Method of operating a network node in a communications network comprising a receiver node

Abstract

Method of operating a network node (101) having parallel processing units (102a, 102b, 102c, 102d), in a communications network comprising a receiver node (104) coupled via a packet network (103). The method comprises receiving (10) incoming jobs (J0, J1, J2, J3, J4, J5) for processing within a production window; distributing (20) jobs (J0, J1, J2, J3, J4, J5) for processing to available processing units (102a, 102b, 102c, 102d); assigning (30) a timer for each job indicating when the set of data packets (P0, P1, P2, P3, P4, P5) should be transmitted, so that sets (P0, P1, P2, P3, P4, P5) are spread evenly over a transmission window; and transmitting (40) the packets of the sets (P0, P1, P2, P3, P4, P5) to the receiver node (104) through the packet network (103) at the corresponding time instant.

Description

[0001] METHOD OF OPERATING A NETWORK NODE IN A COMMUNICATIONS NETWORK COMPRISING A RECEIVER NODE

[0002] TECHNICAL FIELD

[0003] The present disclosure relates generally to communications, and more particularly to methods of operating a network node in a communications network.

[0004] BACKGROUND

[0005] Nowadays, network nodes used in communications networks, for example and without limitations radio nodes and baseband nodes in mobile networks, are often connected to one another via packet networks. To that end, packet-switched network techniques such as proposed for the Evolved Common Public Radio Interfaces (eCPRI) and without being limitative, can be used. These packet-switched network, also known as fronthaul networks, can be point-to-point between baseband and radio nodes, or multi-point packet switched network.

[0006] In mobile networks implementing a split architecture, in contrast to a monolithic architecture where the base station node contains both the radio node parts and the baseband node parts, the radio nodes typically connect to the mobile terminals (User Equipment, UEs) over the radio air interface and implement the lower layers of the radio access network (RAN), whilst the baseband nodes implement the higher layers of the RAN. For uplink traffic from UE to RAN, the radio nodes receive mobile data traffic from the UEs via the antennas or antenna elements associated to each radio node, and process them at the lower physical layer before transmitting the ensuing data towards the baseband nodes for further RAN data processing.

[0007] The air interface between the radio node and the UEs is normally dimensioned to connect many UEs simultaneously to the RAN. In the uplink, each UE can generate a high volume of traffic towards the radio node over the air interface, but it is generally assumed that not all UEs connected to the radio node will reach their traffic peaks simultaneously. Therefore, the packet interface, the fronthaul, connecting the radio node to the baseband node can be dimensioned with the assumption of some statistical multiplexing gains with respect to UE traffic when transmitting uplink traffic from the radio node to the baseband node. In other words, the fronthaul link from the radio node to the baseband node can be underdimensioned in relation to the total capacity of the radio interface to the UEs.Due to the nature of mobile data traffic and the over-the-air protocols used to transmit it, the data transfers over these fronthaul links often come in bursts. When the momentary data rate of a burst exceeds the capacity of the fronthaul network, data can be delayed or lost. This can especially easily happen in a packet switched network such as an Ethernet network or an IP network in support of fronthauling, where many radio nodes may share the same links and packet switches and routers. This is mainly due to the best effort nature of packet switched networks.

[0008] A way of mitigating this problem, is pacing the transmission of data over the fronthaul link or network. In other words, making the sender of data on the fronthaul link spread out transmissions over a time window, sending them at a slower pace rather than all at once.

[0009] This transmission time window is then repeated continuously overtime, splitting the data to be transmitted over the fronthaul link into time windows, each window with a requirement of maximum data that can be transmitted, for example in terms of a maximum peak rate. A time window may serve to represent the number of packets per timing unit that a transmission interface can support for a given size of packet.

[0010] The use of highly parallelized processing in radio nodes and baseband nodes is an implementation method tailored to the nature of radio network transmissions, which contain many independent data streams to or from UEs, that could be processed in parallel.

[0011] However, to implement this parallel processing over a high-performance baseband or radio node, the processing units producing data to be sent over the fronthaul link (herein referred to as “data producers”) must be coordinated in order not to overwhelm the fronthaul link or network with high data rates from or to too many UEs simultaneously.

[0012] Data producers of this kind, running on parallel processing units, will receive data at random times from either the core network or user equipment in the access network, process it in various ways, and transmit the results over the fronthaul link.

[0013] Each data producer requires an allocation of hardware resources while running, including memory buffers for storing its output data. If many of these parallel data producers run at the same time, their combined data output may exceed the capacity of the fronthaul link, even if each process attempts to pace its data output.This undesirable effect can be mitigated by making each data producer run for a longer time, pacing its data over a wider window of time than its processing requires. This, however, would cause each data producer to tie up hardware resources in the baseband or radio node for a longer time, reducing the transmission capacity.

[0014] Alternatively, the data producers could send their data to a central pacing function, e.g. a buffer, in the baseband or radio node which would spread out the data in time before transmitting it over the fronthaul. Nevertheless, this would also consume more hardware resources, and would increase latency, causing lower performance for network users. It would also require some centralized mechanism on the network node to coordinate the buffering of data from the data producing parallel processors. That in turn will counter the benefits of parallel processors on the network node, since they would need to be centrally controlled or coordinated.

[0015] Therefore, still there is a need to develop other methods of operating network nodes in a communications network, that are capable of higher data transmission performance and lower consumption of hardware and software resources.

[0016] SUMMARY

[0017] According to some embodiments, a method of operating a network node in a communications network comprising a receiver node is provided. The receiver node can be communicatively coupled to the network node via a packet network and the network node can comprise a plurality of parallel processing units. The method can comprise receiving a plurality of incoming jobs for processing within a production window. The method can further comprise distributing the plurality of incoming jobs for processing to available processing units, the processing of jobs generating sets of one or more packets to be transmitted to the receiver node. The method can further comprise assigning a timer for each incoming job, the timer indicating when the set generated by each job should be transmitted to the receiver node and the timers being selected so that the sets are spread evenly over a transmission window. The method can further comprise transmitting the packets the sets to the receiver node through the packet network at the time instant corresponding to the assigned timer.According to other embodiments, a baseband node, radio node, computer program, computer program product, and a non-transitory computer readable medium are provided to perform the computer implemented methods above.

[0018] Various embodiments described herein are capable of pacing the transmission of data to avoid overloading of the packet network, while using a minimal amount of additional hardware resources, and without using any communication between data producers Thus, by not using additional hardware these embodiments of the present invention decrease the hardware cost in the baseband or radio node to reach a given capacity. On the other hand, these embodiments of the invention could increase the amount of memory employed for creating packets to be transmitted to the receiver node, since output memory must be held longer until transmission. However, this is a marginal increase compared to data transmission without pacing, where the main determinant for memory consumption is dimensioning for peak processing load. Finally, by not relying on communication between data producers, these embodiments of the present invention decrease the latency of data passing through the baseband network node. Avoiding inter-producer communication also simplifies the system, which makes it easier to understand and work with, in turn reducing software defects and reducing the cost of adding new system features.

[0019] BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this application, illustrate certain non-limiting embodiments of the inventive concepts in the drawings:

[0021] Fig. 1 is a schematic diagram showing a prior art method of operating a network node in a communication network, wherein data is transmitted to a fronthaul link without any pacing;

[0022] Fig. 2 is a schematic diagram showing a prior art method of operating a network node in a communication network, wherein data is transmitted to a fronthaul link with pacing;

[0023] Figs. 3A and 3B are graphs representing the fronthaul link load versus time, according to the operating method of Fig. 1 and according to the operating method of Fig. 2, respectively.

[0024] Fig. 4 is a schematic diagram showing, in a simplified way, a method of operating a network node according to the present invention.Fig. 5 is a flow diagram illustrating the steps of a method of operating a network node according to an embodiment of the present invention.

[0025] Fig. 6 is a schematic diagram showing, the production window and the transmission window corresponding to the method of operating a network node of Fig. 5.

[0026] Fig. 7 is a graph that plots the transmission of data packets over time within the transmission window, both in a prior art method of operating a network node and a method according to Figs. 5 and 6.

[0027] DETAILED DESCRIPTION

[0028] Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art. Inventive concepts may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of present inventive concepts to those skilled in the art. It should also be noted that these embodiments are not mutually exclusive. Components from one embodiment may be tacitly assumed to be present / used in another embodiment. Equal same or similar elements will be designated with the same numerical references throughout the figures.

[0029] Fig. 1 is a schematic diagram illustrating a prior art method of operating a network node 201 in a communication network, wherein data is transmitted to a receiver node (not shown in the figure) via a fronthaul link 203. The fronthaul link 203, can also be viewed as a port of the network node, the port leading toward a physical fronthaul link, i.e. representing an interface of the network node 201 providing access to a fronthaul link. In this particular case, data is transmitted without any pacing to the receiver node. This interface may be an Ethernet interface, an IP interface, an optical interface or other media interface, or any combination of such interfaces, e.g. an Ethernet interface operating over an optical physical interface.

[0030] In the example shown in Fig. 1 , the network node comprises three different processing units 202a, 202b, 202c working in parallel. The three different processing units may represent such units on the network node (in this case a radio node for uplink data traffic) that processdata received over the air interface from three different UEs. Each processing unit generates data packets to be transmitted over the physical link 203 to other entities (e.g. the baseband node) for further processing of the data. In Fig. 1, above each processing unit 202a, 202b, 202c the volume, e.g. in number of bytes, that data packet or packets that said processing unit generates is represented as a dashed rectangle. The height of the dashed rectangle represents the data volume of the packet or packets, while the width thereof represents the time window during which said data packet or packets are transmitted to a receiver node via the interface towards the fronthaul link 203.

[0031] As can be seen in Figure 3A corresponding to this method of network node without pacing, since the packets generated by the three processing units 202a, 202b, 202c are transmitted at the same time to the fronthaul link 203, the total load on the fronthaul link 203 corresponds to the sum of the data packets from each of them. In this case, the total capacity (represented with a dashed horizontal line) of the fronthaul link is exceeded at the moment when the three packets arrive, something that is highly undesirable, since it can lead to data being delayed due to buffering or even lost.

[0032] Fig. 2 is a schematic diagram illustrating a prior art method of operating a network node 301 in a communication network, wherein data is transmitted to a receiver node (not shown in the figure) via a fronthaul link 303. In this particular case, data is transmitted with pacing.

[0033] As in the previous case, in the example shown in Fig. 2, the network node comprises three different processing units 302a, 302b, 302c working in parallel. Above each processing unit 302a, 302b, 302c the data packet that said processing unit generates, to be transmitted to the receiver node, is represented as a dashed rectangle. The height of the dashed rectangle represents the data volume of the packet, while the width thereof represents the time window during which said data packet is transmitted to a receiver node via the fronthaul link 303.

[0034] As it can be clearly seen in Figure 3B (corresponding to this method of network node operation with data pacing), the fact of pacing the data transmission prevents the fronthaul link from exceeding its maximum transmission capacity (represented with a dashed horizontal line). However, this prior art method still has disadvantages, in particular that data packets are transmitted to the receiver node over a wider window of time than its processing strictly requires. This, could cause each processing unit to tie up hardware resources in the network node for a longer time, reducing its processing capacity. Alternatively, it would cause the data produced by each processing unit to be buffered in buffers in the network node.Fig. 4 is a schematic diagram showing, in a simplified way, a method of operating a network node according to the present invention.

[0035] In the example of the invention illustrated in Fig, 4, the network node 101 comprises three different processing units, 102a, 102b and 102c working in parallel and forms part of a communications network. A receiver node 104 is communicatively coupled to the network node 101 via a packet network, the packet network comprising in this particular case, a fronthaul link 103.

[0036] The network node 101 also comprises a memory 105 coupled to the processing units 102a, 102b and 102c and having instructions stored therein that are executable by the processing units 102a, 102b, 102c to cause the network node 101 to perform operations comprising any of the operations of a method of operating a network according to the present invention.

[0037] The network node 101 may be, for example, but not limited to a radio node or a baseband node. In some Open Radio Access Network architectures including O-RAN Alliance’s O-RAN architecture radio nodes and baseband nodes may be named differently. In O-RAN, a baseband node may correspond to an O-DU (Distributed Unit), or a combination of O-DU and O-CU (Centralized Unit), and a radio node may correspond to an O-RU (Radio Unit). A baseband node may also be referred to as a Digital Unit (DU). In cloudified RAN, the baseband node or O-DU may also be defined as a RAN network function (NF) implemented in a cloud. As a further non-limiting example, a network node may be a WiFi access point.

[0038] User equipment (UE) include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless cameras, gaming console or device, music storage device, playback appliance, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment, laptop-mounted equipment, smart device, wireless customer-premise equipment (CPE), vehicle-mounted or vehicle embedded / integrated wireless device, etc. Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-loT) UE, a machine type communication (MTC) UE, and / or an enhanced MTC (eMTC) UE.

[0039] As in the previous cases, above each processing unit 102a, 102b, 102c the data packet that said processing unit generates, to be transmitted to the receiver node 104, is represented asa dashed rectangle. The height of the dashed rectangle represents the data volume of the packet, while the width thereof represents the time window during which said data packet is transmitted to the receiver node 104 via the fronthaul link 303.

[0040] As schematically depicted in figure 4, thanks to the special pacing contemplated by the method of the invention shown in Fig 4, it is possible to prevent the fronthaul link 103 from exceeding its maximum work capacity (represented with a dashed horizontal line). In addition, the method of the invention is also capable of transmitting the corresponding data packets in a shorter time window than prior art pacing methods, resulting in higher transmission capacity and less hardware resources consumption.

[0041] Fig. 5 is a flow diagram illustrating the steps of a method of operating a network node 101 according to an embodiment of the present invention.

[0042] Fig. 6 gives additional details of the method of Fig. 5 such as, for example, a time representation of the production window and the transmission window applied in this method.

[0043] In the example of the invention illustrated in Figs. 5 and 6, the network node 101 is identical to the one illustrated in Fig. 4, with the only exception that it comprises four different processing units, 102a, 102b, 102c and 102d working in parallel and coupled to a memory 105 (as schematically represented in the left-hand part of Fig 6). The communications network is also identical to the one illustrated in Fig. 4: the receiver node 104 (not shown in Figs 5 and 6) is communicatively coupled to the network node 101 via a packet network, and the packet network comprising in this particular case, a fronthaul link 103.

[0044] The network node 101 can optionally be a baseband node transmitting data in the downlink direction over the packet network 103 towards a radio node. Alternatively, the network node 101 can also optionally be a radio node transmitting data in the uplink direction over the packet network 103 towards a baseband node. In some topologies, e.g. a tree topology where the root of the tree is a baseband unit and the branches are fronthaul links, radio nodes can act as forwarding devices for other radio nodes and for the baseband node.

[0045] The method of Figs. 5 and 6 starts with step 10, wherein the network node 101 receives a plurality of incoming jobs JO, J1 , J2, J3, J4 and J5 for processing within a production window, (as shown in the right-hand part of fig. 6). A job may be created or scheduled by data, received on the network node, which needs to be processed by the network node beforebeing sent over the fronthaul interface or the air interface. Next, in step 20, the plurality of incoming jobs is distributed to available processing units 102a, 102b, 102c and 102d.

[0046] In an embodiment of the method of the present invention, an incoming job JO, J1, J2, J3, J4, J5 is assigned to one processing unit 102a, 102b, 102c, 102d. Moreover, the method according to the present invention also contemplates the possibility that the processing unit 102a, 102b, 102c, 102d processes one job JO, J1, J2, J3, J4, J5 at the time.

[0047] Each processing unit eventually generates a set (or group) P0, P1 , P2, P3, P4 and P5 consisting of one or more transmission packets, to be transmitted to the receiver mode, for each incoming job. In an embodiment of the method of the present invention, each job JO, J1, J2, J3, J4, J5 generates a maximum of one set P0, P1, P2, P3, P4, P5.

[0048] In an embodiment of the invention, the set P0, P1, P2, P3, P4, P5 is ready for transmission at the end of the processing of the corresponding job JO, J1, J2, J3, J4, J5. Moreover, typically, at the end of the processing of a job, the packets to send as a result of that job are ready for sending and all the packets of the set are transmitted simultaneously.

[0049] In step 30 a timer, to, t1 , t2, t3, t4, and t5 indicating when the packets generated by each job should be transmitted to the receiver node, is assigned for each incoming job, JO, J1 , J2, J3, J4 and J5. In this particular example, in the sake of simplicity timer to is assigned to incoming job JO, timer t1 is assigned to incoming job J1 and so on.

[0050] Preferably, the timer to, t1 , t2, t3, t4, t5 for a job JO, J1 , J2, J3, J4, J5 is assigned when the job JO, J1, J2, J3, J4, J5 is assigned to a processing unit 102a, 102b, 102c, 102d.

[0051] Note that the model does not exclude one job to trigger the creation of other jobs. Since the processing of a job involves RAN related processing at some level, e.g. layer 1 in a radio node, that processing may result in the conclusion that further jobs need to be generated on the radio node as a consequence of the first job. For example, if the processing of a job results in the identification of packets of very high priority to be sent, a new job could be created in order for the packets of the first job to be replicated for reasons of resiliency.

[0052] The present invention contemplates selecting timers to, t1, t2, t3, t4, and t5, so that the set of packets P0, P1 , P2, P3, P4 and P5 are spread evenly over a transmission window.In addition, in the embodiment of the invention shown in Figs. 5 and 6, the transmission window, is divided into bins 0, 1, 2, 3, 4, 5, 6, 7, 8 and 9 and the timers tO, t1 , t2, t3, t4, and t5 relate to which bin 0, 1, 2, 3, 4, 5, 6, 7, 8 or 9 shall be used for transmission of data from the job. In the present invention, the bins 0, 1, 2, 3, 4, 5, 6, 7, 8 and 9 of the transmission window are, preferably, equally sized bins, as it is the case in the example shown in Figs. 5 and 6. By ensuring that traffic generated by the jobs are spread evenly across the bins, the load on fronthaul links or interfaces is spread out.

[0053] Each timer to, t1, t2, t3, t4, and t5 preferably corresponds to a semi-unique identifier assigned to each incoming job, said identifier spanning from the earliest bin number which has not yet expired to the latest existing bin number which has not yet expired.

[0054] More preferably, the semi-unique identifier assignment is performed by means of a fast hash function, the input of the fast hash function being at least part of the data of the transmission packet in the set P0, P1 , P2, P3, P4 and P5 to be transmitted to the receiver node (104), as it is the case in the example shown in Figs. 5 and 6.

[0055] It is also preferred that the production window partially overlaps with the transmission window. More preferably, the bins 0, 1, 2, 3, in the beginning of the transmission window where the production windows overlaps the transmission window, are repeated at the end of the transmission window, as happens in the example shown in Figs. 5 and 6.

[0056] In the example shown in figs. 5 and 6, timers are related with bins according to the table below:

[0057]

[0058] Finally, the method of Figs. 5 and 6 ends with step 40, wherein the data packets of the sets P0, P1 , P2, P3 P4 and P5 corresponding to each incoming job JO, J1 , J2, J3, J4 and J5 aretransmitted to the receiver node 104 through the packet network 103 at the time instant corresponding to the assigned timer to, t1 , t2, t3, t4 and t5.

[0059] Fig. 7 illustrates the transmission of data packets over time within the transmission window, both in a prior art method and in the method of Figs. 5 and 6 according to the present invention.

[0060] As can be clearly seen in Fig. 7 the dotted line representing prior art method shows a massive spike at the start of the transmission window. This is from all the transmissions that were ready to be sent before the transmission window opened. After the initial burst, there is still a bias towards the start of the transmission window.

[0061] Conversely, in the method of Figs. 5 and 6 according to the present invention, illustrated with the solid line in Figure 7, there is a more spread-out load throughout the transmission window. The peak load is significantly lower compared to the prior art method.

[0062] Although the computing devices described herein (e.g., network nodes, receiving nodes) may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and / or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and / or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination. Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of the components may be partitioned between the processing circuitry and the communication interface. In another example, non-computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry executing instructions stored on in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer-readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device- readable storage medium, such as in a hard-wired manner. In any of those particular embodiments, whether executing instructions stored on a non-transitory computer-readable storage medium or not, the processing circuitry can be configured to perform the described functionality. The benefits provided by such functionality are not limited to the processing circuitry alone or to other components of the computing device, but are enjoyed by the computing device as a whole, and / or by end users and a wireless network generally.

Claims

CLAIMS1. Method of operating a network node (101) in a communications network comprising a receiver node (104), the receiver node communicatively coupled to the network node via a packet network (103), the network node (103) comprising a plurality of parallel processing units (102a, 102b, 102c, 102d), the method comprising:-receiving (10) a plurality of incoming jobs (JO, J1, J2, J3, J4, J5) for processing within a production window;-distributing (20) the plurality of incoming jobs (JO, J1, J2, J3, J4, J5) for processing, to available processing units (102a, 102b, 102c, 102d), the processing of jobs (JO, J1, J2, J3, J4, J5) generating sets (P0, P1 , P2, P3, P4, P5) of one or more packets to be transmitted to the receiver node (104);-assigning (30) a timer (to, t1, t2, t3, t4, t5) for each incoming job (JO, J1 , J2, J3, J4, J5), the timer (to, t1 , t2, t3, t4, t5) indicating when the set (P0, P1 , P2, P3, P4, P5) generated by each job (JO, J1 , J2, J3, J4, J5) should be transmitted to the receiver node (104), the timers (to, t1 , t2, t3, t4, t5) being selected so that the sets (P0, P1 , P2, P3, P4, P5) are spread evenly over a transmission window; and- transmitting (40) the packets of the sets (P0, P1 , P2, P3, P4, P5) to the receiver node (104) through the packet network at the time instant corresponding to the assigned timer (tO, t1, t2, t3, t4, t5).

2. The method of any of the preceding claims, wherein an incoming job (JO, J1, J2, J3, J4, J5) is assigned to one processing unit (102a, 102b, 102c, 102d).

3. The method of any of the preceding claims, wherein the processing unit (102a, 102b, 102c, 102d) processes one job (JO, J1, J2, J3, J4, J5) at the time.

4. The method according to any of the preceding claims, wherein each job (JO, J1 , J2, J3, J4, J5) generates a maximum of one set (P0, P1, P2, P3, P4, P5).

5. The method according to any of the preceding claims, wherein the set (P0, P1, P2, P3, P4, P5) is ready for transmission at the end of the processing of the corresponding job (JO, J1, J2, J3, J4, J5).

6. The method of any of the preceding claims, wherein all packets of a set (P0, P1, P2, P3, P4, P5) are transmitted simultaneously.

7. The method of any of the preceding claims, wherein the timer (to, t1 , t2, t3, t4, t5) for a job (JO, J1, J2, J3, J4, J5) is assigned when the job (JO, J1, J2, J3, J4, J5) is assigned to a processing unit (102a, 102b, 102c, 102d).

8. The method according to claim 1 or 2, wherein the production window ends before the transmission window, ends.

9. The method according to claim 1 or 2, wherein the production window starts before the transmission window.

10. The method according to any of claims 1 to 3, wherein the transmission window is divided into bins (0, 1, 2, 3, 4, 5, 6, 7, 8, 9) and the timers (tO, t1, t2, t3, t4, t5) relate to which bin (0, 1, 2, 3, 4, 5, 6, 7, 8, 9) shall be used for transmission of data from the job (JO, J1, J2, J3, J4, J5).

11. The method according to claim 4, wherein each timer (to, t1 , t2, t3, t4, t5) corresponds to a semi-unique identifier assigned to each incoming job (JO, J1 , J2, J3, J4, J5), said identifier spanning from the earliest bin number which has not yet expired to the latest existing bin number which has not yet expired.

12. The method according to claim 5, wherein the identifier assignment is performed by means of a fast hash function, the input of the fast hash function being at least part of the data of the transmission packet in the set (P0, P1 , P2, P3, P4, P5) to be transmitted to the receiver node (104).

13. The method according to any of claims 4 to 6, wherein the bins (0, 1, 2, 3, 4, 5, 6, 7, 8, 9) of the transmission window are equally sized bins.

14. The method according to any of claims 4 to 7, wherein the production window partially overlaps with the transmission window .1515. The method according to claim 8, wherein the bins (0, 1 , 2, 3) in the beginning of the transmission window where the production windows overlaps the transmission window are repeated at the end of the transmission window.

16. The method according to any of claims 1 to 9, wherein the network node (101) is a baseband node transmitting data in the downlink direction over the packet network (103).

17. The method according to any of claims 1 to 9, wherein the network node (101) is a radio node transmitting data in the uplink direction over the packet network (103).

18. Baseband node (101) in a communications network that includes a receiver node (104) communicatively coupled to the baseband node via a packet network, the baseband node comprising:-a plurality of parallel processing units (102a, 102b, 102c, 102d); and-memory coupled to the processing units and having instructions stored therein that are executable by the processing units (102a, 102b, 102c, 102d) to cause the baseband node to perform operations comprising any of the operations of claims 1 to 17.

19. Radio node in a communications network that includes a receiver node (104) communicatively coupled to the baseband node via a packet network, the radio node comprising:- a plurality of parallel processing units (102a, 102b, 102c, 102d); and-memory coupled to the processing units and having instructions stored therein that are executable by the processing units (102a, 102b, 102c, 102d) to cause the radio node to perform operations comprising any of the operations of claims 1 to 17.

20. A computer program, comprising program code to be executed by a plurality of parallel processing units (102a, 102b, 102c, 102d) of a network node (101) in a communications network that includes a receiver node (104) communicatively coupled to the network node via a packet network (103), whereby execution of the program code causes the network node (101 )to carry out operations comprising any of the operations of claims 1 to 17.

21. A computer program product, comprising a non-transitory storage medium including program code to be executed by a plurality of parallel processing units (102a, 102b, 102c, 102d) of a network node (101) in a communications network that includes a receiver node (104) communicatively coupled to the network node via a packet network (103), whereby execution of the program code causes the network node to carry out operations comprising any of the operations of claims 1 to 17.

22. A non-transitory computer readable medium having instructions stored therein that are executable by a plurality of parallel processing units of a network node (101) in a communications network that includes a receiver node (104) communicatively coupled to the network node via a packet network (103), to cause the network node (101) to carry out operations comprising any of the operations of claims 1 to 17.

Images