measured offset as illustrated in Figure 11, where estimated offset <p(ty )= A<p(tx) + 2n Af(f) dt and ε is the difference between measured and

[0221]

[0222] predicted offset.

[0223] The predicted value based in equation (5) includes known FCW updates and hence a deviation between predicted and measured RF phase offset, when measured offset compensated for potential wrap around effects as previously described, will include both RIB-RPS measurement errors and drift components not covered or corrected by FCW updates i.e. the error sources included in earlier defined phase error e.

[0224] Hence from the phase error e and time between two subsequent measurements (ty-tx) a frequency offset error (Aferror) in RIB-RPS determined frequency offset could be derived from below expressions.

[0225] ε = 2π Δferror(ty- tx) → Δferror

[0226]

[0227] Frequency offset errors may, depending on its characteristics when derived from historical e, e.g. if there exists a deterministic bias error component or a predictable drift caused by oscillator temperature drift, be used to provide better frequency offset estimates between measurements.

[0228] As an alternative or a combination to above method with historical measurements to reveal and compensate a RIB-RPS measurement bias error, inter TRP Time Alignment Error (TAE) over the air measurements (RIB-TAE) could be used to reveal and quantify such bias i.e. if a change of TAE over time does not correspond to measured RIB-RPS frequency offsets (e.g. an unknown frequency offset bias of 0,1 ppb over 100 s will explain a 10ns delta between measured change of TAE and estimated change based on RIB-RPS measurements). Whether such method is feasible depends on how accurate RIB-TAE can perform such measurements. Quality index indicator

[0229] The quality of the RIB-RPS frequency offset estimates will impact the quality for services depending on the accuracy of the frequency synchronization or accuracy of the knowledge of frequency offsets. Assessment of RIB-RPS performance and providing status towards end applications will be a key component supporting a service especially in a mobile system where both actual and required performance could vary. This could also be related to verifying a subscribed performance level for the end application, e.g., a positioning or radar sensing service.

[0230] Earlier mentioned deviation between modelled and measured offset could be used including forming statistics from historical measurements to serve as an indication of RIB-RPS measurement quality.

[0231] Possible examples of other factors as input to a quality index are:

[0232] SINR for RIB-RPS observations, including interference estimate, and RIBS channel characteristics with relation to system bandwidth, BW and time resolution, multipath resolution.

[0233] As mentioned earlier assessment of measurement quality may also be used by the RIB-RPS entity itself to perform measurement sufficient frequent to avoid earlier mentioned phase wrap around issue and potentially improve estimates.

[0234] The quality index would be dependent on potential frequency offset estimate improvements as described above in relation to “Automated measurement period”.

[0235] Also here earlier mentioned relative comparison from RIB-TAE measurements could be used as an alternative or complementary measure to assess RIB-RPS performance.

[0236] Interaction with applications

[0237] In the following it will be shown how RIB-RPS function may interact with applications.

[0238] Application Carrier Phase Positioning (CPP)

[0239] Carrier Phase Positioning (CPP) can be done from measurements in UL direction or in DL direction. In both cases, it relies on that the RF phase between TRPs is very stable over time, or in other words, that the RF carrier frequencies are exactly the same. Any difference in carrier frequency introduces a phase drift over time, that has the effect that the UE appears to move although it is actually standing still. This may cause positioning error, or distortion to a tracked UE route, so that the beginning of a closed loop route appears different from the end of the route. A frequency offset could hence give a positioning error that increases over a tracking period.

[0240] It is disclosed that FCW updates are continuously reported to the CPP entity, with timestamp, and this can be done either from TRP or from RIB-RPS.

[0241] RIB-RPS may report measured instantaneous frequency offset at a point in time to the CPP entity, comprising a time stamp.

[0242] It is also disclosed that, optionally, a predicted absolute frequency drift component, which is not FCW related, and quality index for RIB-RPS measurements may be measured and reported.

[0243] The CPP entity may use received information in the compensation of each new relative positioning phase measurement for accumulated phase drift, as illustrated in Figure 12 for an UL case shown. Examples of this is illustrated in Figures 12 to 15 as exemplified below:

[0244] UL CPP compensates phase with accumulated phase drift from the frequency offset and the FCW in the CPP entity, as illustrated in Figure 12.

[0245] DL CPP alt 1: UE reports Phase difference in DL and sends the report to the CPP entity. The CPP entity compensates for accumulated phase drift, see Figure 13.

[0246] DL CPP alt 2: UE does positioning autonomously, accumulated phase drift from measured frequency offset and FCW which is reported to UE on side channel, see Figure 14.

[0247] DL CPP alt 3: UE does positioning autonomously as shown in Figure 15. Positioning reference symbols precoded to compensate for accumulated phase drift from measured frequency offset and the FCW or the TRPs are synchronized towards each other as shown in Figure 10. Optionally the quality index could be reported to the UE.

[0248] Application ISAC Radar

[0249] For ISAC Radar one key aspect is to detect a moving object in a cluttered environment by detecting changes in received RF phase over time when a sensing signal is reflected at a moving object. One typical challenging use case is a slow-moving small object close to a large object like a building i.e. small difference in relative speed and large difference in Radar Cross Section (RCS).

[0250] If distributed TRPs involved in bi-static radar sensing, i.e. one TRP act as a TX node and the other TRP as a RX node, any RF phase drift between the transmitting and receiving node, e.g. due to a frequency offset, will cause a stationary object to falsely be detected as it will be moving or similar a moving object falsely detected as stationary, even small frequency offsets like few ppb would be in same order as doppler from a pedestrian.

[0251] As illustrated in Figure 16, it is disclosed that:

[0252] FCW updates are continuously reported to a radar sensing entity 161, with timestamp, and this may be done either from TRP1 / TRP2211, 212 or from RIB- RPS 213.

[0253] RIB-RPS 213 reports measured instantaneous frequency offset to the radar sensing entity 161, comprising time stamp.

[0254] Optionally: RIB-RPS 213 may report predicted an absolute frequency drift component, not FCW related, and a quality index for the RIB-RPS measurements. The radar sensing entity 161 uses received information and compensates each received sensing signal phase for inter TRP drift.

[0255] Alternatively, as shown in Figure 17 sensing reference symbols may be precoded to compensate for accumulated phase from measured frequency offset and FCWor, as shown in Figure 18, the TRPs 211, 212 may be synchronized towards each other.

[0256] Application DL C-JT

[0257] Interaction with DL Coherent Joint Transmission (C-JT) will now be exemplified, as shown in Figure 19.

[0258] A UE does a channel estimate with regular intervals, and the result is fed back to a DL C-JT precoder 191 via channel state information (CSI) feedback channel. This CSI feedback contains a co-phasing component that is applied between the two TRPs, 211, 212, joint transmission. Since the feedback has some inherent delay, and the channel estimation is done with some periodicity, the CSI feedback co-phasing component may become aged due to the frequency offset between TRPs, 211, 212.

[0259] The RIBS-RPS 213 may improve this aging by providing an estimated frequency offset to the DL C-JT precoder entity 191. The precoder can compensate for the phase variation from frequency offset in the precoder on the co-phasing component, so that the effect of frequency offset is neutralized.

[0260] The frequency offset quality index may be used by the DL C-JT to decide whether or not to apply C-JT at a particular instance, or if feedback period should be changed or if it should revert to a more robust but lower performance Non-Coherent Joint Transmission scheme (NC-JT).

[0261] Figure 20 is a flowchart summary of an exemplifying embodiment of RIB-RPS. The RIB-RPS entity, CPP positioning, and Radar sensing functions could be distributed implemented in cloud nodes if the latency between them and the TRPs are in the same region as for low layer split, e.g less than 200 ps.

[0262] The RIB-RPS entity may be implemented in O-DU. For RIB-RPS to be an O-RAN extension the communication between them should be in O-RAN M-plane.

[0263] A number of embodiments will now be described, some of which may be seen as alternatives, while some may be used in combination.

[0264] The method comprises the following steps, which steps may be taken in any suitable order. Optional steps are referred to as dashed boxes in Figures.

[0265] Embodiments herein relates to a method performed by a first node to handle frequency synchronization between two or more second nodes, as illustrated in Figure 21.

[0266] With the purpose of setting up, or finding an optimal value of the observation interval avoiding unknown wrap around, it is disclosed that the observation interval, Atobs, may be determined in different ways. One embodiment to find an optimal observation interval discloses that the method may comprise that the first node 213:

[0267] Step 2101: estimates a maximum frequency offset Δfmax estbased on at least one of:

[0268] information from a synchronisation system of the at least two second nodes, on estimated clock frequency error, historical frequency updates and their statistics; information about the synchronization system of the at least two second nodes such as synchronization source(s), characteristics of fronthaul synchronization distribution, reference oscillator designs of the at least two second nodes, temperature variations etc; and

[0269] an existing standardized maximum value of the maximum frequency offset, e.g. from 3GPP defined frequency error in TS 38.104; and

[0270] Step 2102: determines the observation interval, Atobs, based on the maximum frequency offset Afmaxest- Another disclosed embodiment related to finding an optimal observation interval teaches a method comprising that the first node:

[0271] Step 2103: determines the observation interval, Atobs, by starting from a fixed or previous observation interval, Atobs, value and increasing the observation interval, Atobs, value.

[0272] It is disclosed that the method comprises that the first node 213:

[0273] Step 2104: receives, from at least two second nodes 211, 212, reports on updates of a frequency, such as a Frequency Control Word, FCW, indicating a value for adjustment of a reference clock frequency, Step 2105: may receive, from at least two second nodes 211, 212, respective timestamps, indicating a time of adjustment of the reference clock frequency relating to the received reports on updates of a frequency indicating a value for adjustment of a reference clock frequency at the respective second node.

[0274] Step 2106: receives, from at least one of the two second nodes, a phase p or frequency A measurement, and

[0275] Step 2107: estimates a frequency offset between the at least two second nodes 211, 212 related to frequency synchronization taking the reports and the phase A<p or frequency A measurement during an observation interval, Atobs, into account.

[0276] It is disclosed that estimating a frequency offset, Step 5, between the at least two second nodes 211, 212 related to frequency synchronization taking the reports and the phase A<p or frequency A measurement during an observation interval, Atobs, into account may further comprise that the first node 213 estimates a frequency offset based on a known time of adjustment of the reference clock frequency.

[0277] It is disclosed that in the estimation of the frequency offset between the at least two second nodes, may comprise that the first node:

[0278] Step 2108: determines the estimated frequency offset as un-valid if an update of the frequency was made during the observation interval, Atobs, and determines the estimated frequency offset as valid if an update of the frequency was not made during the observation interval, Atobs.

[0279] It is disclosed that estimating a frequency offset between the at least two second nodes may comprise that the first node:

[0280] Step 2109: compensates a measurement for an accumulated frequency update effect during the observation interval, Atobs, where the observation interval, Atobsmay stretch over one or several frequency updates.

[0281] It is further disclosed that the method may comprise that the first node:

[0282] Step 21010: determines an instantaneous frequency offset at a point in time between the at least two second nodes, based on the reports on updates of a frequency and phase A<p or frequency A measurement.

[0283] As illustrated in Figures 8 and 9, it is disclosed that the method may comprise that the first node 213:

[0284] Step 21011: updates the estimated frequency offset using received reports on updates of a frequency 8a, 8b, 94 in-between determined instantaneous frequency offsets 92a, 92b.

[0285] It is disclosed that the method may comprise that the first node: Step 21012: estimates or detects hardware, HW, related changes in frequency, i.e. frequency drift effects not caused by frequency updates, such as changes due to an oscillator, temp changes etc., and

[0286] Step 21013: determines a prediction model for compensation of the changes in frequency offset (Aferror) due to HW related effects based on a time interval between at least two subsequent measurements of the instantaneous frequency offset and estimated frequency offsets.

[0287] Disclosed embodiments teach that the method may comprise that the first node: Step 21014: use the prediction model to determine the observation interval, Atobs. It is also disclosed that the method may comprise

[0288] Step 21015: using the prediction model as basis to derive a frequency offset estimate quality index.

[0289] It is also disclosed that the method may comprise:

[0290] Step 21016: transmitting the compensation obtained from the prediction model, to at least one of the second nodes.

[0291] Embodiments herein teach that the first node 213 may act as a Radio Interface Based Radio Phase Stability, RIB-RPS, entity between a number of second nodes 211, 212, and where the second nodes 211, 212 may be acting as Transmission and Reception points, TRPs, in a multi-TRP system.

[0292] With renewed reference to Figure 4, it is shown that to perform the method steps above, the first node 213 is further configured to handle frequency synchronization between two or more second nodes 211, 212, the first node 213 being adapted to:

[0293] receive, from at least two second nodes 211, 212, reports 4a, 4b on updates of a frequency, such as a Frequency Control Word, FCW, indicating a value for adjustment of a reference clock frequency,

[0294] receive, from at least one of the two second nodes 212 a phase p or frequency A measurement 4c, and

[0295] estimate a frequency offset between the at least two second nodes 211, 212 related to frequency synchronization taking the reports 4a, 4b and the phase p or frequency A measurement 4c during an observation interval, Atobs, into account.

[0296] It is disclosed that in the reception of reports 4a, 4b on updates of a frequency indicating a value for adjustment of a reference clock frequency, from at least two second nodes 211, 212, the first node 213 may be adapted to also receive respective timestamps, indicating a time of adjustment of the reference clock frequency, at the respective second node.

[0297] It is also disclosed that, in the estimation of a frequency offset between the at least two second nodes 211, 212 related to frequency synchronization taking the reports 4a, 4b and the phase p or frequency A measurement 4c during an observation interval, Atobs, into account, the first node 213 may be further adapted to estimate a frequency offset based on a known time of adjustment of the reference clock frequency.

[0298] Disclosed embodiments teach that the first node 213, in the estimation of a frequency offset between the at least two second nodes, may be adapted to determine the estimated frequency offset:

[0299] as un-valid if an update of the frequency was made during the observation interval, and

[0300] as valid if an update of the frequency was not made during the observation interval, Atobs.

[0301] It is disclosed that the first node 213 may be adapted to compensate a measurement for an accumulated frequency update effect during the observation interval, Atobs, where the observation interval, Atobsmay stretch over one or several frequency updates as indicated in Figure 6a and 6b.

[0302] Embodiments teach that the first node 213 may be adapted to determine an instantaneous frequency offset at a point in time between the at least two second nodes 211, 212, based on the reports on updates of a frequency 4a, 4b and phase A<p or frequency A measurement 4c.

[0303] It is disclosed that the first node 213 may be adapted to find the optimal value of the observation interval, Atobs. This can be done in n different ways.

[0304] One disclosed embodiment teaches that the first node may be adapted to estimate a maximum frequency offset Afmaxestbased on at least one of:

[0305] - 73 information from a synchronisation system of the at least two second nodes 211, 212 on estimated clock frequency error, historical frequency updates and their statistics;

[0306] - 74 information about the synchronization system of the at least two the second nodes 211, 212 such as synchronization source(s), characteristics of fronthaul synchronization distribution, reference oscillator designs of the at least two second nodes 211, 212, temperature variations etc; and

[0307] - an existing standardized maximum value of the maximum frequency offset ^fmax est,e9- from 3GPP defined frequency error in TS 38.104. It is further disclosed that a determination of the observation interval, Atobs, may be based on the estimated maximum frequency offset Afmax est.

[0308] Another disclosed embodiment teaches that the first nosed 213 may be adapted to determine the observation interval, Atobs, by starting from a fixed observation interval, Atobs, value and increase the observation interval, Atobs, value.

[0309] As indicated in Figures 8 and 9, it is disclosed that the first node 81 may be adapted to update the estimated frequency offset using received reports on updates of a frequency 8a, 8b, 94 in-between determined instantaneous frequency offsets.

[0310] Embodiments teach that the first node may be adapted to:

[0311] Step 21012: estimate or detect hardware, HW, related changes in frequency, i.e. frequency drift effects not caused by frequency updates or FCW, such as changes due to an oscillator, temp changes etc., and

[0312] Step 21013: determine a prediction model for compensation of the changes in frequency offset (Aferror) due to HW related effects based on a time interval between at least two subsequent measurements of the instantaneous frequency offset and estimated frequency offsets.

[0313] It is also disclosed that the first node 213 may be adapted to use the prediction model to determine the observation interval, Atobs.

[0314] It is also disclosed that the first node may be adapted to use the prediction model as basis to derive a frequency offset estimate quality index.

[0315] It is also disclosed that the first node 213 may be adapted to use the prediction model to determine the time period between observations, tperiod.

[0316] The first node 213 may also be adapted to transmit the compensation obtained from the prediction model to at least one of the at least two second nodes 211, 212.

[0317] Figure 22, illustrates a first node, where it is exemplified that embodiments herein may be implemented through a processor or one or more processors, such as the processor 2210 of a processing circuitry in the first node 213 depicted in Figure 22 together with respective computer program code for performing the functions and steps of the embodiments herein. The program code mentioned above may also be provided as a computer program product, for instance in the form of a data carrier carrying computer program code for performing the embodiments herein when being loaded into the first node 213. One such carrier may be in the form of a CD ROM disc. It is however feasible with other data carriers such as a memory stick. The computer program code may furthermore be provided as pure program code on a server and downloaded to the first node 213. The first node 213 may further comprise a memory 2220 comprising one or more memory units. The memory 2220 comprises instructions executable by the processor in the first node 213. The memory 2220 is arranged to be used to store e.g., media functions, indications, tags, information, data, configurations, communication data, and applications to perform the methods herein when being executed in the first node 213.

[0318] In some embodiments, a computer program 2230 comprises instructions, which when executed by the at least one processor 2210 cause the at least one processor of the first node 213 to perform the steps above.

[0319] In some embodiments, a carrier 2240 comprises the computer program 2230 wherein the carrier 2240 is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer-readable storage medium.

[0320] Those skilled in the art will appreciate that units in the first node 213 described above may refer to a combination of analog and digital circuits, and / or one or more processors configured with software and / or firmware, e.g. stored in the first node 213 that when executed by the one or more processors such as the processors described above. One or more of these processors, as well as the other digital hardware, may be included in a single Application-Specific Integrated Circuitry ASIC, or several processors and various digital hardware may be distributed among several separate components, whether individually packaged or assembled into a System-on-a-Chip (SoC).

[0321] Although the computing devices described herein (e.g., UEs, network nodes) may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and / or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and / or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination. Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of the components may be partitioned between the processing circuitry and the communication interface. In another example, non-computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.

[0322] In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry executing instructions stored on in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer-readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hard-wired manner. In any of those particular embodiments, whether executing instructions stored on a non-transitory computer-readable storage medium or not, the processing circuitry can be configured to perform the described functionality. The benefits provided by such functionality are not limited to the processing circuitry alone or to other components of the computing device, but are enjoyed by the computing device as a whole, and / or by end users and a wireless network generally.

[0323] When using the word "comprise" or “comprising” it shall be interpreted as nonlimiting, i.e. meaning "consist at least of".

[0324] The embodiments herein are not limited to the preferred embodiments described above. Various alternatives, modifications and equivalents may be used.

[0325] Abbreviations

[0326] C-JT Coherent Joint T ransmission

[0327] CPP Carrier Phase Positioning

[0328] CU central unit

[0329] DU distributed unit

[0330] eCPRI enhanced Common Public Radio Interface

[0331] FCW Frequency Control Word

[0332] GNSS Global Navigation Satellite System

[0333] HW hardware

[0334] ISAC Integrated Sensing and Communication

[0335] LTE Long Term Evolution

[0336] OTA Over the air

[0337] ppb Part per billion PTP Precision Time Protocol

[0338] RIBS Radio Interface Based Synchronization RIB-RPS RIB-Radio Phase Stability

[0339] RIB-RPS-M RIB-RPS monitoring

[0340] RIB-RPS-S RIB-RPS control

[0341] Rll radio unit

[0342] SINR Signal to interference noise ratio TAE Time Alignment Error

[0343] TRP Transmission and Reception Point 5G NR 5G New Radio

Claims

CLAIMS1. A method performed by a first node (213) to handle frequency synchronization between two or more second nodes (211, 212), the method comprising: receiving (2104), from at least two second nodes (211, 212), reports on updates of a frequency indicating a value for adjustment of a reference clock frequency, receiving (2106), from at least one of the two second nodes (212), a phase p or frequency A measurement, andestimating (2107) a frequency offset between the at least two second nodes (211, 212) related to frequency synchronization taking the reports and the phase Δφ̂ or frequency Δf̂ measurement during an observation interval, Δtobs, into account.

2. The method according to claim 1, wherein receiving, from at least two second nodes, reports on updates of a frequency indicating a value for adjustment of a reference clock frequency, further comprises:receiving (2105) respective timestamps, indicating a time of adjustment of the reference clock frequency, at the respective second node.

3. The method according to claim 1, wherein estimating a frequency offset between the at least two second nodes related to frequency synchronization taking the reports and the phase A<p or frequency A measurement during an observation interval, Atobs, into account further comprises:- estimating (21071) a frequency offset based on a known time of adjustment of the reference clock frequency.

4. The method according to any of claims 1-3, wherein estimating a frequency offset between the at least two second nodes comprises:- determining (2108) the estimated frequency offset as un-valid if an update of the frequency was made during the observation interval, Atobs, and- determining (2108) the estimated frequency offset as valid if an update of the frequency was not made during the observation interval, Atobs.

5. The method according to any of claims 1-3, wherein estimating a frequency offset between the at least two second nodes comprises:compensating (2109) a measurement for an accumulated frequency update effect during the observation interval, Atobs, where the observation interval, Atobsmay stretch over one or several frequency updates.

6. The method according to any one of claims 1 to 5, wherein the method comprises:determining (21010) an instantaneous frequency offset at a point in time between the at least two second nodes, based on the reports on updates of a frequency and phase Δφ̂ or frequency Δf̂ measurement.

7. The method according to any one of claim 1 to 6, the method comprising:estimating (2101) a maximum frequency offset Δfmax estbased on at least one of:information from a synchronisation system of the at least two second nodes, on estimated clock frequency error, historical frequency updates and their statistics;information about the synchronization system of the at least two second nodes such as synchronization source(s), characteristics of fronthaul synchronization distribution, reference oscillator designs of the at least two second nodes, temperature variations etc; andan existing standardized maximum value of the maximum frequency offset; anddetermining (2102) the observation interval, Atobs, based on the maximum frequency offset Afmaxest-8. The method according to any one of claims 1 to 6, the method comprising:determining (2103) the observation interval, Atobs, by starting from a fixed observation interval, Atobs, value and increasing the observation interval, Atobs, value.

9. The method according to any preceding claim, the method comprising:updating (21011) the estimated frequency offset using received reports on updates of a frequency in-between determined instantaneous frequency offsets.

10. The method according to any preceding claim, the method comprising:estimating or detecting (21012) hardware, HW, related changes in frequency; anddetermining a prediction model for compensation of the changes in frequency offset, Aferror, due to HW related effects based on a time interval between at least two subsequent measurements of the instantaneous frequency offset and estimated frequency offsets.

11. The method according to claim 10, the method comprising:using (21014) the prediction model to determine the observation interval, Atobs.

12. The method according to any one of claims 10 to 11, the method comprising:using (21014) the prediction model as basis to derive a frequency offset estimate quality index.

13. The method according to any one of claims 10 to 12, the method comprising:transmitting (21016) the compensation obtained from the prediction model to at least one of the two second nodes.

14. The method according to any preceding claim, wherein the first node is acting as a Radio Interface Based Radio Phase Stability, RIB-RPS, entity between a number of second nodes, and where the second nodes are acting as Transmission and Reception points, TRPs, in a multi-TRP system.

15. A first node adapted to handle frequency synchronization between two or more second nodes, the first node being adapted to:receive, from at least two second nodes, reports on updates of a frequency, indicating a value for adjustment of a reference clock frequency,receive, from at least one of the two second nodes, a phase p or frequency A measurement, andestimate a frequency offset between the at least two second nodes related to frequency synchronization taking the reports and the phase Δφ̂ or frequency Δf̂ measurement during an observation interval, Δtobs, into account.

16. The first node according to claim 15, being adapted to, in the reception, from at least two second nodes, of reports on updates of a frequency indicating a value for adjustment of a reference clock frequency:receive respective timestamps, indicating a time of adjustment of the reference clock frequency, at the respective second node.

17. The first node according to claim 15, being further adapted to, in the estimation of a frequency offset between the at least two second nodes related to frequency synchronization taking the reports and the phase A <p or frequency A f measurement during an observation interval, Atobs, into account:- estimate a frequency offset based on a known time of adjustment of the reference clock frequency.

18. The first node according to any of claims 15 to 17, being adapted to, in the estimation of a frequency offset between the at least two second nodes:- determine the estimated frequency offset as un-valid if an update of the frequency was made during the observation interval, Atobs,- determine the estimated frequency offset as valid if an update of the frequency was not made during the observation interval, Δtobs.

19. The first node according to any of claims 15 to 17, being adapted to, in the estimation of a frequency offset between the at least two second nodes:compensate a measurement for an accumulated frequency update effect during the observation interval, Δtobs, where the observation interval, Δtobsmay stretch over one or several frequency updates.

20. The first node according to any one of claims 15 to 18, being adapted to:determine an instantaneous frequency offset at a point in time between the at least two second nodes, based on the reports on updates of a frequency and phase A<p or frequency A measurement.

21. The first node according to any one of claims 15 to 20, being adapted to:estimate a maximum frequency offset Δfmax estbased on at least one of:information from a synchronisation system of the at least two second nodes, on estimated clock frequency error, historical frequency updates and their statistics;information about the synchronization system of the at least two second nodes such as synchronization source(s), characteristics of fronthaulsynchronization distribution, reference oscillator designs of the at least two second nodes, temperature variations etc; andan existing standardized maximum value of the maximum frequency update; anddetermine the observation interval, Atobs, based on the maximum frequency offset fmax est-22. The first node according to any one of claims 15 to 20, being adapted to:determine the observation interval, Atobs, by starting from a fixed observation interval, Atobs, value and increase the observation interval, Atobs, value.

23. The first node according to any one of claims 15 to 22, being adapted to:update the estimated frequency offset using received reports on updates of a frequency in-between determined instantaneous frequency offsets.

24. The first node according to any one of claims 15 to 23, being adapted to:estimate or detect hardware, HW, related changes in frequency; and determine a prediction model for compensation of the changes in frequency offset, 1 ferror, due to HW related effects based on a time interval between at least two subsequent measurements of the instantaneous frequency offset and estimated frequency offsets.

25. The first node according to claim 24, being adapted to:use the prediction model to determine the observation interval, Atobs.

26. The first node according to any one of claims 24 to 25, being adapted to:use the prediction model as basis to derive a frequency offset estimate quality index.

27. The first node according to any one of claims 24 to 26, being adapted to:transmit the compensation obtained from the prediction model to at least one of the two second nodes.

28. The first node according to any one of claims 15 to 27, being adapted to act as a Radio Interface Based Radio Phase Stability, RIB-RPS, entity between a number of secondnodes, and where the second nodes are adapted to act as Transmission and Reception points, TRPs, in a multi-TRP system.

29. A computer program comprising instructions, which when executed by a processor, causes the processor to perform steps according to any of the claims 1-14.

30. A carrier comprising the computer program of claim 30, wherein the carrier is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer-readable storage medium.

Images