METHODS PERFORMED BY A BASEBAND UNIT (BBU) SYSTEM OF A WIRELESS COMMUNICATION NETWORK; METHODS PERFORMED BY A FIRST RU OF A DISTRIBUTED BASE STATION SYSTEM, AND SAID BBU SYSTEM AND FIRST RU

AR125504B1Active Publication Date: 2026-08-26TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
ARP20220100512
Authority / Receiving Office
AR · AR
Patent Type
Patents
Current Assignee / Owner
Priority Date
2021-03-08
Filing Date
2022-03-07
Publication Date
2026-08-26
Estimated Expiration
2042-03-07

AI Technical Summary

Technical Problem

The increasing number of antennas in massive MIMO systems for 5G and future 6G wireless communication leads to a significant increase in fronthaul link capacity and cost due to the need to transport beamforming weights (BFWs) over the fronthaul, especially in cascaded RU topologies, where BFW bursts further exacerbate the peak rate requirements.

Method used

A method to decompose BFWs into a common first part and RU-specific second part, where the first part is sent only once over the fronthaul, and the second part is sent separately to each RU, reducing the overall data sent over the fronthaul link by compressing the BFWs to a higher degree than usual, especially for the second part, which is less sensitive to errors.

Benefits of technology

This approach reduces the fronthaul link capacity and cost by minimizing the data transmitted while maintaining manageable BFW bursts, thus optimizing the deployment of cascaded RU topologies in distributed base station systems.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

A method implemented by a baseband unit (BBU) system of a wireless communication network, comprising a distributed base station system (100) comprising a BBU (110), a first radio unit (RU) (120) connected to the BBU (110) via a fronthaul link (140). The base station system (100) further comprises a second RU (160) connected to the first RU (120) via an RU link (165), wherein the first and second RUs are cascaded to the BBU.The method comprises determining a first portion of beamforming weights (BFW) in the frequency domain based on an estimate of the DL channel of the first RU (120), wherein the BFW will be used to send the K user layer beamforming DL signals to a number of UEs (131, 132, 133); determining a second portion of the BFW based on an estimate of the DL channel of the second RU (160); and determining a first portion of the BFW based on the DL channel estimates of the first RU (120) and the second RU (160), to perform interference cancellation between the user layer signals. The method further comprises sending the first portion of the BFW and the first and second portions of the BFW to the first RU (120) and sending the K user layer signals in the frequency domain to the first RU (120).Also a method in the first BBU for beaming the K user layer signals comprising using the first portion and the first portion of the BFW and sending the first portion and the second portion of the BFW and the K user layer signals to the second RU (160).
Need to check novelty before this filing date? Find Prior Art

Description

METHODS, BASEBAND UNIT SYSTEM AND RADIO UNIT OF A DISTRIBUTED BASE STATION SYSTEM ADAPTED FOR BEAM FORMATION TECHNICAL FIELD This disclosure generally relates to methods, baseband unit (BBU) systems, and radio units (RUs) of a distributed base station system adapted for beamforming. More specifically, this disclosure addresses such methods, systems, and units when the RU has a plurality of antennas for transmitting downlink (DL) signals and receiving uplink (UL) signals, as in multiple-input multiple-output (MIMO) systems. This disclosure also relates to software and related software for the aforementioned methods, systems, and units. BACKGROUND In a centralized radio access network (C-RAN), also called a distributed base station system, processing is performed by two separate units: a radio unit (RU) and a baseband unit (BBU). The BBU is connected to the RU via a fronthaul link. The RU may also be called a remote radio unit (RRU). The baseband unit may also be called a base unit (BU), digital unit, or distributed unit (DU). The RU is connected to one or more antennas through which it communicates wirelessly with at least one user equipment (UE). The BBU, in turn, is connected to other base station systems or base stations, and to a core network of a wireless communication system. The BBU is centralized, and there is typically more than one RU connected to each BBU.Traditionally, the BBU performs advanced radio coordination functions, such as joint detection, joint decoding, coordinated multipoint transmission (CoMP), to increase spectrum efficiency and network capacity, as well as baseband processing, while the RU performs radio frequency (RF) processing and transmission / reception of the RF-processed signals. Originally, the RU was designed to reduce cable loss in coaxial cables between the top of an antenna tower, where the actual antennas are located, and the bottom of the tower, where the base station functionality is housed. Therefore, prior to fifth-generation communications 1689323 of 44 mobiles (5G), that is, in 4G, for example, Long Term Evolution (LTE), the RU was quite simple and mainly performed RF processing with limited baseband processing, if any. As 4G transitioned to 5G, there was a need to increase wireless communication capacity to the User Entities (UEs) to deliver the data volumes required per time period in 5G. A key enabler of the mobile evolution to 5G is massive multiple-input multiple-output (MIMO), in which each User Entity (RU) has a plurality of antennas. In other words, massive MIMO leverages spatial multiplexing to improve spectrum efficiency by using antenna arrays within the RU. Each antenna array is equipped with N antennas that simultaneously serve K user layers within the same time-frequency resource. The typical scenario is N >> K; for example, N might be 64, 128, or 256, while K might be 8 or 16. As shown, the number of antennas is quite large.Massive MIMO is often referred to as massive beamforming, which can form narrow beams and focus them in different directions, mitigating the increased path loss in higher frequency bands. It also benefits multi-user MIMO, enabling simultaneous transmissions to and from multiple UEs across separate spatial channels resolved by massive MIMO technologies, while maintaining high capacity for each UE. Therefore, it can significantly increase spectrum efficiency and cell capacity. In the evolution of 5G and the future sixth generation of mobile communications (6G), massive MIMO is expected to support even more antennas, as the cost per transceiver chain decreases over time. To address this trend, MIMO processing is anticipated to become more distributed and scalable, with a larger MIMO system being processed by multiple Routing Units (RUs), each processing only a subset of antennas. With such a scalable design, the MIMO system can easily scale up in terms of the number of antennas. In the legacy Common Public Radio Interface (CPRI) type of time-domain (FH) fronthaul, the BBU performs beamforming and determines the time-domain IQ samples per antenna branch. These time-domain IQ samples per antenna branch are then transported, in DL, across the fronthaul link to the RUs. Because there are many antennas, i.e., antenna branches, in massive MIMO systems, the required FH link capacity increases dramatically with the number of antenna branches, causing FH costs to rise significantly. To address this challenge, several initiatives have been developed. 1689323 of 44 adopted different Lower Layer Splitting (LLS) options. The basic idea is to move the beamforming function to the RU and transport the samples or data from the user layers in the frequency domain across the FH link. In this way, the number of FH streams is reduced from the number of antennas N to the number of user layers K; remember that in the typical scenario N >> K. One LLS option is called 3GPP Option 7-3 and is presented in 3GPP TR 38.801 V14.0.0, clause 11.1.2.7. In 3GPP Option 7-3, the channel estimation of the radio channel between the RU and the UEs is performed in the BBU, based, for example, on UL reference signals, as well as the calculation of beamforming weights (BFW) in the frequency domain. Next, the BFW in the frequency domain is sent by the FH to the RU.The BBU also performs forward error correction (FEC) encoding and sends the encoded bits from the user layers through the FH to the RU. The RU modulates the received user layer signals and beamforms them into antenna signals using the BFW received from the BBU. The RU further transforms the antenna signals in the time domain using, for example, Inverse Fast Fourier Transform (IFFT), and performs the necessary RF processing before the antenna signals are wirelessly transmitted to the UEs via their respective antennas on the antenna branches. Another LLS option is called O-RAN LLS and is described in ORAN.WG4.CUS.0-v04.00, O-RAN Fronthaul Working Group, Control, User and Synchronization Plane Specification. This option is very similar to 3GPP option 7-3.The difference is that modulation is performed in the BBU instead of the RU in O-RAN, so the RU receives the modulated user layer signals from the BBU. O-RAN defines two types of RUs: CAT-A RU and CAT-B RU, which are called CAT-A O-RU and CAT-B O-RU, respectively, in O-RAN. For CAT-A O-RU, frequency-domain beamforming is implemented in the BBU, which is called the O-DU in O-RAN. For CAT-B O-RU, frequency-domain beamforming is implemented in the CAT-B O-RU. In O-RAN, frequency-domain beamforming is called precoding, and beamforming is called precoding in CAT-B O-RU. Note that the BBU and RU are referred to as O-DU and O-RU, respectively, in O-RAN. In eCPRI terminology, the BBU and RU are referred to as eREC (eCPRI Radio Equipment Control) and eRE (eCPRI Radio Equipment), respectively.In other terminology, the BBU and the RU can be called LLS-CU and LLS-DU, respectively. 1689323 of 44 Although the LLS options described above reduce the amount of DL user plane data sent through the FH, the BFWs that must be sent in the control plane create high bursts on the FH, which is then the primary driver for increasing the peak FH rate. This is because the transmission window for carrying BFWs is very short. Therefore, it creates high bursts to transport so many BFWs in such a short time. In international patent application WO2020 / 256609 by the same applicant, a scheme was proposed to reduce the number of BFWs carried by the FH. The idea described there is to split the BFWs into two parts. The BFWs of the second part are significantly reduced by performing a specific layer beam selection. Then, the BFWs of the first part are calculated based on the reduced BFWs of the second part. The amount of data from both parts is much less than without reduction, while the performance degradation is small. In document WO2020 / 256609, the BBU sends the BFWs of both the first and second parts to the RU. The RU then calculates, or combines, the BFWs based on the first and second parts received and then performs beamforming using the combined BFWs. However, WO2020 / 256609 only addresses a point-to-point FH topology, where each RU has a dedicated FH link to the BBU, as shown in Figure 1. In the point-to-point FH topology example of a distributed base station system 10 as shown in Figure 1, a first RU 30 is connected to a BBU 20 via a first FH link 25, a second RU 40 is connected to BBU 20 via a second FH link 35, and a third RU 50 is connected to BBU 20 via a third FH link 45. The first, second, and third RUs 30, 40, and 50 are arranged to transmit and receive user scheduler data as antenna signals to / from UEs 31, 32, and 33. Such a point-to-point FH topology would require many fiber connections and the same number of BBU ports, even if those RUs are configured as a larger joint MIMO system. This description, on the other hand, addresses a cascaded RU topology, as shown in Figure 2. In a cascaded topology, a first RU 120 is connected to the BBU 110 via a fronthaul link 140, as in the point-to-point topology. However, the second RU 160 connects to the first RU 120 via an independent RU-RU link 165, and a third RU 170 connects to the second RU 160 via another independent RU-RU link 175. Furthermore, if there are any additional RUs, they are in turn connected to another RU, one after the other, as in a line. In Figure 2, Figure 1689323 of 44 illustrates only three RUs, but there can be many more RUs in a cascaded topology of this type. Implementing cascaded RUs would reduce the number of FH fiber links and the number of BBU ports to 1. This would help reduce implementation costs (i.e., fiber connections) and system complexity (i.e., BBU ports). However, in the cascaded topology, especially when using the methods described in WO2020 / 256609, BFW bursts across the FH 140 link increase even further, since all BFWs sent by the BBU to all the cascaded RUs are sent over the same FH 140 link. Therefore, the peak FH rate in the cascaded FH chain increases even more, particularly on the FH 140 link between BBU 110 and the first RU 120. Consequently, there is a need for a solution to manage distributed base station systems with RUs cascaded to the BBU. Preferably, such a solution should keep the total amount of data sent across the FH link low, while also keeping BFW bursts sent across the FH link at a manageable level. SYNTHESIS It is an object of the invention to address at least some of the problems and issues described above. These and other objects can be achieved through the use of methods, network nodes, and wireless devices as defined in the appended independent claims. According to one aspect, a method is provided that is implemented using a BBU system of a wireless communication network, which wireless communication network comprises a distributed base station system. The distributed base station system comprises a BBU, a first RU connected to the BBU via a fronthaul link and comprising N1 antennas, and a second RU connected to the first RU via an RU link and comprising N2 antennas. The method comprises determining a first BFW portion in the frequency domain, based on an estimate of the DL channel of the first RU, wherein the BFW will be used to send K beamforming user-layer DL signals to various UEs. The first portion of the BFW is determined to expand, in the frequency domain, the K user-layer signals to the antenna signals of the N1 antennas of the first RU.The method also includes determining a second portion of the BFW in the frequency domain, based on an estimate of the DL channel of the second RU; the second portion of the BFW is determined to expand, in the. 1689323 of 44 frequency domain, the K user layer signals to the N2 antenna signals of the second RU. The method further comprises determining a first part of the BFW in the frequency domain, based on the DL channel estimates of the first RU and the second RU, where the first part of the BFW is determined to perform interference cancellation between the user layer signals. In addition, the method comprises activating the transmission of the first part of the BFW and the first and second portions of the BFW to the first RU, and activating the transmission of the K user layer signals in the frequency domain to the first RU. According to another aspect, a method is provided that is implemented by a first RU of a distributed base station system, where the first RU comprises N1 antennas. The distributed base station system further comprises a BBU connected to the first RU via a fronthaul link and a second RU connected to the first RU via an RU link, where the second RU comprises N2 antennas. The method comprises receiving, from the BBU, K frequency-domain user-layer DL signals for transmission to several UEs. The method further comprises receiving, from the BBU, a first portion of BFW, and first and second portions of BFW for beamforming the K frequency-domain user-layer signals.The first portion is based on a determined DL channel estimate from the first RU, and this portion is designed to expand, in the frequency domain, the user layer signals to antenna signals from the N1 antennas of the first RU. The second portion is based on a determined DL channel estimate from the second RU, and this portion is designed to expand, in the frequency domain, the user layer signals to antenna signals from the N2 antennas of the second RU. The first part of the BFW is based on the determined DL channel estimates from the first and second RUs, and this first part of the BFW is designed to perform interference cancellation between the user layer signals.The method further comprises determining the BFWs for the first RU based on the first part and the first portion received from the BFWs, and sending the K received user layer DL signals in the frequency domain to the second RU via the RU link. The method further comprises beamforming the K received user layer signals into antenna signals, using the BFWs determined for the first RU, and sending the antenna signals to the UE number via antennas N1. According to another aspect, a method is provided that is carried out by 1689323 of 44 a BBU system of a wireless communication network, wherein the wireless communication network comprises a distributed base station system comprising a BBU, a first RU connected to the BBU via a fronthaul link, the first RU comprising N1 antennas. The distributed base station system further comprises a second RU connected to the first RU via an RU link, wherein the second RU comprises N2 antennas. The method comprises determining a first BFW portion in the frequency domain, based on an estimate of the UL channel of the first RU, wherein the first BFW portion will be used to combine, in the frequency domain, the antenna signals received at the N1 antennas of the first RU from several UEs to K user layer signals of the first RU.The method further comprises determining a second portion of the BFW in the frequency domain, based on an estimate of the UL channel of the second RU, where the second portion of the BFW will be used to combine the frequency domain signals received at the N2 antennas of the second RU from the number of UEs to K user layer signals of the second RU. The method further comprises determining a first portion of the BFW in the frequency domain based on the UL channel estimates of the first and second RUs, where the first portion of the BFW is determined to perform interference cancellation between the user layer signals of the first and second RUs and to enable the transmission of the first portion of the BFW and the first and second portions of the BFW to the first RU. According to another aspect, a method is provided that is carried out by a first RU of a distributed base station system, where the first RU comprises N1 antennas. The distributed base station system further comprises a BBU connected to the first RU via a fronthaul link and a second RU connected to the first RU via an RU link, where the second RU comprises N2 antennas.The method comprises receiving, from several UE, UL antenna signals at the N1 antennas, and receiving, from the BBU, a first part of BFW and a first and second part of BFW, wherein the first part is based on a determined UL channel estimate from the first RU, the first part is determined to combine, in the frequency domain, the UL antenna signals received at the N1 antennas with user layer signals, the second part is based on a determined UL channel estimate from the second RU and the second part is determined to combine, in the frequency domain, antenna signals received at the N2 antennas of the second RU with user layer signals, the first part of the BFW is based on the determined UL channel estimates from the first RU and the second RU and the. 1689323 of 44 The first part of the BFW is determined to perform interference cancellation between the user layer signals. The method further comprises determining the BFW for the first RU based on the first part and the first portion received from the BFW, and sending to the second RU, via the RU link, the first part and the second portion received from the BFW, and combining, in the frequency domain, the UL antenna signals received at the N1 antennas into the first K user layer signals, using the BFW determined for the first RU. The method further comprises receiving, from the second RU, the second K user layer signals combined by the second RU from antenna signals received at the N2 antennas of the second RU from various UEs, the combination by the second RU being based on the first part and second portion of the BFW sent to the second RU by the first RU.The method further comprises combining the first and second K user layer signals into K combined user layer signals, and sending the K combined user layer signals to the BBU. According to another aspect, a BBU system is provided that is configured to operate in a wireless communication network. The wireless communication network comprises a distributed base station system comprising a BBU, a first RU connected to the BBU via a fronthaul link, where the first RU comprises N1 antennas. The distributed base station system further comprises a second RU connected to the first RU via an RU link, where the second RU comprises N2 antennas. The BBU system also comprises a processing circuit and a memory.This memory contains instructions executable by this processing circuit, so the BBU system is operational to determine a first portion of BFW in the frequency domain, based on an estimate of the DL channel of the first RU, where the BFW will be used for the formation of K beams of the user layer DL signals that will be sent to several UE, the first portion of the BFW is determined to expand, in the frequency domain, the K user layer signals to antenna signals of the N1 antennas of the first RU.The BBU system is also operational for determining a second portion of the BFW in the frequency domain, based on an estimate of the DL channel of the second RU. The second portion of the BFW is determined to expand, in the frequency domain, the K signals of the user layer to the antenna signals of the N2 antennas of the second RU. It is also used to determine a first part of the BFW in the frequency domain, based on the DL channel estimates of the first RU and the second RU, where the first part of the BFW is... 1689323 of 44 determines how to perform interference cancellation between user layer signals. The BBU system is also operational to activate the transmission of the first part of the BFW and the first and second portions of the BFW to the first RU, and to activate the transmission of the K user layer signals in the frequency domain to the first RU. According to another aspect, a first RU is provided that is configured to operate in a distributed base station system, where the first RU comprises N1 antennas. The distributed base station system further comprises a BBU connected to the first RU via a fronthaul link and a second RU connected to the first RU via an RU link, where the second RU comprises N2 antennas. The first RU comprises a processing circuit and a memory. This memory contains instructions executable by the processing circuit, so the first RU is operational for receiving, from the BBU, K frequency-domain user-layer DL signals to be sent to various UEs, and for receiving, from the BBU, a first portion of BFW and a first and second portion of BFW for beamforming the K frequency-domain user-layer signals.The first portion is based on a determined DL channel estimate from the first RU. This portion is designed to expand, in the frequency domain, the user layer signals to the antenna signals from the N1 antennas of the first RU. The second portion is based on a determined DL channel estimate from the second RU. This portion is designed to expand, in the frequency domain, the user layer signals to the antenna signals from the N2 antennas of the second RU. The first part of the BFW is based on the determined DL channel estimates from the first and second RUs, and this first part of the BFW is designed to perform interference cancellation between the user layer signals. The first RU is also responsible for determining the BFW for itself based on the first part and the first portion of the received BFW and sending it to the second RU via the RU link.the first and second portions received from the BFWs, and send to the second RU, via the RU link, the K DL signals from the user layer received in the frequency domain, and beam the K signals received from the user layer into antenna signals, and use the BFWs determined for the first RU, and send the antenna signals to the UE number via the N1 antennas. According to another aspect, a BBU system is provided that is configured to operate on a wireless communication network. The network of 1689323 of 44 Wireless communication comprises a distributed base station system comprising a BBU, a first RU connected to the BBU via a fronthaul link, wherein the first RU comprises N1 antennas. The distributed base station system further comprises a second RU connected to the first RU via an RU link, wherein the second RU comprises N2 antennas. The BBU system also comprises a processing circuit and a memory.The memory contains instructions executable by the processing circuits, so the BBU system is operational to determine a first portion of BFW in the frequency domain, based on an estimate of the UL channel of the first RU, where the first portion of BFW will be used to combine, in the frequency domain, antenna signals received on the N1 antennas of the first RU from a number of UEs to K user layer signals of the first RU, and to determine a second portion of BFW in the frequency domain, based on an estimate of the UL channel of the second RU, where the second portion of BFW will be used to combine frequency domain signals received on the N2 antennas of the second RU from the number of UEs to K user layer signals of the second RU.The BBU system is also operational for determining a first part of the BFW in the frequency domain based on the UL channel estimates of the first RU and the second RU, where the first part of the BFW is determined to perform interference cancellation between the user layer signals in the first and second RU and activate the sending of the first part of the BFW and the first and second portion of the BFW to the first RU. According to another aspect, a first RU is provided that is configured to operate in a distributed base station system, where the first RU comprises N1 antennas. The distributed base station system further comprises a BBU connected to the first RU via a fronthaul link and a second RU connected to the first RU via an RU link, where the second RU comprises N2 antennas. The first RU comprises a processing circuit and a memory.The memory contains instructions executable by the processing circuits, so the first RU is operational to receive, from several UEs, UL antenna signals on the N1 antennas, and to receive, from the BBU, a first part of BFW, and a first and second portion of the BFW, where the first portion is based on a determined UL channel estimate from the first RU, the first portion is determined to combine, in the frequency domain, the UL antenna signals received on the N1 antennas with user layer signals, the second portion is based on a channel estimate. 1689323 of 44 The first RU is also responsible for determining the BFW for itself based on the first and first portions of the received BFWs and sending the first and second portions of the received BFWs to the second RU via the RU link. This BFW then combines, in the frequency domain, the antenna signals received at the N2 antennas of the second RU with user layer signals. The first portion of the BFWs is based on the UL channel estimates determined for the first RU and the second RU, and is used to perform interference cancellation between the user layer signals.The first RU is also operational to receive, from the second RU, the second K user layer signals combined by the second RU from antenna signals received on the N2 antennas of the second RU from the UE number, the combination of the second RU is based on the first part and the second portion of the BFW sent to the second RU by the first RU, combines the first and second K user layer signals into K combined user layer signals, and sends the K combined user layer signals to the BBU. In accordance with other aspects, computer programs and media are also provided, the details of which will be described in the claims and the detailed description. Other potential features and benefits of this solution will become clear from the detailed description below. BRIEF DESCRIPTION OF THE DRAWINGS The solution will now be described in more detail by means of examples of embodiments and with reference to the accompanying drawings, in which Figure 1 is a block diagram illustrating a point-to-point topology of a distributed base station system comprising a plurality of RUs connected to a BBU. Figure 2 is a block diagram illustrating a cascade topology of a distributed base station system comprising a plurality of RUs connected to a BBU. Figure 3 is a flowchart illustrating a method performed by a BBU system for DL ​​communication, according to possible implementation forms. Figure 4 is a flowchart illustrating a method performed by a 1689323 of 44 first RU for DL ​​communication, according to possible forms of realization. Figure 5 is a flowchart illustrating a method performed by a BBU system for UL communication, according to possible implementation forms. Figure 6 is a flowchart illustrating a method performed by a first RU for UL communication, according to possible implementation forms. Figure 7 is a more detailed block diagram of a possible embodiment of a distributed base station system for DL ​​communication according to the present invention. Figure 8 is a schematic block diagram illustrating a BBU system in more detail, according to other possible embodiments. Figure 9 is a schematic block diagram illustrating a first RU in more detail, according to other possible embodiments. DETAILED DESCRIPTION Figure 2 illustrates a wireless communication network in which the present invention can be used. The wireless communication network comprises a distributed base station system 100, which in turn comprises a BBU 110 and a first RU 120. The BBU 110 has connections to other base station nodes or other RAN nodes and also to a core network (symbolized by 150 in Figure 2) so that the distributed base station system 100 can communicate with other nodes in the communication network. The BBU 110 connects to the first RU 120 via a fronthaul link 140. The fronthaul link 140 can be any type of connection, such as a dedicated wired or wireless connection or a connection through a network, provided that the connection meets the fronthaul requirements, for example, in capacity and latency.The first RU 120 further comprises a plurality of antennas 121, 122, and 123 through which wireless signals are communicated to and from one or more UE 131, 132, and 133. The wireless signals comprise data to be communicated to or from UE 131, 132, and 133. The distributed base station system 100 further comprises a second RU 160, which is connected to the first RU 120 via link RU 165. Note that the second RU 160 does not have a direct connection to BBU 110 but is connected to the BBU via link RU 165, the first RU 120, and fronthaul link 140. The second RU 160 further comprises a plurality of antennas 161, 162, and 163 through which wireless signals are communicated to and from one or more UE 131, 132, 133. The distributed base station system 100 may further comprise a third RU 170 that is connected to the second RU 160 via a second link RU 175. Note that the third RU 170 has no connection. 1689323 of 44 directly to BBU 110 but is connected to the BBU via the second RU link 175, the second RU 160, the RU link 165, the first RU 120 and the fronthaul link 140. The third RU 170 further has a plurality of antennas 171, 172, 173 through which wireless signals are communicated to and from one or more UE 131, 132, 133. The distributed base station system 100 may comprise other RUs cascaded to the third RU 170 in a similar manner. BBU 110, the first RU 120, the second RU 160, the third RU 170, and any other possible RUs each comprise RAN functionality for handling the data and signals that will be communicated between BBU 110, RUs 120, 160, and 170, and UEs 131, 132, and 133. RAN functionality is distributed between BBU 110 and the RUs as described later in this description. It can be observed that in 3GPP, the BBU can be further divided into two units called the Distributed Unit (DU) and the Central Unit (CU), where the DU is arranged to perform lower-layer processing, e.g., L1 and L2 of the BBU, and the CU is arranged to perform upper-layer processing of the BBU, e.g., L3 and above. A wireless communication network can be any type of wireless communication network that can provide radio access to wireless devices. Examples of such wireless communication networks include networks based on the Global System for Mobile Communications (GSM), Enhanced Data Rates for GSM Evolution (EDGE), Universal Mobile Telecommunications System (UMTS), Code Division Multiple Access 2000 (CDMA 2000), Long Term Evolution (LTE), LTE Advanced, Wireless Local Area Networks (WLANs), Worldwide Interoperability for Microwave Access (WiMAX), WiMAX Advanced, as well as fifth-generation (5G) wireless communication networks based on technologies such as New Radio (NR) and any potential future sixth-generation (6G) wireless communication networks. UE 131, 132, and 133 can be any type of communication device capable of wirelessly communicating with RU 120, 160, and 170 using radio signals. For example, UEs can be machine-type UEs or UEs capable of machine-to-machine (M2M) communication, sensors, tablets, mobile terminals, smartphones, laptops with embedded equipment (LEE), laptop-mounted equipment (LME), USB drives, customer premises equipment (CPE), and so on. Figure 3, together with Figure 2, describes a method implemented by a BBU system of a wireless communication network, where the communication network 1689323 of 44 wireless comprises a distributed base station system 100. The distributed base station system 100 comprises a BBU 110, a first RU 120 connected to the BBU 110 over a fronthaul link 140, wherein the first RU comprises N1 antennas 121, 122, 123, and a second RU 160 connected to the first RU 120 via a RU link 165, wherein the second RU 160 comprises N2 antennas 161, 162, 163. The method comprises determining 204 a first BFW portion in the frequency domain, based on an estimate of the DL channel of the first RU 120, wherein the BFWs will be used for beamforming of K user-layer DL signals to be sent to various UEs 131, 132, 133, where the The first portion of the BFW is determined to expand, in the frequency domain, the K user layer signals to the antenna signals of antennas N1 121, 122, 123 of the first RU 120.The method further comprises determining 206 a second portion of the BFW in the frequency domain, based on an estimate of the DL channel of the second RU 160, where the second portion of the BFW is determined to expand, in the frequency domain, the K user layer signals to the antenna signals of antennas N2 161, 162, 163 of the second RU 160. The method further comprises determining 212 a first part of the BFW in the frequency domain, based on the estimates of the DL channel of the first RU 120 and the second RU 160, where the first part of the BFW is determined 212 to perform interference cancellation between the user layer signals. In addition, the method comprises activating the 214 transmission of the first part of the BFW and the first and second portion of the BFW to the first RU 120, and activating the 216 transmission of the K user layer signals in the frequency domain to the first RU 120. The RU 165 link is a communication link or connection between two RUs, where only the first RU has a direct link to the BBU. The RU link can also be called an RU-RU link or a RU-RU fronthaul link. The DL channel estimate of the first RU is an estimate of the communication channel between UEs 131, 132, 133 and the first RU 120, and the DL channel estimate of the second RU is an estimate of the communication channel between UEs 131, 132, 133 and the second RU 160. Even if one or more of the UEs seen from the first RU are not seen from the second RU, or vice versa, UEs 131, 132, and 133 can all be considered connected to their respective RUs, but in that case, the channel estimates of the unseen UEs of a RU are set to zero. The respective DL channel estimate is determined based on channel measurements performed by the UE on the actual DL signals sent by the respective RU or, more commonly, the BBU determines an estimate 1689323 of 44 of the uplink (UL) channel from the UL signals sent from the UE to the respective RU and uses the known reciprocity of UL and DL. As claimed, the first part of the BFW is determined in 212 to perform interference mitigation / cancellation between the user layer signals in the beamforming at the first and second RUs, and the first and second portions of the BFW, which together may be called a second part of the BFW, are determined in 204, 206 to expand the user layer signals to antenna signals of the respective first and second RUs. The beamforming weights (BFW) may also be called precoding coefficients, and the beamforming may also be called precoding. Beamforming signals (BFWs) can be defined as beam-domain BFWs or antenna-domain BFWs, both within the frequency domain. The antenna or beam-domain BFWs are located at the top of the frequency domain; that is, antenna or beam-domain BFWs are applied to user-layer signals in the frequency domain. If the BBU system determines beam-domain BFWs, it can send these to the RU, which then transforms them back into antenna-domain BFWs and applies them to the user-layer signals on one or more subcarriers to the N1 antennas.Alternatively, the RU receives the beam-domain BFWs from the BBU system and applies the beam-domain BFWs to the user layer signals, then transforms the beam-formed beam-domain signals back into antenna-domain signals on the subcarrier(s) to the N1 antennas. Alternatively, the BBU system determines the beam-domain BFWs, transforms the determined beam-domain BFWs into antenna-domain BFWs, and then sends the antenna-domain BFWs to the RU. The RU applies the antenna-domain BFWs to the user layer signals on one or more subcarriers to the N1 antennas. When a cascaded topology is used to connect the RUs to the BBU, as previously described, there will be a high load on the fronthaul link 140 between the BBU 110 and the first RU 120 in the cascade, since the BFWs for all the RUs (in this case, the first RU 120 and the second RU 160), as well as the user data from all the RUs, must be sent across the fronthaul link 140 to the first RU 120. According to the invention, the RU BFWs are decomposed into a first part that is common to the cascaded RUs and a second part that is determined separately for each of the cascaded RUs. The first part is sent only once. 1689323 of 44 times from BBU 110 to the first RU 120. The first RU 120 copies the common first part of the BFWs and sends it to the subsequent RU 160. On the other hand, the second part with its RU-specific portions is sent across the fronthaul link 140 to the first RU 120 so that the first RU 120 can receive its own portion, called the first portion, and send the other portions of the second part to the subsequent RU 160. By such a separation of the BFWs into a first part of BFW that is common to both the first RU 120 and the second RU 160 (as well as any subsequent cascaded RUs), the first part of BFW only needs to be sent once across the fronthaul link. As a result, less data is sent via the fronthaul link compared to sending separate BFWs for each UK, compared to document WO2020 / 256609. The BBU system in the wireless communication network that implements the method can be the BBU 110, a unit within the BBU, or a distributed base station system. Alternatively, the BBU system implementing the method can be located in any other node of the communication network, such as a node further away from the UEs, for example, another network element in or near the RAN, or another RAN node.In this alternative, and in the cloud-based implementation described below, BBU 110 receives uplink signals from the first and second RUs 120 and 160, respectively. These signals, such as reference signals like SRS or DMRS, are received by the RUs from the UEs, or information related to these uplink signals. BBU 110 then communicates these uplink signals / information to the other network node. This node determines the DL channel estimate for each RU and, based on this estimate, calculates the first and second parts of the BFW, as shown above. The other network node then sends the determined first and second parts of the BFW back to BBU 110 for subsequent distribution to the first RU, 120.Alternatively, the wireless communication network system that implements the method can be a group of network nodes, where the functionality to perform the method is distributed across different physical or virtual nodes in the network. This latter approach can be called a “cloud solution.” According to one embodiment, the method further comprises compressing the first and second portions of the BFWs to a greater degree than any compression of the first portion of the BFWs. The compression is performed before the first and second portions of the BFWs are sent over the fronthaul link 140 to the first RU 120. The first part of the beamforming for the 1689323 of 44 Interference mitigation or cancellation has proven to be quite sensitive to the accuracy of the BFW, while the first and second portions, also called the second part of the BFW for expanding user layer signals to antenna signals, are comparatively robust. In other words, the second part of the BFW is less sensitive to errors, for example, due to lossy compression, compared to the first part of the BFW. Based on this understanding, the second part of the BFW (i.e., the first and second portions) is more heavily compressed before being sent to the first RU than the first part of the BFW to save fronthaul link capacity without any noticeable loss of beamforming accuracy due to its robustness. Therefore, even less data needs to be sent across the fronthaul link 140 compared to the previous implementation, without a significant performance loss.The fact that the first and second portions of the BFW are compressed to a greater degree than any compression of the first portion of the BFW means that there is stronger compression of the first and second portions of the BFW than of the first portion. The term "compression to a greater degree" is intended to encompass both "compression with fewer bits" and "compression by selecting only a subset of the BFW." "Compression to a greater degree" also means "more lossy compression." According to another embodiment, the first part of the BFW is determined based on the DL channel estimate of the first RU, the DL channel estimate of the second RU, and also on the first and second compressed portions of the BFW. This achieves an even better determination of the first part of the BFW, as it can mitigate interference caused by any inaccuracies due to the compression of the first and second portions. At the same time, the amount of data sent over the fronthaul link is kept to a low level. According to another embodiment, the first and second portions of the BFWs are compressed by selecting only a subset of the first and second portions, where the subset is a number of BFWs with a higher magnitude for each user layer signal. Furthermore, triggering the transmission of the compressed first and second portions of the BFWs to the first RU involves sending the selected subset of the first and second portions of the BFWs to the first RU. The selected subset can be sent either by sending only the values ​​of the selected subset or as a bitmask indicating the selected entries in the original array of the second portion of the BFWs. 1689323 of 44 The first and second portions are combined into a matrix; the bitmask has a first and second portion arranged similarly. For example, the bitmask is assigned to an antenna or beam branch / element for different user layers. For each user layer and antennas / beams N1 / N2, the bitmask is N1 / N2 bits per layer, where “1” represents the selected beam / antenna and “0” represents the unselected beam / antenna. The BFW number with the highest magnitude may correspond to one or more of the strongest channel components or coefficients. Often, the energy of a user layer is more concentrated in some channel coefficients than in others. For example, and especially when the channel estimation is in the beam domain, such compression would be quite effective. A channel coefficient is a complex number that represents the channel attenuation and phase shift.In the antenna element domain, it refers to the channel attenuation and phase shift from a transmitting antenna to a receiving antenna. In the beam domain, it refers to the channel attenuation and phase shift of a transmitting beam, i.e., a beam transmitted to a receiving antenna. Beams are typically predefined by a series of directional beams pointing in different directions across the covered angles. These beams can be obtained using the basis functions of the Discrete Fourier Transform (DFT). According to an alternative to the previous embodiment, the first and second portions of the BFW are compressed using 208 bits, employing fewer bits to represent individual BFWs than the number of bits used for the first and second portions before 208 compression. This compression method is feasible because the first and second portions of the BFW are not as error-sensitive as the first portion. Fewer bits for a beamforming weight can be achieved by using fewer quantization levels to describe a beamforming weight sample. Furthermore, the first and second portions of the BFW can be represented by floating-point blocks or block scaling, which have proven to be efficient quantization schemes using fewer bits. According to one embodiment, the method further comprises sending 209 to the first RU 120, first compression information concerning the compression 208 of the first portion of the BFW and second compression information concerning the compression 208 of the second portion of the BFW, wherein the first and second compression information are for the decompression in the first and second RU, respectively. This compression information contains details about how the compression was performed so that the first RU and the second RU can 1689323 of 44 decompress / reconstruct the first and second compressed portions of BFW, respectively. The compression information can be the bitmask mentioned earlier, which indicates which bits were sent when only a subset of the first and second portions (204, 206) of the specified BFWs was selected. When the first and second portions of the BFWs were compressed using fewer bits, the compression information may or may not need to be sent by the FH. According to another embodiment, the method further comprises combining the first and second portions of the BFW into a second part of the BFW. Compression can be performed individually for the first portion (i.e., the first RU) and the second portion (i.e., the second RU), respectively, of the second part of the BFW, i.e., before the combination, or it can be performed after the first and second portions have been combined. “Combining” here refers to placing the first and second portions of the BFW into an array that is larger than each of the arrays describing each of the first and second portions of the second part of the BFW. The larger array represents the complete second part of the BFW. The larger array formulation simplifies the calculation of the first part of the BFW when it is based on the first and second portions. According to another embodiment, the method further comprises determining 202 the DL channel estimate of the first RU 120 based on the UL signals originating from UE numbers 131, 132, 133 and received at antennas 121, 122, 123 of the first RU 120, and determining 203 the DL channel estimate of the second RU 160 based on UL signals originating from UE numbers 131, 132, 133 and received at antennas 161, 162, 163 of the second RU 160. For wireless communication networks based on reciprocal over-the-air technology, such as Time-Division Duplex (TDD) technologies, the DL channel estimate can be determined quite accurately from the UL signals. When the channel estimate is accurate, for example, when SINR is high, this approach works even better than a codebook-based approach.In TDD, for example, any difference between the UL and DL channels is primarily due to the radio interface difference between them, which can be calibrated. The UL signals can be reference signals originating from the UE number, such as the Sound Reference Signal (SRS) and Demodulation Reference Signal (DMRS), etc. These reference signals can be the same as those used by the UEs or different signals. They can be transmitted simultaneously or separately. 1689323 of 44 moments. According to another embodiment, the method further comprises normalizing the first part and the first and second portions of the BFW, and sending, to the first RU 120, information about a scale factor per user layer related to normalization. According to another embodiment, the distributed base station system 100 further comprises a third RU 170 connected to the second RU 160 via a second link RU 175, wherein the third RU 170 comprises antennas N3 171, 172, and 173. Furthermore, the method comprises determining 207 a third portion of the BFW in the frequency domain based on a determined DL channel estimate of the third RU 170, wherein the third portion of the BFW is determined to expand, in the frequency domain, the K user layer signals to signals from antennas N3 171, 172, and 173 of the third RU 170. Furthermore, the activation transmission 214 of the first portion of the BFW and the first and second portions of the BFW to the first RU 120 further comprises the activation transmission of the third portion of the BFW to the first RU 120. Figure 4, together with Figure 2, describes a method performed by a first RU 120 of a distributed base station system 100, wherein the first RU 120 comprises N1 antennas 121, 122, 123. The distributed base station system 100 further comprises a BBU 110 connected to the first RU 120 over a fronthaul link 140 and a second RU 160 connected to the first RU 120 via a link RU 165, wherein the second RU comprises N2 antennas. The method comprises receiving 302, from BBU 110, K user layer DL signals in the frequency domain, to be sent to several UE 131, 132, 133. The method further comprises receiving 304, from BBU 110, a first part of BFW, and a first and second portion of BFW for beamforming the K user layer signals in the frequency domain.The first portion is based on a determined DL channel estimate from the first RU 120, and the first portion is determined to expand, in the frequency domain, the user layer signals to the antenna signals of antennas N1 121, 122, 123 of the first RU 120. The second portion is based on a determined DL channel estimate from the second RU 160, and the second portion is determined to expand, in the frequency domain, the user layer signals to the antenna signals of antennas N2 161, 162, 163 of the second RU 160. The first part of the BFW is based on the determined DL channel estimates from the first RU 120 and the second RU 160, and the first part of the BFW is determined. 1689323 of 44 to perform interference cancellation between user layer signals. The method further comprises determining 310 BFW for the first RU based on the first received part and the first portion of the BFW, and sending 312, to the second RU 160 via link RU 165, the first and second received portions of the BFW, and sending 314, to the second RU 160 via link RU 165, the K received user layer DL signals in the frequency domain. The method further comprises beamforming 316 the K received user layer signals into antenna signals, using the BFW determined for the first RU 160, and sending 318 the antenna signals to UE numbers 131, 132, 133 via antennas N1 121, 122, 123. Using this method, the first RU can receive and use BFW tailored to the first RU and send BFW tailored for subsequent cascaded RUs (i.e., RU2) to the subsequent cascaded RUs. Because the first part of the BFW is only sent once across the fronthaul link, even though it is used by all RUs cascaded to the BBU, the FH link capacity is saved. The 310 determination of BFW for the first RU is typically performed by multiplying the first part of the BFW by the first portion of the BFW. When the first part and the first portion of the BFW are arrays, the multiplication is an array multiplication. The first and second RUs can be implemented as separate RUs on different printed circuit boards (PCBs). Alternatively, the first and second RUs can be implemented as separate radio processors on a single PCB. According to one embodiment, the method further comprises extracting the first portion and the first part of the BFW from the first received portion, as well as the first and second portions of the BFW. The extraction can be performed such that when the first and second portions are combined into a common array, the first RU extracts the portion of the array belonging to the first portion. For example, the common array is a four-column array where the first two columns belong to the first portion and the last two columns belong to the second portion of the BFW. The first RU then uses or extracts the first two columns and sends the remaining two columns to the second RU. Alternatively, the extraction can be performed such that if the first and second portions are sent in separate messages, the first RU takes the message containing the first portion of the BFW and sends the message containing the second portion to the second RU.Thus, it is possible that the first RU only uses the BFWs determined for it and sends, or forwards, the second portion that it has not used, and the first part that both the first RU and the first RU use. 1689323 of 44 as the second, to the second UK. According to one embodiment, the method further comprises decompressing the first portion of the BFWs when the first portion of the BFWs has been compressed in the BBU. This step is performed when the BBU has compressed the first portion of the BFWs to a higher degree than any compression applied to the first part of the BFWs. The decompression can be either dequantization or reconstruction by filling in zeros in an array according to a received bitmask (see below). In the case of dequantization, the first RU requires information about the format used, such as the compression scheme and configuration parameters. The format used is usually preconfigured so that the BBU and the first RU are aware of it beforehand. According to another embodiment, the method further comprises receiving from BBU 110, first compression information concerning the compression of the first portion of the BFW and second compression information concerning the compression of the second portion of the BFW. The first and second portions of the BFW are compressed by the BBU to a higher degree than any compression of the first portion of the BFW, wherein the decompression of the first portion of the BFW is performed according to the first compression information. The method further comprises sending the second compression information to the second RU 160 via the RU 165 link. The first and second compression information may be a bitmask.This is applicable when the first and second portions of the BFWs were compressed in the BBU, for example, by selecting only a subset of a certain first and second portion of the BFWs, where the subset is a number of stronger BFWs for each user layer signal. According to another embodiment, in which the first received part 304 and the first and second portions of the BFW are normalized, the method further comprises receiving 306, from the BBU, information on a per-user-layer scaling factor relating to the normalization and scaling 307 of the first received normalized part and first portion of the BFW in accordance with the information on the scaling factors prior to beamforming 316. The first received part and the first and second portions of the BFW are normalized by the BBU in accordance with a per-user-layer scaling factor that may be different from the scaling factor used by the first RU for scaling 307. According to one alternative, the method also involves sending the scaling factor information per user layer to the second RU 160. 1689323 of 44 mode, also the second RU can scale the first part and the second portion of the BFW. According to another embodiment, the distributed base station system 100 further comprises a third RU 170 connected to the second RU 160 via a second link RU 175, wherein the third RU 170 comprises antennas N3 171, 172, and 173. The BFW receiver 304 further comprises receiving a third portion of the BFW, determined based on a DL channel estimate for the third RU 170 to expand, in the frequency domain, the user layer signals to the antenna signals of antennas N3 171, 172, and 173 of the third RU 170, and the first portion of the BFW is based on DL channel estimates for the first RU 120, the second RU 160, and the third RU 170. The method further comprises sending, to the second RU 160 via link RU 165, the third portion of the BFW for its additional transmission to the third RU 170 via the second link RU 175. Figure 5, together with Figure 2, describes a method implemented by a BBU system of a wireless communication network, where the wireless communication network comprises a distributed base station system 100 comprising a BBU 110, a first RU 120 connected to the BBU 110 via a fronthaul link 140, where the first RU comprises antennas N1 121, 122, 123. The distributed base station system 100 further comprises a second RU 160 connected to the first RU 120 via a link RU 165, where the second RU 160 comprises antennas N2 161, 162, 163. The method comprises determining a first BFW portion in the frequency domain, based on an estimate of the UL channel of the first RU 120, wherein the first BFW portion will be used to combine, in the frequency domain, antenna signals received on the N1 antennas of the first RU 120 from several UE 131, 132, 133 to K user layer signals of the first RU.The method further comprises determining a second portion of the BFW in the frequency domain, based on an estimate of the UL channel of the second RU 160, wherein the second portion of the BFW will be used to combine the frequency domain signals received at the N2 antennas of the second RU 160 from UE numbers 131, 132, 133 to K user layer signals of the second RU. The method further comprises determining a first portion of the BFW in the frequency domain based on the UL channel estimates of the first RU 120 and the second RU 160, wherein the first portion of the BFW is determined to perform interference cancellation between the user layer signals of the first and second RUs, and to enable the transmission of the first portion of the BFW and the first and second portions. 1689323 of 44 the BFW to the first RU 120. This method in the BBU is similar to the method in the BBU of previous embodiments; however, this method is adapted to UL signal communication, whereas the previous embodiments were adapted to DL signal communication. The advantages, etc., are similar to those of DL signal communication. Furthermore, the embodiments mentioned earlier in this description in relation to Figure 3 are also applicable to this UL signal communication scenario. Figure 6, together with Figure 2, describes a method performed by a first RU 120 of a distributed base station system 100, wherein the first RU 120 comprises N1 antennas 121, 122, 123. The distributed base station system 100 further comprises a BBU 110 connected to the first RU 120 via a fronthaul link 140 and a second RU 160 connected to the first RU 120 via a link RU 165, wherein the second RU comprises N2 antennas 161, 162, 163. The method comprises receiving 452, from various UE 131, 132, 133, UL antenna signals at the N1 antennas, and receiving 454, from BBU 110, a first part of BFW and a first and second portion of BFW, wherein the first portion is based on a channel estimate. UL determined from the first RU 120, and where the first portion is determined to combine, in the frequency domain,the UL antenna signals received at antennas N1 with user layer signals; and wherein the second portion is based on a determined UL channel estimate from the second RU 160 and the second portion is determined to combine, in the frequency domain, antenna signals received at antennas N2 from the second RU 160 to user layer signals, wherein the first part of the BFWs is based on the determined UL channel estimates from the first RU 120 and the second RU 160, and wherein the first part of the BFWs is determined to perform interference cancellation between the user layer signals. The method further comprises determining 456 BFWs for the first RU based on the first part and the first received portion of the BFWs, sending 458 to the second RU 160, via link RU 165, the first part and the second received portion of the BFWs, and combining 460, in the frequency domain,the UL antenna signals received at the N1 antennas in the first K user layer signals, using the BFW determined for the first RU 120. The method further comprises receiving 462, from the second RU 160, the second K user layer signals combined by the second RU from antenna signals received at the N2 antennas of the second RU 160 from UE numbers 131, 132, 133, wherein the combination of the second RU is based on, 1689323 of 44 the first part sent and the second portion of the BFW sent to the second RU by the first RU. The method further comprises combining 464 the first and second K user layer signals into K combined user layer signals, and sending 466 the K combined user layer signals to BBU 120. This method, implemented by the first RU 120, is adapted to UL signal communication, while the implementations described in relation to Figure 4 were adapted to DL signal communication. The combination of the antenna signal beamforming weights with the user layer signals can be achieved by pre-equalizing the antenna signals, which effectively performs the combination. This pre-equalization / combination can achieve a coherent combination, such as a maximum ratio combination. Furthermore, by combining the first and second K user layer signals, already combined or pre-equalized into K combined user layer signals, only one set of combined signals is sent through the base station system via the fronthaul link 140 to BBU 110 for all RUs that are cascaded together.In this way, the use of the fronthaul connection capacity can be kept low. Furthermore, the embodiments mentioned earlier in this description in relation to Figure 4 are also applicable to this UL signal communication scenario. The following describes a system model for reciprocal DL-based beamforming according to one embodiment. For DL ​​in 3GPP Option 7-3 and Cat-B O-RU in O-RAN architecture (where beamforming is performed in the RU), channel estimation and BFW calculation are performed in the BBU system, while beamforming is executed in the RU. In the case of cascaded RUs, which is the subject of this invention, the calculated BFWs need to be transported from the BBU system to the first RU via the fronthaul link. The first RU extracts its corresponding BFWs and then forwards the remaining BFWs to the next RU, i.e., the second RU. This process is repeated at each RU in the cascade chain until the BFWs reach the last RU. This is typically done in the control plane of the fronthaul interface / link.Meanwhile, multi-layer user-layer data, such as the encoded user-layer bits in 3GPP option 7-3 or the O-RAN-compressed modulation symbols, is typically transported in the data plane from the BBU to each RU in the same way—that is, cascading through the first RU. Each RU in the cascade converts the received user-plane data into modulated symbols and applies the received BFW. 1689323 of 44 directed to the determined RU in the modulated symbols of the multiple layers and then converts the beam-shaped symbols into analog signals to transmit wirelessly to the different antennas towards the UE. Now consider the scenario with K user layers (i.e., UEs to receive user data) in a desired cell. There are L RUs in the local area cascaded in a daisy chain, effectively forming a large antenna array. The first RU (RU1) is the one with a fronthaul interface connecting to the BBU. For Z = 1, RU1 perceives the DL channel to the K UEs as H1 e^(-KxNi), where Ni is the number of antennas equipped in RU1. The effective large antenna network therefore has Σι = ιM antenna elements in total. The corresponding effective DL channel estimate comprising L RUs will be H = [H — HJ e tfxÜ=ii Note that reciprocity-assisted transmission (RAT) is performed with respect to the effective antenna network. The base force readouts (BFWs) for a zero-forcing (ZF)-based method can be expressed as PZF= H (HH*)-1*-1(1) while the BFW for a method based on the minimum mean square error (MMSE) can be expressed as, P mmse = H* (HH* +σ^I)-! P2 Pt (2) where σ2 is a measure of interference and noise (or error terms) and I is a K x K identity matrix. Then, the BFWs are divided into a first part Pi and a second part P2 and the BFWs are generally written as P = P2P1, where the first part Pi is a common part for all L RUs, and the second part P2 = H* = [(H*)r- (H*) ] contains the respective coefficient part specific to each of the L RUs. The first part Pi, which performs the pre-mitigation of interference between user layers, is composed of K x K complex values. The second part p2 = [Pj,RU1 — P{RU1 —p2^RU1]T contains different coefficients for each RU, where the coefficient P2,RU1 = H* expands signals from different user layers previously combined by 1 to transmit signals on antenna elements or beam directions in RU-Z. Consequently, the BFW for RU-Z can be written as PRU l = P2RU (pi.Therefore, the RU-Z transmitted signal after beam formation can be expressed as. Yί =PRU íx=P2,RU íP1x 1689323 of 44 where x = [x1 x2. xK]T denotes the signal vector where each element xt represents the modulated symbols of layer iyy(= [y^ yÍ2-yiN] denotes the transmitted signal vector of RU-Z, where each element yt represents the transmitted symbols of antenna io beam i of RU-Z. Then, according to one embodiment, the DL BFWs are compressed. The BFWs at P consist of (ΣΪ=1^) x ^ complex values. If each complex value is represented by Q bits, transporting P directly requires (ΣΪ^NI) x ^ x Q bits on the fronthaul link. This invention transports P1 and P2 respectively from the BBU to the first RU. Two different embodiments are provided for compressing P2 at the BBU so that the total number of bits required to transport P1 and P2 can be significantly reduced compared to transporting P directly. Figure 7 shows a block diagram of a distributed base station having a BBU 510 and RUs 520, 560, and 570 cascaded to the BBU 510, according to one embodiment with respect to a 3GPP architecture option 7-3. Details relating to the operation of the BBU 510 and the cascaded RUs 520, 560, and 570, respectively, will be described below. BBU 510 comprises a forward error correction (FEC) encoding unit for encoding user layer signals to be transported forward (DL) across the fronthaul link 540 to RUs 520, 560, and 570. BBU 510 further comprises a BFW calculation unit 514 for determining or calculating the BFWs for each RU. The BFW calculation unit 514 may alternatively be located in a BBU system that is located elsewhere than within the BBU, such as when deployed as a cloud solution. One implementation for compressing the second part of the BFW2 in the BBU is a so-called layer-specific BFW selection. In this implementation, before calculating the first and second parts of the BFW, i.e., P1 and P2, the channel coefficients can be transformed into the beam / direction domain if the channel is not estimated in the beam / direction domain. Then, the second part of the BFW2 is determined as separate parts for each RU: P2 = [P^RU 1 *” P^RU l P^RU Ϊ]whereP2.RU l = Hj The second compressed part of BFW P2 is determined as follows: P2 is obtained by keeping M selected entries from each column of P2 and setting the unselected entries to zero. The M selected entries from each 1689323 of the 44 columns of P2 are the inputs with the greatest amplitude (absolute value), which corresponds to the strongest channel coefficients for this layer. Indeed, for the compressed BFWs of each RU, P2ζκυι, there are corresponding non-zero inputs in each column, and the rest of the inputs are zeros, where 0 <M<M y Μ = Σϊ=ιΜι. Elp2comprimido se puede expresar comoP2 =[P2,RU 1 *”P2,RU í *”Ρ2,RUϊ]. The first common part of the BFW P1 is then calculated according to p2 based on Eqs. (1) or (2), depending on whether the ZF method or the MMSE method is used. For the ZF-based method, P1 = (HP2)1. For the MMSE-based method, P1 = (Hp2 + σ2I) 1. Note that the ZF-based method and the MMSE-based method are only example methods; any other possible method may be used. From then on, the first part of BFW P1 and the non-zero values ​​of P2, along with a bitmask indicating the selected inputs in p2, are sent to the first RU 520. The bitmask size for all RUs is (Σϊ=1 A¡)xK bits, which can be shared among multiple resource blocks (RBs), and therefore the overhead can be small. Here, without considering the bitmask overhead, the number of bits required to carry the BFW is (M + K)KQ. To ensure a well-conditioned matrix inversion for the calculation of p1, it is suggested that M>K. For good performance, the selected inputs should capture most of the channel energy from the DL H channel estimate. As an example, if each of the L RUs has A antennas and each selects K of the A beams, the number of BFWs carried is (L + 1)K². For the uncompressed case, the number of BFWs is LAK. Therefore, the compression ratio is equal to (ϊ + NK). When L is large, it will approach K / A. For L = 4, A = 32, and K = 16, compression reduces the BFWs to 62.5% of the uncompressed case. If K = 8, it is reduced even further to 31.25%. Therefore, it can significantly reduce the link capacity required to carry BFWs. Regarding performance, our studies have shown that K-beam selection for a K-layer system performs quite close to the full system without beam selection. A second embodiment for compressing the second part of the P2 BFWs in the BBU is a P2 compression using fewer bits. Originally, when transporting the P BFWs, the BFWs had to be quantized with high precision, since large quantization errors would degrade the interference cancellation performance between layers. In this invention, for the same reason, the P1 BFWs are also 28 1689323 of 44 should be quantized with the same level of precision as the BFWs of P. However, P2 can allow some loss of precision with fewer bits in the quantization. Therefore, transporting P2 across the FH link with fewer bits can also reduce the required FH capacity. Note that this implementation does not require the BFWs to be calculated based on beam / direction domain channels. The second part of the BFWs can be determined as P2 = [PIrU1 ·” P^RUi *“ P^RUL] where P2,RU^ = H¡, that is, in the same way as in the first embodiment. Then, the second part of the BFWs is compressed with fewer bits per BFW, the result of which is indicated as p2. This embodiment can be implemented in addition to the first embodiment, that is, to represent non-zero weights with fewer bits. The compressed BFW can be expressed as ~ r~T CST 1TP2 = [P2,RU 1 ·”P2,RU l *”P2,RU L]. The first common part of BFW P1 is then calculated according to P2 based on Eqs. (1) or (2), depending on whether the ZF method or the MMSE method is used. For the ZF-based method, P1 = (HP2)1. For the MMSE-based method, P1 = (Hp2 + σ2I) 1. Note that the ZF-based and MMSE-based methods are only example methods; any other possible method may be used. The first part of BFW P1 and the second compressed part of BFW P2 are then sent to the first RU 520. Returning to Figure 7, we will now examine what happens in each of RUs 520, 560, and 570 when the FEC-encoded user layer signals and the first and second calculated BFW portions are received from BBU 510. The first RU, 520, comprises a modulation unit 522 that modulates the FEC-encoded user layer signals received via link FH 540. The first RU, 520, further comprises an extraction unit 524 that extracts the portion of the second BFW portion received via link FH 540 that is specific to the first RU. This portion is referred to as the first BFW portion and is denoted as Pen in Figure 7. The extraction unit 524 then sends the remaining second BFW portion, that is, without the first portion, to the second RU, 560. The first RU also sends the first BFW portion. to the second RU, after copying it.The first RU also sends the FEC-encoded user layer signals received from the BBU to the second RU 560. The first RU 520 then calculates the BFW by itself as RU 1= 1689323 of 44 P2,ruiPi... that is, based on the first part of the BFWs and the first compressed portion of the BFWs. Furthermore, the first RU 520 comprises a beamforming unit 526 that beamforms the modulated user layer signals x with the BFWs of the first RU PRU1 to create antenna signals y1 = PRU1x. The first RU 520 further has Inverse Fast Fourier Transform (IFFT) and related RF circuitry 528 that performs IFFT on the beamformed user layer signal in the frequency domain so that the signal is transformed into the time domain. The time-domain signal is then converted from a digital to an analog signal by a digital-to-analog converter, and the frequency is then converted to an RF signal before being wirelessly transmitted from the antennas N1 of the first RU to the UE 531, 532. Furthermore, for the first embodiment of compression of the second part of the BFW P2 performed in the BBU, the first RU 520 receives the first part of the BFW P1 and non-zero weights of the second part P2 along with the bitmask of P2 from the BBU. In this embodiment, the extraction unit 524 not only extracts the first portion of the second part of the BFW but also extracts the bitmask for the first portion of the BFW. Similarly, the extraction unit 524 not only sends the remaining portions of the second part of the BFW to the second RU 560 but also sends the remaining bitmask for the remaining portions of the second part of the BFW. Additionally, the first portion of the BFW is reconstructed by filling in zeros in the array using the extracted bitmask for the first portion of the BFW. For the second form of realization, the first portion of the BFW is reconstructed by dequantization. Note that in 3GPP option 7-3, the encoded bits are received by the first RU 520 of BBU 510 via the fronthaul interface along with the BFWs. The modulated x symbols are modulated onto the received encoded bits. In O-RAN, the modulated x symbols may be compressed. In this case, the first RU decompresses the compressed symbols to obtain the modulated x symbols. The second RU 560 has units similar to the first RU, namely a modulation unit 562, a extraction unit 564, a beamforming unit 566, IFFT and RF-related circuitry 568, and antennas N2. The modulation unit 562 modulates the FEC-encoded user layer signals received from the first RU. The second RU 560 receives the remaining portions of the second part of the BFW from the first RU 520; that is, the second part of the BFW without the first part. 1689323 of portion 44. Extraction unit 564 of the second RU extracts the portion of the second part of the BFW that is specific to the first RU; this portion is called the second portion of the BFW and is indicated as P2 RU2 in Figure 7. Extraction unit 564 also sends the remaining second part of the BFW P2, that is, without the first and second portions, to the remaining RUs, in this case, RU L 570. The second RU 560 also sends the first part of the BFW to RU L, after copying it. The second RU 560 also sends the FEC-encoded user layer signals received from the first RU 510 to RU L 570. The second RU 560 then calculates the BFWs itself as PRU2 = P2,RU2P1, that is, based on the first part of the BFWs and the second compressed portion of the BFWs. Additionally, the beamforming unit 566 beamforms the modulated user layer signals x with the BFWs from the second PRU2 to create antenna signals y2 = PRU2x. The IFFT and RF circuitry 568 performs IFFT on the antenna signals in the frequency domain so that the signal is transformed into the time domain and then wirelessly transmitted from the N2 antennas of the second RU to the UE 531 and 532. Furthermore, for the first embodiment of compression of the second part of the BFW2 performed in the BBU, the second RU 560 receives the first part of the BFW P1 and non-zero weights of the remaining portions of the second part P2, along with the remaining bitmask of P2, from the first RU. In this embodiment, the extraction unit 564 not only extracts the second portion of the second part of the BFW but also extracts the bitmask for the second part of the BFW. Similarly, the extraction unit 564 not only sends the remaining portions of the second part of the BFW to RU L 570 but also sends the remaining bitmask for the remaining portions of the second part of the BFW. Additionally, the second part of the BFW is reconstructed by filling in zeros in the array using the extracted bitmask for the second part of the BFW. For the second form of realization, the second portion of the BFW is reconstructed by dequantization. The RU L 570 has units similar to the first and second RUs, namely a modulation unit 572, a beamforming unit 576, IFFT and RF-related circuitry 578, and NL antennas, which serve a similar function to the first and second RUs. However, since the RU L is the last in the cascade, it does not require an extraction unit. Similarly, there is no need to extract any data and send other data, as the received data will be used in the last RU 31. 1689323 of 44 of the waterfall. According to another embodiment, there can be a scaling factor between the MIMO user layers. Once the BFWs have been calculated, the DL implementation may require adjusting the power over the number of jointly scheduled UEs or MIMO layers to allocate power to each UE or MIMO layer. Let D be a KxK diagonal matrix where each diagonal element represents a power allocation factor with respect to the corresponding column. The power allocation factor not only controls the power allocation between the MIMO layers but also normalizes the BFWs, reducing the BFW dynamic range and facilitating efficient quantization. Including the scaling matrix D, the transmit signal can be expressed as y = PDx According to previous embodiments, the BFWs consist of a first and a second part P = P2P1. The two parts of the BFWs at P1 and P2, sent respectively from BBU 510 to the first RU 520, must also be normalized so that the dynamic range of the BFWs is reduced when transported over the fronthaul link 540. To achieve the same transfer signal, and as previously stated, the normalized first part and the second part of the BFWs can be composed as ?! = D-1?^! ?2 = P2D2 where D1 is a KxK diagonal matrix where each diagonal element represents a scale factor with respect to each column of D-1Pi, and D2 is a KxK diagonal matrix where each diagonal element represents a scale factor with respect to each column of P2. To reconstruct y, the BBU 510 needs to additionally send to the first RU 520, the scale factors in D = D⁻¹D, where D is a diagonal matrix containing K real values. The first RU 520 constructs P1 = P1D and forwards it to the second RU 560 along with the remaining part of P2, and so on to the remaining RU, i.e., RU L, as described above. Upon receiving P1, P2, and D, the first RU can obtain the transmission signal from the effective antenna array as D2?'x = P2D2D-1P1D1D-1Dx = P2P1Dx = PDx = y This results in the transmission signal as originally intended. 1689323 of 44 A method similar to that described above for DL ​​transmission of user-layer signals can also be used for UL transmission of user-layer signals, given an architecture in which channel estimation and BFW calculation are performed in the BBU, while beamforming is performed in each of the cascaded RUs. Similar to DL, the BBU sends the BFWs to the cascaded RUs, through the first RU, then the second RU, and so on. Each RU receives the BFWs, extracts its own BFWs, and forwards the remaining BFWs to the next RU. Let Hule cÍSLi^xk be the effective UL channel comprising the L RU. If ZF-based beamforming is used, the BFW can be expressed as WZF = (HulHul)1HjjL ----------,--------' '------,-----W1 W2 If MMSE-based beamforming is used, the BFWs can be expressed as wMMSE = (Hu lHU L + ^21)^HUL 12 where σ2 is a measure of interference and noise (or error terms) and I is a K × K identity matrix. W1 denotes the first part of the BFWs for UL and W2 denotes the second part of the BFWs for UL, similarly to the BFWs for DL. Both W1 and W2 are calculated in the BBU. According to the invention, the second part of the BFWs W2 can be compressed as W2 using one of the embodiments described above for DL. Note that P2 e £(Σί, ^ήχκwhile W2 e EΚχΣ[=ι¡, therefore, the selection of the M entries from each column of P2 in the DL must be done as a selection of the M columns from each row of W2 in the UL. Then W1 is calculated based on W2.The BBU sends the compressed W2 in terms of the non-zero part of W2 plus the bitmask, for the first compression realization, or the complete W2 for the second compression realization, along with W1 through the fronthaul link to the first RU. In the first RU, the first portion of the second part of the BFWs, W2,RU1, is reconstructed based on the first portion extracted from the received BFWs and a bitmask when using the first compression realization. The first RU forwards the remaining portion of the second part of the BFW2s and the remaining portion of the bitmask (for the first compression realization) along with the first portion of the BFW1s to the second RU. The first RU calculates the BFWs for the first RU as WRU1 = W1W2,RU1 and performs beamforming accordingly. This procedure is repeated in each of the cascaded RUs until 33 1689323 of 44 BFWs arrive at RU L. To perform beamforming on the UL signals received at its antennas from the UEs, the first RU combines the UL antenna signals received at the N1 antennas into the first K user layer signals, using the BFWs calculated for the first RU. Then, the first RU receives, from the second RU, the second K user layer signals that the second RU has beamformed, i.e., combined in a similar way, which is combined by the second RU from antenna signals received at the N2 antennas of the second RU from the UEs. The combination of the second RU is based on the BFWs calculated by the second RU from the first part and a second portion of the second part of the BFWs, which the first RU forwards when it receives the BFWs from the BBU. The first RU then combines the first and second K user layer signals into K combined user layer signals and sends the K combined user layer signals to the BBU.In the event that there are more than two cascaded RUs, the subsequent RUs in the chain-beam antenna signals are received at their antennas in a similar manner, i.e., they combine into user layer signals using the first part of the BFW and some part of the second part of the BFW and send the combined user layer signals upwards in the cascade towards the BBU, to combine with the user layer signals of the RUs located closer to the BBU, on their way to the BBU. Except for their use in a distributed base station with separately located RUs, where each RU is arranged on a separate printed circuit board (PCB), the embodiments described above can also be used in a single large RU design with multiple radio processors, where each radio processor would assume the role of an RU in the previous embodiments. In this case, the radio processors / RUs can be arranged on the same PCB. The radio processors are then implemented in a cascade topology, i.e., as in Figure 2. This can significantly reduce the number of SerDes lanes on the PCB between the radio processors and a front-end interface that will connect to the BBU, compared to a star topology design. The overall capacity required of the fronthaul link is also significantly reduced. Figure 8, together with Figure 2, describes a BBU 600 system configured to operate in a wireless communication network. The wireless communication network comprises a distributed base station system 100 comprising a BBU 110, a first RU 120 connected to the BBU 110 via a fronthaul link 140, wherein the first RU comprises N1 antennas 121, 122, 123. The base station system 1689323 of 44 distributed further comprises a second RU 160 connected to the first RU 120 via a link RU 165, wherein the second RU 160 comprises antennas N2 161, 162, 163. The BBU 600 system also comprises a processing circuit 603 and a memory 604. Said memory contains instructions executable by said processing circuit, by which the BBU 600 system is operational to determine a first portion of BFW in the frequency domain, based on an estimate of the DL channel of the first RU 120, wherein the BFW will be used for the formation of beams of K user layer DL signals to be sent to various UE 131, 132, 133, wherein the first portion of the BFW is determined to expand, in the frequency domain, the K user layer signals to the antenna signals of antennas N1 121, 122, 123 of the first RU 120.The BBU 600 system is also operational for determining a second portion of the BFW in the frequency domain, based on an estimate of the DL channel of the second RU 160, where the second portion of the BFW is determined to expand, in the frequency domain, the K user layer signals to antenna signals of antennas N2 161, 162, 163 of the second RU 160 and for determining a first part of the BFW in the frequency domain, based on the estimates of the DL channel of the first RU 120 and the second RU 160, where the first part of the BFW is determined to perform interference cancellation between the user layer signals. The BBU 600 system is also operational for activating the transmission of the first part of the BFW and the first and second portion of the BFW to the first RU 120, and activating the transmission of the K user layer signals in the frequency domain to the first RU 120. The BBU 600 system can be the actual BBU 110, a unit in the BBU 110, or in the distributed base station system 100. Alternatively, the BBU 600 system can be deployed in any other node of the communication network, such as a node further away from the UEs, for example, another network element in or near the RAN, or another RAN node.In this alternative, and in the cloud-based implementation described below, BBU 110 is configured to receive uplink signals from the first and second RUs 120 and 160, respectively. These signals, such as reference signals like SRS or DMRS, are received by the RUs from the UEs, or information related to these uplink signals. BBU 110 then communicates these uplink signals / information to the other network node. This other node is designed to estimate the DL channel of each RU and, based on this estimate, determine the first and second parts of the BFW, as shown above. The other network node then forwards the first and second parts of the BFW. 1689323 of 44 second part determined from the BFW back to BBU 110 for subsequent distribution to the first RU 120. Alternatively, the BBU 600 system can be implemented as a group of network nodes, where the functionality of the BBU system is distributed across different physical or virtual nodes of the network. The latter can be called a “cloud solution”. According to one embodiment, the BBU 600 system is also operational for compressing the first and second portions of the BFW to a higher degree than any compression of the first part of the BFW. According to another embodiment, the BBU 600 system is operational for the determination of the first part of the BFW based on the estimation of the DL channel of the first RU 120, the estimation of the DL channel of the second RU 160, and on the compressed value of the first and second portion of the BFW. According to another embodiment, the BBU 600 system is operational for the compression of the first and second portions of the BFWs by selecting only a subset of the first and second portions of the BFWs, where the subset is a number of BFWs with greater magnitude for each user layer signal, and operational for activating the sending of the compressed first and second portions of the BFWs to the first RU 120 by sending the selected subset of the first and second portions of the BFWs to the first RU. According to another embodiment, the BBU 600 system is operational for compressing the first and second portions of the BFWs using fewer bits to represent individual BFWs than the number of bits used for the first and second portions before compression. According to yet another embodiment, the BBU 600 system is also operational for sending, to the first RU 120, a first compression information on the compression of the first portion of the BFW and the second compression information on the compression of the second portion of the BFW, where the first and second compression information are for decompression in the first and second RU, respectively. According to yet another embodiment, the BBU 600 system is also operational for combining the first portion and the second portion of the BFW into a second part of the BFW. According to yet another embodiment, the BBU 600 system is also operational for determining the DL channel estimate of the first RU 120 based on the UL signals originating from UE numbers 131, 132, 133 and received at 1689323 of 44 antennas 121, 122, 123 of the first RU 120, and determine the DL channel estimate of the second RU 160 based on the UL signals originating from UE number 131, 132, 133 and received on antennas 161, 162, 163 of the second RU 160. According to yet another embodiment, the BBU 600 system is also operational for normalizing the first part and the first and second portions of the BFW, and sending, to the first RU 120, information on a scale factor per user layer related to normalization. According to another embodiment, the distributed base station system 100 further comprises a third RU 170 connected to the second RU 160 via a second link RU 175, wherein the third RU 170 comprises antennas N3 171, 172, and 173. The BBU 600 system is further operative for determining a third portion of the BFW in the frequency domain based on a determined DL channel estimate from the third RU 170, wherein the third portion of the BFW is determined to expand, in the frequency domain, the K user layer signals to the antenna signals of antennas N3 171, 172, and 173 of the third RU 170. Furthermore, the BBU 600 system is operative for activating the transmission of the first portion of the BFW and the first and second portions of the BFW to the first RU 120 by also activating the transmission of the third portion from the BFW to the first RU 120. According to an alternative aspect, memory 604 of the BBU system in Figure 8 contains instructions executable by processing circuit 603, so the BBU system 600 is operational to determine a first portion of BFW in the frequency domain, based on an estimate of the UL channel of the first RU 120, where the first portion of BFW will be used to combine, in the frequency domain, antenna signals received at antennas N1 of the first RU 120 from several UE 131, 132, 133 with K user layer signals from the first RU, and to determine a second portion of BFW in the frequency domain, based on an estimate of the UL channel of the second RU 160, where the second portion of BFW will be used to combine signals in the frequency domain received at antennas N2 of the second RU 160 from UE 131, 132, 133 with K user layer signals of the second RU.The BBU 600 system is also operational for determining a first part of the BFW in the frequency domain based on the UL channel estimates of the first RU 120 and the second RU 160, where the first part of the BFW is determined to perform interference cancellation between the user layer signals in the first and second RU, and to activate the sending of the first part of the BFW and the first and second portion of the BFW to the first RU. 1689323 of 44 120. This alternative aspect handles the UL transmission of user layer signals. The embodiments mentioned above in relation to Figure 8 are also applicable to this aspect. According to other embodiments, the BBU 600 system may further comprise a communication unit 602, which may be considered to comprise conventional means for communication with other network nodes of the wireless communication network 100, such as the BBU, if the system is not the actual BBU, and with the first RU 120. The instructions executable by said processing circuit 603 may be arranged as a computer program 605 stored, for example, in said memory 604. The processing circuit 603 and the memory 604 may be arranged in a sub-arrangement 601. The sub-arrangement 601 may be a microprocessor and suitable software and storage, hence a programmable logic device, PLD, or other electronic components / processing circuits configured to perform the methods mentioned above.The 603 processing circuitry may comprise one or more programmable processors, application-specific integrated circuits, field-programmable gate assemblies, or combinations thereof adapted to execute instructions. The computer program 605 can be arranged so that when its instructions are executed in the processing circuit, they cause the BBU 600 system to perform the steps described in any of the embodiments of the BBU 600 system and its method described. The computer program 605 can be carried by a computer program product connectable to the processing circuit 603. The computer program product can be memory 604, or at least be arranged in memory. Memory 604 can be implemented as, for example, RAM (random access memory), ROM (read-only memory), or EEPROM (electrically erasable programmable ROM). In some embodiments, a storage medium can hold the computer program 605.The medium can be an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electrical signal, a radio signal, a microwave signal, or a computer-readable storage medium. The computer-readable storage medium can be, for example, a CD, DVD, or flash memory, from which the program could be downloaded to memory 604. Alternatively, the computer program can be stored on a server or any other entity that the BBU 600 system accesses via communication unit 602. The computer program 605 can then... 1689323 of 44 to be downloaded from the server to memory 604. Figure 9, together with Figure 2, describes a first RU 120 configured to operate in a distributed base station system 100, wherein the first RU 120 comprises N1 antennas 121, 122, 123. The distributed base station system 100 further comprises a BBU 110 connected to the first RU 120 via a fronthaul link 140 and a second RU 160 connected to the first RU 120 via a link RU 165, wherein the second RU comprises N2 antennas. The first RU 120 comprises a processing circuit 703 and a memory 704. This memory contains instructions executable by said processing circuit, so the first RU 120 is operative to receive, from the BBU 110, K frequency domain user layer DL signals, to be sent to several UE 131, 132, 133, and to receive, from the BBU 110, a first part of BFW and a first and second portion of BFW for beamforming the K frequency domain user layer signals.where the first portion is based on a determined DL channel estimate of the first RU 120 and the first portion is determined to expand, in the frequency domain, the user layer signals to the antenna signals of antennas N1 121, 122, 123 of the first RU 120; and where the second portion is based on a determined DL channel estimate of the second RU 160 and the second portion is determined to expand, in the frequency domain, the user layer signals to the antenna signals of antennas N2 161, 162, 163 of the second RU 160, where the first part of the BFW is based on the determined DL channel estimates of the first RU 120 and the second RU 160,and where the first part of the BFW is determined to perform interference cancellation between the user layer signals. The first RU 120 is further operational to determine BFW for the first RU 120 based on the first part and the first portion received of the BFW and send to the second RU 160 via the RU 165 link, the first part and the second portion received of the BFW and send, to the second RU 160 via the RU 165 link, the K DL signals received from the user layer in the frequency domain and beam the K signals received from the user layer into antenna signals, using the BFW determined for the first RU 160, and send the antenna signals to UE number 131, 132, 133 via antennas N1 121, 122, 123. According to one embodiment, the first RU 120 is also operational for extracting the first portion and the first part of the BFW from the first part received and the first and second portion of the BFW. According to another embodiment, the first RU 120 is also 1689323 of 44 operational to decompress the first portion of the BFW when the first portion of the BFW was compressed in the BBU. According to another embodiment, the first RU 120 is further operative to receive, from BBU 110, the first compression information on the compression of the first portion of the BFW and the second compression information on the compression of the second portion of the BFW, the BBU compresses the first and second portions of the BFW to a higher degree than any compression of the first part of the BFW, wherein the first RU is operative to perform the decompression of the first portion of the BFW according to the first compression information and send the second compression information to the second RU 160 via the RU 165 link. According to another embodiment, the first received part and the first and second portions of the BFW are normalized. The first RU is also operative to receive, from BBU 110, information about a per-user-layer scaling factor related to normalization, and to scale the first normalized received part and the first portion of the BFW according to the scaling factor information before beamforming. According to yet another embodiment, the first RU 120 is also operational for sending information about the scale factor per user layer to the second RU 160. According to another embodiment, the distributed base station system 100 further comprises a third RU 170 connected to the second RU 160 via a second link RU 175, wherein the third RU 170 comprises antennas N3 171, 172, and 173. The first RU 120 is operational for BFW reception by also receiving a third portion of BFW, determined based on a DL channel estimate for the third RU 170, to expand, in the frequency domain, the user layer signals to the antenna signals of antennas N3 171, 172, and 173 of the third RU 170, wherein the first portion of the BFW is based on DL channel estimates for the first RU 120, the second RU 160, and the third RU 170. The first RU 120 is also operational for sending to the second RU 160 through the RU 165 link, the third portion of the BFW for additional transmission to the third RU 170 over the second RU 175 link. According to an alternative aspect, memory 704 of the first RU 120 in Figure 9 contains instructions executable by the processing circuit 703, so the first RU is operational to receive signals from various UE 131, 132, 133 1689323 of 44 UL antenna on the N1 antennas, and receive, from the BBU 110, a first part of BFW, and a first and second portion of the BFW, where the first portion is based on a determined UL channel estimate from the first RU 120 and the first portion is determined to combine, in the frequency domain, the UL antenna signals received on the N1 antennas to user layer signals; and where the second portion is based on a determined UL channel estimate from the second RU 160 and the second portion is determined to combine, in the frequency domain, antenna signals received on the N2 antennas of the second RU 160 with user layer signals; where the first part of the BFW is based on the UL channel estimates determined from the first RU 120 and the second RU 160, and the first part of the BFW is determined to perform interference cancellation between user layer signals.The first RU 120 is also operational to determine the BFWs for the first RU based on the first part and the first portion received from the BFWs, and to send, to the second RU 160 via the RU 165 link, the first part and the second portion received from the BFWs and to combine, in the frequency domain, the UL antenna signals received on the N1 antennas into the first K user layer signals, using the BFWs determined for the first RU 120.The first RU 120 is also operational for receiving, from the second RU 160, second K user layer signals combined by the second RU from antenna signals received at the N2 antennas of the second RU 160 from UE numbers 131, 132, 133, where the combination by the second RU is based on the first and second portions of the BFWs sent to the second RU by the first RU and combining the first and second K user layer signals into K combined user layer signals, and sending the K combined user layer signals to BBU 110. The embodiments mentioned above in relation to Figure 9 are also applicable to this aspect. According to other embodiments, the first RU 120 may further comprise a communication unit 702, which may be considered to include conventional means for wireless communication with UE 131, 132, and 133, such as a transceiver for wireless transmission and reception of signals. The communication unit 702 may also include conventional means for communication with BBU 110 and with the second RU 160. The instructions executable by said processing circuit 703 may be arranged as a computer program 705 stored, for example, in said memory 704. The processing circuitry 703 and the memory 704 may be arranged in a sub-arrangement 701. The sub-arrangement 701 may be a microprocessor and software. 1689323 of 44 and suitable storage, therefore, a programmable logic device, PLD, or other electronic components / processing circuits configured to perform the methods mentioned above. The 703 processing circuitry may comprise one or more programmable processors, application-specific integrated circuits, field-programmable gate assemblies, or combinations thereof adapted to execute instructions. The computer program 705 can be arranged so that when its instructions are executed in the processing circuit, they cause the first RU 120 to perform the steps described in any of the embodiments of the first RU 120 and its method. The computer program 705 can be carried by a computer program product connectable to the processing circuit 703. The computer program product can be memory 704 or at least be disposed in memory. Memory 704 can be realized as, for example, RAM (random access memory), ROM (read-only memory), or EEPROM (electrically erasable programmable ROM). In some embodiments, a carrier can hold the computer program 705. The carrier can be an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electrical signal, a radio signal, a microwave signal, or a computer-readable storage medium.The computer-readable storage medium can be, for example, a CD, DVD, or flash memory, from which the program could be downloaded into memory 704. Alternatively, the software program can be stored on a server or any other entity to which the first RU 120 has access via communication unit 702. Software program 705 can then be downloaded from the server to memory 704. Although the preceding description contains a number of specific details, these should not be interpreted as limiting the scope of the concept described herein, but rather as simply illustrating some exemplary embodiments of the concept. It will be appreciated that the scope of the concept described herein fully encompasses other embodiments that may be obvious to those skilled in the art, and that the scope of the concept described herein should not be limited. Reference to a singular element is not intended to mean "one and only one" unless explicitly stated, but rather "one or more." All structural and functional equivalents of the elements of the embodiments described above that are known to those skilled in the art are expressly incorporated herein as References to 1689323 of 44 are intended to be included in this document. Furthermore, it is not necessary for an apparatus or method to address each and every problem that the currently described concept seeks to solve in order to be covered herein. In the illustrative figures, a dashed line generally signifies that the feature within the dashed line is optional.

Claims

1. A method implemented by a baseband unit (BBU) system of a wireless communication network, the wireless communication network comprising a distributed base station system (100) comprising a BBU (110), a first radio unit (RU) (120) connected to the BBU (110) via a fronthaul link (140), the first RU comprising antennas N1 (121, 122, 123), and a second RU (160) connected to the first RU (120) via a RU link (165), the second RU (160) comprising antennas N2 (161, 162, 163), characterized in that the method comprises: determining (204) a first portion of beamforming weights (BFW) in the frequency domain, based on a downlink channel estimate (DL) of the first RU (120), wherein the BFWs are to be used to the formation of K-band user layer DL signals to be sent to several UEs (131, 132, 133), where the first portion of the BFW is determined to expand,in the frequency domain, the K user layer signals to antenna signals from antennas N1 (121, 122, 123) of the first RU (120); determine (206) a second portion of the BFW in the frequency domain, based on an estimate of the DL channel of the second RU (160), wherein the second portion of the BFW is determined to expand, in the frequency domain, the K user layer signals to antenna signals from antennas N2 (161, 162, 163) of the second RU (160); determine (212) a first part of the BFW in the frequency domain, based on the DL channel estimates of the first RU (120) and the second RU (160), wherein the first part of the BFW is determined (212) to perform interference cancellation between the user layer signals, activate the sending (214) of the first part of the BFW and the first and second portions of the BFW to the first RU (120),and activate the transmission (216) of the K user layer signals in the frequency domain to the first RU (120). Twenty claims follow,