Method and apparatus for supporting estimation of delays on a communication path in a communication network
By sending sample bursts in a wireless communication network and identifying acknowledge packets of specific modes, the problem of delay estimation under the QUIC protocol is solved, the delay and RTT of QUIC packets are accurately measured, and the network load detection is improved accuracy.
Patent Information
- Application Number
- CN202080103229.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-08-18
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2040-08-18
AI Technical Summary
The existing technology is difficult to effectively estimate delays in wireless communication networks, especially when using the QUIC protocol, which cannot perform RTT measurements through spin-bit mechanisms like the TCP protocol, resulting in difficulty in congestion detection and load level assessment.
By sending sample bursts on the communication path, using the detection node to identify acknowledge packets in a specific mode, estimating the delay, avoiding the difficulty of acknowledge packet identification of QUIC packets, and realizing the delay estimation of spin-free QUIC packets.
It realizes effective estimation of delay in wireless communication networks, and can accurately calculate RTT when supporting QUIC packets, avoids undesired plastic surgery and confusion, and improves the accuracy of network load detection.
Smart Images

Figure CN115918043B_ABST
Abstract
Description
Technical Field
[0001] Embodiments herein relate to methods and apparatus relating to estimation of delays on communication paths in a communication network (eg, a wireless communication network such as a telecommunications network), such as based on round trip time (RTT). Background Art
[0002] Communication devices such as wireless communication devices (which may be referred to as wireless devices for short) may also be referred to as, for example, user equipment (UE), mobile terminals, wireless terminals and / or mobile stations. Wireless devices are enabled to perform wireless communications in a wireless communication network, a wireless communication system, or a radio communication system (e.g., a telecommunications network, sometimes also referred to as a cellular radio system, a cellular network, or a cellular communication system). Communications may be performed, for example, between two wireless devices, between a wireless device and a conventional phone, and / or between a wireless device and a server via a radio access network (RAN) and possibly one or more core networks included in a cellular communication network. To name just a few other examples, wireless devices may also be referred to as mobile phones, cellular phones, laptop computers, personal digital assistants (PDAs), tablet computers. Wireless devices may be so-called machine-to-machine (M2M) devices or machine communication type (MTC) devices, i.e., devices that are not associated with conventional users.
[0003] The wireless device may be, for example, a portable, pocket-storable, handheld, computer-containing, or vehicle-mounted mobile device enabled to communicate voice and / or data with another entity (such as another wireless device or a server) via the RAN.
[0004] A wireless communication network may cover a geographical area divided into cell areas, wherein each cell area is served by at least one base station, or base station (BS), e.g., a radio base station (RBS), which may sometimes be referred to as, for example, "eNB", "eNodeB", "NodeB", "B node", or BTS (base transceiver station), depending on the technology and terminology used. Base stations may belong to different categories, such as, for example, macro eNodeB, home eNodeB, or pico base station, based on transmission power and thus also based on cell size. A cell is typically identified by one or more cell identities. A base station at a base station site may provide radio coverage for one or more cells. Thus, a cell is typically associated with a geographical area in which radio coverage of the cell is provided by a base station at the base station site. Cells may overlap, so that several cells cover the same geographical area. Being provided or served by a base station typically means that the base station provides radio coverage so that one or more wireless devices located in the geographical area in which the radio coverage is provided can be served by the base station in the cell. When a wireless device is described as being served in a cell or being served by a cell, this means that the wireless device is served by the base station providing radio coverage for the cell. One base station may serve one or several cells. Furthermore, each base station may support one or several communication technologies.The base stations communicate with wireless devices within range of the base station over an air interface operating on radio frequencies.
[0005] In some RANs, several base stations may be connected to a radio network controller (e.g., a radio network controller (RNC) in a universal mobile telecommunications system (UMTS)) and / or to each other, for example, by landline or microwave. A radio network controller (sometimes also referred to as a base station controller (BSC), e.g., in GSM) may supervise and coordinate various activities of the multiple base stations connected to it. GSM is an abbreviation for Global System for Mobile Communications (formerly Groupe Special Mobile), which may be referred to as the second generation or 2G.
[0006] UMTS is a third generation mobile communication system (which may be referred to as third generation or 3G and evolved from GSM) and provides improved mobile communication services based on wideband code division multiple access (WCDMA) access technology. The UMTS Terrestrial Radio Access Network (UTRAN) is essentially a radio access network using wideband code division multiple access for wireless devices. High Speed Packet Access (HSPA) is a merger of two mobile phone protocols defined by 3GPP, namely High Speed Downlink Packet Access (HSDPA) and High Speed Uplink Packet Access (HSUPA), which uses WCDMA to expand and improve the performance of existing third generation mobile telecommunication networks. Such a network may be named WCDMA / HSPA.
[0007] The expression downlink (DL) may be used for the transmission path from the base station to the wireless device. The expression uplink (UL) may be used for the transmission path in the opposite direction, ie, from the wireless device to the base station.
[0008] In 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE), base stations (which may be referred to as eNodeBs or eNBs) may be directly connected to other base stations and may be directly connected to one or more core networks. LTE may be referred to as 4th Generation or 4G.
[0009] 3GPP has embarked on further development of UTRAN and GSM based radio access network technologies, for example, into the Evolved UTRAN (E-UTRAN) used in LTE.
[0010] Work is underway to develop the next generation of wide area networks, which may be referred to as Next Generation (NX), New Radio (NR), or Fifth Generation (5G).
[0011] QUIC is a stream multiplexing and secure transport protocol based on the User Datagram Protocol (UDP) with integrity protection headers and encrypted payloads. Unlike traditional transport protocol stacks using the Transmission Control Protocol (TCP), which reside in the operating system kernel, QUIC can be easily implemented in the application layer. This brings great flexibility in evolution, congestion control, and new feature implementation. QUIC is being organized by the Internet Engineering Task Force (IETF), see for example "QUIC: A UDP-Based Multiplexed and Secure Transport" (Draft-ietf-quic-transport-29, June 10, 2020) and "Using TLS to Secure QUIC" (Draft-ietf-quic-tls-29, June 9, 2020).
[0012] QUIC is likely to become the main transport protocol in the user plane of wireless communication networks and other communication networks. It is expected that most web applications that currently run mainly over HTTP / HTTPS will migrate to QUIC.
[0013] An important difference between QUIC and TCP is that QUIC is an encrypted protocol. Protocols running on top of TCP are usually protected by the Transport Layer Security (TLS) protocol, which provides protection for the payload but not for the TCP header. QUIC, on the other hand, provides protection for both the payload and the packet header. Two separate keys are used to protect QUIC packets, one for protecting the packet header and the other for protecting the packet payload. In addition, the QUIC packet header includes fewer fields than its TCP counterpart. Instead, QUIC uses frames that are part of the protected packet payload to transmit relevant information such as acknowledgments.
[0014] QUIC allows the migration of connections between different Internet Protocol (IP) addresses and ports. This migration can happen implicitly (e.g., due to rebinding Network Address Translation (NAT) state) or explicitly (e.g., because the client decides to use a separate access network when it becomes available).
[0015] Latency measurements are often performed by network nodes to understand the characteristics of an end-to-end flow traversing a network and the behavior of the network segments that the flow is traversing. The most common form of latency measured on an end-to-end flow is or is based on packet round trip time (RTT). RTT is usually defined as the time from sending a packet to receiving an acknowledgment of that packet.
[0016] Since QUIC is an encrypted protocol, it is impossible for a passive observer who is not an explicit part of the QUIC conversation to see which acknowledgment refers to which packet. Therefore, it is not possible to measure RTT for QUIC in the same way as TCP.
[0017] The QUIC protocol defines a mechanism called the spin bit, which is designed to allow passive observers on the path to measure RTT by observing the period of change in the spin bit value. However, this explicit signal is an optional part of the QUIC protocol and cannot be assumed to be always present.
[0018] In wireless communication networks, latency measurements and RTT are also important metrics and are used, for example, to understand the current situation in the network, for example to obtain an indication of whether there is congestion or other problems in the RAN or parts thereof, or the potential for such problems ahead. Wireless communication networks can be configured to take various actions based on latency measurements and RTT. For example, current congestion detection techniques use RTT as the main source of information for determining the load level in the RAN. Therefore, when the observed traffic is not TCP-based or QUIC-based (with the spin bit enabled), current congestion detection techniques will not work.
[0019] Therefore, when data and data streams (e.g., containing streaming video) are provided via QUIC packets, estimating latency can become an issue. Summary of the invention
[0020] In view of the above, an object is to provide one or more improvements relating to the prior art, and in particular to provide improvements regarding the estimation of delays on communication paths in a communication network, and which will also work when the data packets are QUIC packets.
[0021] According to a first aspect of an embodiment of the present invention, the purpose is achieved by a method for estimating a delay on a communication path between a data flow shaping node and a detection node in a communication network, performed by one or more devices. The communication path includes a data flow target node configured to receive a data flow from a data flow source node via the communication path. The data flow shaping node is configured to provide a data flow to the data flow target node in a burst. The one or more devices initiate the sending of a sample burst by the data flow shaping node, the sample burst being a burst in the burst that is different from the previous and subsequent bursts in a predefined manner, so that the data flow target node sends a confirmation packet of a specific pattern in response to the sample burst, and the specific pattern is different from the pattern of the confirmation packet sent by the data flow target node in response to the previous and subsequent bursts. The one or more devices further initiate the reception of a potential confirmation packet identified by the detection node according to the specific pattern.
[0022] According to a second aspect of an embodiment of the present invention, the purpose is achieved by one or more devices for supporting the estimation of the delay on the communication path between a data flow shaping node and a detection node in a communication network. The communication path includes a data flow target node configured to receive a data flow from a data flow source node via the communication path. The data flow shaping node is configured to provide a data flow to the data flow target node in a burst. The one or more devices are configured to initiate the sending of a sample burst by the data flow shaping node, the sample burst being a burst in the burst that is different from the previous and subsequent bursts in a predefined manner, so that the data flow target node sends a confirmation packet of a specific pattern in response to the sample burst, and the specific pattern is different from the pattern of the confirmation packet sent by the data flow target node in response to the previous and subsequent bursts. The one or more devices are further configured to initiate the reception of a potential confirmation packet identified by the detection node according to the specific pattern.
[0023] The one or more devices may in turn or may be configured to initiate determining an estimate of the delay based on timing information regarding transmission of the sample burst and receipt of said identified potential acknowledgment packet by the probing node.
[0024] In some embodiments, the probing node is a data shaping node and the communication path corresponds to a round trip and the estimate of latency may correspond to a calculation of a round trip time (RTT).
[0025] By identifying the specific pattern in potential ACK packets, these packets are likely to be actual ACK packets, and the risk of mixed ACK packets between bursts can also be avoided. It is even possible to find out which exact data packet of a sample burst generated a certain ACK packet of the pattern. This means that latency can be estimated. Therefore, embodiments of the present invention support the estimation of latency, which also works when the data packet is a QUIC packet without a spin bit.
[0026] Another advantage of embodiments herein is that the sample burst itself (i.e. above the packet level, e.g. its data rate, length, etc.) need not be affected or changed compared to a regular burst. Thus, a sample burst may appear identical to previous and subsequent bursts, since only, e.g., the order of the data packets may have changed. This means that undesirable shaping may be avoided when implementing embodiments herein.
[0027] Thus, embodiments herein support estimation of latency (eg, calculation of RTT) and also function when the data packet is a QUIC packet without a spin bit. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Examples of embodiments herein are described in more detail with reference to the accompanying schematic drawings, which are briefly described below.
[0029] Figure 1 is a block diagram schematically depicting an example of a first communication system for discussing embodiments herein.
[0030] Figure 2 is another block diagram schematically depicting an example of a second communication system for discussing embodiments herein.
[0031] FIG. 3 schematically illustrates an example of how a data stream may be shaped and sent in bursts.
[0032] FIG. 4 schematically illustrates an example for explaining how packets of a regular burst result in acknowledgement packets.
[0033] Figure 5 An example is schematically illustrated to illustrate how a sample burst with reordered packets may result in a particular pattern of acknowledgment packets according to some embodiments herein.
[0034] Figure 6 Examples are schematically illustrated to further illustrate how sampling windows and matching windows may be related to each other and used in embodiments herein.
[0035] Figure 7 is a flow chart for schematically illustrating an exemplary method according to some embodiments herein.
[0036] Figure 8 is a flow chart for schematically illustrating a method according to an embodiment of the present invention.
[0037] Fig. 9 is a schematic block diagram for illustrating an embodiment of how one or more devices may be configured to perform a first method.
[0038] Fig.10 It is a schematic diagram illustrating some embodiments related to a computer program and a carrier thereof for causing a device to perform the first method and related actions. DETAILED DESCRIPTION
[0039] In the following description, like reference numerals may be used to identify like elements, units, modules, circuits, nodes, components, items or features when applicable.When embodiments are illustrated in the drawings, features that appear only in some embodiments are generally indicated by dashed lines.
[0040] The embodiments herein are described by exemplary embodiments. It should be noted that these embodiments are not mutually exclusive. Components from one embodiment may be assumed to exist in another embodiment by default, and it will be apparent to those skilled in the art how these components may be used in other exemplary embodiments.
[0041] As the embodiments of this document are developed, some ideas that are the basis of the embodiments of this document will first be discussed.
[0042] In order to deliver services (i.e., data) from a host (e.g., a server) to a client (e.g., a wireless communication device), data shaping is often applied. It is, for example, a common practice to perform throttling of adaptive bit rate (ABR) video streams in cellular networks. Nodes in a communication network (e.g., CN) or in a local area network (LAN) detect streams delivering videos with ABR characteristics and shape the set of streams to the configured bit rate. The shaping of ABR video streams is likely to result in a reduction in the capacity consumed by the video services carried by these streams. ABR clients can continuously estimate the downlink bandwidth and select the video resolution that best matches the available bandwidth.
[0043] For wireless communication networks, the volume reduction that can be achieved by shaping the video is most useful when the RAN is under high load. Therefore, the network nodes that perform the shaping are also typically associated with a load or congestion detection mechanism. Such mechanisms typically use RTT-based latency estimates as the primary source of information for determining the load level in the RAN.
[0044] When shaping traffic, for example in wireless communication networks (e.g., cellular and / or telecommunication networks), it has proven beneficial to send data in large bursts consisting of several maximum transmission unit (MTU) sized data packets at a time (i.e., sending data in transmission periods with silent periods in between). The benefit of sending data in bursts is that it mitigates some of the negative impacts on RAN throughput metrics that are typically observed when sending shaped data packets packet by packet or in small bursts. In addition, there are some shaping techniques that can mitigate the impact of large data flows on the loaded RAN by sending data in bursts. Bursts are typically formed (i.e., data streams are transformed into bursts) by downstream nodes or devices, for example, a node or device in the CN or RAN may correspond to a data flow shaper or data flow shaping node and may be configured to convert the received data flow into bursts for further delivery to a client (such as a wireless device).
[0045] Embodiments herein may be described in a simplified manner as exploiting the presence of bursts and / or the ability to form bursts (e.g., when there is a data shaping node in the communications network configured to shape the data stream into bursts) to find latency or a measure thereof (e.g., RTT) in the portion of the network where the bursts are delivered (e.g., the RAN) and to do so in a manner that is equally effective in the case of QUIC packets.
[0046] Figure 1 1 is a block diagram schematically depicting an example of a first communication system 10 for discussing embodiments herein and in which embodiments herein may be implemented. The first communication system 10 includes a communication network 100, which may be or be based on an IP network, or in other words, a data communication network based on (such as implemented to support) an Internet Protocol (e.g., version 4 or version 6, i.e., IPv4 or IPv6). The figure shows a client device 120 in the communication network 100, for example, served by it and / or enabled to receive and / or send data traffic in and / or via the communication network 100. The figure also shows a server device 130 (e.g., a host device), which, for example, provides data for communication to / from the client device 120 via the communication network 100 (such as media for streaming, e.g., audio and / or video data). The figure also schematically illustrates a data flow 150 from the server device 130 to the client device 120 in and via the communication network 100. The data flow 150 may include the data provided by the server device 130. Therefore, the client device 120 is an example of a data flow target node and the server device 130 is an example of a data flow source node.
[0047] In addition, communication paths 160a-b are shown between the data flow shaping node 111 and the detection node 110. It should be understood that the data flow shaping node 111 can be such a data shaping node mentioned above, which can exist for forming bursts. The detection node 110 can be a node different from or the same as the data flow shaping node 111.
[0048] Also shown in the figure are remote devices 141 (e.g., remote network nodes) and remote computer networks 140, to which the remote devices 141 may be connected or part. The remote computer networks 140 may correspond to a so-called computer cloud (or simply a cloud), thereby providing certain services. The remote devices 141 and / or the remote networks 140 may, for example, be communicatively connected to the multi-access communication network 100 and, for example, the detector nodes 110 and / or the data stream shaping nodes 111, as shown in the figure.
[0049] Figure 2 2 is a block diagram schematically depicting an example of a second communication system 20 for discussing embodiments of the present invention and in which embodiments of the present invention may be implemented. The second communication system 20 includes a wireless communication network 200 (e.g., a telecommunication network). The wireless communication network 200 may include a radio access network (RAN) 201 portion and a core network (CN) 202 portion. The wireless communication network 200 is typically a telecommunication network or system, such as a cellular communication network supporting at least one radio access technology (RAT) (e.g., LTE or 4G, also known as New Radio (NR) for 5G).
[0050] The wireless communication network 200 comprises network nodes that are communicatively interconnected. These network nodes may be logical and / or physical and located in one or more physical devices. The wireless communication network 200 (typically a RAN 201) may comprise a radio network node 210, i.e. a network node that is or comprises a radio transmitting network node (such as a base station) and / or a network node that is or comprises a control node that controls one or more radio transmitting network nodes. The radio network node may, for example, be communicatively connected, such as being configured to communicate with other radio network nodes (not shown) included in the RAN by or via a so-called X2-U communication interface or communication link.
[0051] Furthermore, the wireless communication network 200 (e.g., CN 202) shown comprises a network node 211 (e.g., a core network node, such as a packet gateway (P-GW) in LTE / 4G or a user plane function (UPF) node in NR / 5G), which may be communicatively connected, such as configured to communicate with a radio network node of the RAN 201 (e.g., with the radio network node 210) by or via a communication interface or communication link (such as the so-called S1-U). The network node 211 shall here correspond to a data shaping node, i.e., a node having (e.g., configured to) transform a data stream it receives into bursts for further transmission. In other words, it may correspond to the data stream shaping node 111, but in the context of a wireless communication network.
[0052] S1-U, X2-U are examples of user plane protocols based on IP / UDP and used in wireless communication networks such as LTE and NR. These user plane protocols can be considered to generally correspond to application layer protocols in terms of IP networks.
[0053] Therefore, the wireless communication network 200 can be considered to be based on and / or include one or more IP networks. For example, the radio network node 210 can be communicatively connected (e.g., via X2-U) to another radio network node in the IP network portion of the RAN 201. In addition, the network node 210 can be communicatively connected (e.g., via S1-U) to the network node 211. The connection is in the IP network of the wireless communication network 200 connecting the RAN 201 and the CN 202.
[0054] The wireless communication network 200 (or specifically, one or more network nodes thereof, e.g., the radio network node 210) is generally configured to serve and / or control and / or manage one or more wireless communication devices (such as a first communication device 220 and a second communication device 221) in a radio coverage area (i.e., an area in which radio coverage is provided for communication with one or more communication devices). The communication device 220 may alternatively be named a wireless communication device (or simply a wireless device), and it may correspond to a user equipment (UE). Each radio coverage may be provided and / or associated with a specific radio access technology (RAT). The radio coverage may be a radio coverage of a radio beam (which may be simply referred to as a beam). Those skilled in the art will recognize that a beam is a more dynamic, relatively narrow and directional radio coverage compared to a conventional cell, and may be implemented by so-called beamforming. A beam is generally used to serve one or several communication devices at the same time, and may be specifically set to serve this one or these several communication devices. A beam may be dynamically changed by beamforming to provide desired coverage to one or more communication devices served by the beam. There may be more than one beam provided by the same network node.
[0055] The wireless communication network (e.g., the CN 202 thereof) may also be communicatively connected (e.g., via the core network node 212) to an external network 240 (e.g., the Internet), and thereby, for example, provide access to the external network 240 for the communication device 220. The external network 240 includes and is connected to other network nodes (e.g., the external network node 230). Here, external refers to the outside relative to the wireless communication network 200. The external network node 230 may, for example, correspond to a server that provides services to one or more other Internet-connected devices, for example, the first communication device 220 may be provided with access to the external network 230 (such as the Internet) via the wireless communication network 200 (e.g., specifically via the core network node 220 mentioned above). Therefore, the first communication device 220 may be communicatively connected to the external network node 230 via the wireless communication network 200 and the external network 240, for example, by means of TCP / UDP / IP and application layer protocols. The data stream 250 (as indicated by the dashed line in the figure) may be provided to the first communication device 220, for example, from the external network node 230 via the external network 240 and the wireless communication network 200. The external network node 230 may be, for example, a server that provides an accessed video streaming service via an application or app executed on the first communication device 220. A similar situation may occur on another or the same communication device that is connected to the Internet via a local area network (LAN) (e.g., a WiFi network at home) rather than the wireless communication network 200 (e.g., 5G) as discussed above.
[0056] Additionally, a communication path 260 is shown from the network node 211 to the first client device and back again.
[0057] and Figure 1 In contrast, the first communication device 220 may be considered to correspond to the client device 120 and thus be an example of a data flow destination node, the wireless communication network 200 (possibly together with the external network 240) may be considered to correspond to the communication network 100, and the network node 211 may be considered to correspond to the data flow shaping node 111 and the probing node 110. At least in the case where the probing node 110 and the data flow shaping node 111 are the same node, the external network node 230 may be considered to correspond to the server device 130, the data flow 250 corresponds to the data flow 150, and the communication path 260 corresponds to the communication paths 160a-b.
[0058] It may be noted that a server (e.g., a device or network node corresponding to server device 130 or external network node 230 or external network 240) may actually correspond to, for example, one or more physical nodes or devices associated with a server through or from which a data stream (e.g., 150 or 250) is provided. The server may alternatively be referred to as, for example, a host computer, a server system, or a communication system.
[0059] Notice, Figure 1 and Figure 2 It should be apparent to those skilled in the art that the drawings are merely schematic and for illustrative purposes, and that not everything shown in the drawings is necessary for all embodiments herein. In addition, communication networks and wireless communication networks corresponding to those shown in the drawings will typically include several other devices, network nodes, and details as understood by those skilled in the art, but are not shown herein for simplicity.
[0060] FIG. 3 schematically illustrates an example of how a data stream can be shaped and sent in a burst, i.e., illustrating what has been described above. Shown is how data 301 of a data stream (e.g., any of the data streams 150, 250) can be sent in a data burst 302a-d. Note that the data 301 in this figure is schematic and can be considered to show the amount of data. In practice, as data is received over time, the data rate will vary. Shaping the data 301 into bursts can be performed, for example, by a data shaping node 111 or a network node 211. As can be seen from this figure, the data in a burst is sent under a transmission period, followed by a silent period until the data transmission of the next burst in time (i.e., the subsequent burst) participates.
[0061] The burst interval (i.e., the time between bursts) may be referred to as the burst interval, and therefore corresponds to a transmission period and a silent period. For example, the burst interval of the first burst 302a shown is a burst interval #1, etc. In the illustrated drawings, the burst intervals between bursts appear the same, and the transmission period, silent period, and burst bit rate during transmission are also the same. However, in practice there may be variations, and the node providing the burst is typically configurable so that the burst interval and / or transmission period and / or silent period and / or bit rate may be changed and varied between bursts as needed.
[0062] QUIC is a reliable protocol in which data packets are confirmed similarly to TCP. However, as already explained above, the problem with QUIC is that it cannot be tracked exactly which packet(s) the confirmation is referencing. However, there are situations in which the RTT of a QUIC stream (i.e., a data stream in which data packets are QUIC packets) can be inferred. If the path between the measurement point (i.e., where the confirmation packets are received, such as at the detection node 110 or the network node 211, and the receiver of the QUIC packet (e.g., the client device 120 or the first communication device 220)) is empty, the time from when a group of packets is sent downstream to the receiver to when another group of packets is observed from the receiver in the reverse direction (i.e., upstream) can be calculated, and the other group of packets can then be assumed to be confirmation packets. This provides information that can be used to estimate latency. This situation and situation "naturally" occurs, for example, when a connection is established during the initial handshake packet exchange and when the connection has been application-limited (which means that the application has stopped sending data to the network for some time). However, this is probably the only naturally occurring situation that can be used "like" for this purpose, but of course it is interesting to be able to estimate the delay in other situations as well. This situation as described may also occur naturally in other situations, but it is usually difficult or impossible to know when it will happen, so it is difficult to take advantage of this.
[0063] Embodiments herein may generally be considered based on manipulating some bursts (which would otherwise be sent like other bursts in a sequence, for example, as discussed in and with respect to FIG. 3 , for example, manipulating burst 302 b) so that it becomes another type of burst that is different from regular bursts and other bursts in the sequence, thereby generating another pattern of acknowledgment packets, which in turn may be identified. In this way, timing information about the data packets of the bursts that generate the acknowledgment packets identified according to the pattern and timing information about these identified acknowledgment packets may be used to estimate latency, for example, in the form of RTT, using the data packet transmission time by a data shaping node (e.g., 111 or 211) and the acknowledgment packet reception time in the same node.
[0064] Further details, illustrations and examples follow.
[0065] When a QUIC endpoint (e.g., corresponding to a data flow target node, such as 120 or 220 in the embodiments herein) receives packets in sequence, it generates an acknowledgment packet for every Nth packet, where N is typically 2 or 10. In other words and in general, an acknowledgment packet is not always sent by the endpoint for each received data packet, but is sent for each Nth data packet received.
[0066] A data shaping node (e.g., 111 or 211) that sends packets in a burst may switch the order of a group of packets in the burst so that a receiving endpoint (e.g., 120 or 220) will generate more acknowledgment packets than it would if the burst were delivered in order.
[0067] For example: consider a burst of 10 packets "in order" [1,2,3,4,5,6,7,8,9,10] and the receiving endpoint sends one acknowledgment for every 10 in-order packets. Under normal circumstances, each burst will result in a single acknowledgment being sent in the opposite direction. However, the data shaping node may define (e.g., be configured to send) a sample burst where the first two packets have the order swapped and the last two packets have the order swapped, i.e., [2,1,3,4,5,6,7,8,10,9]. For an endpoint that follows the principle of sending an immediate acknowledgment upon receiving an out-of-order packet, this sequence will generate four acknowledgments instead of one. After having sent such a sample burst, the data shaping node may then start searching for a corresponding acknowledgment pattern, e.g., listening for the reception of such an acknowledgment pattern. When an acknowledgment packet or a potential acknowledgment packet (as it may not always be certain which packets are acknowledgment packets) is observed (i.e., identified) after the sample burst has been sent, a match window timer may be started. If in turn the expected acknowledgment pattern is observed within the matching window (eg, the expected number of acknowledgment packets are received), it is considered a match and the RTT may be calculated, for example, as the time from sending a burst to receiving a corresponding acknowledgment burst.
[0068] Fig. 4 illustrates this point and schematically illustrates how the grouping for explaining conventional burst can cause the example of confirmation packet.Three bursts 402a-c are shown and can correspond to burst 302a-c, for example.In the example shown, each burst is made up of 4 data packets, burst 402a of data packet 404a-1 ... 404a-4, burst 402b or data packet 404b-1 ... 404b-4 and burst 402c of data packet 404c-1 ... 404c-4.The data packet of each burst is sent with a certain numbering sequence, and the endpoint expects to receive the data packet in this order.In the example shown, just for illustrating principle, every 4th data packet generates an confirmation packet. That is, the 4th data packet of burst 402a (i.e., data packet 404a-4) generates confirmation packet 406a, the 4th data packet of burst 402b (i.e., data packet 404b-4) generates confirmation packet 406b, and the 4th data packet of burst 402c (i.e., data packet 404c-4) generates confirmation packet 406c. In order to ensure that each burst can be fully confirmed, the data flow shaping node (e.g., 111 or 211) providing the burst is usually configured so that each burst consists of an integer N data packets, where N is the number of data packets per confirmation packet. The problem with the confirmation packet sequence as shown in Figure 4 is that it may be difficult to identify whether a packet is a confirmation packet, although this can be assumed in some cases, but at least it can be identified whether a packet is a potential confirmation packet. However, even if a packet can be identified as a confirmation packet or a potential such packet, it is still a problem to identify which burst and / or burst of packets (in response to which such (potential) confirmation packets have been sent) is still a problem, which is what delay estimation and, for example, RTT calculations expect to know. For example, although the example is simplified in Figure 4, it is in principle impossible (e.g., in the case of QUIC packets) to know whether acknowledgment packet 406b was sent in response to a data packet of burst 402a or 402b, and it is impossible to know which specific data packet generated it.
[0069] When packets are received out of order (i.e., in an incorrect sequence), a QUIC endpoint will send an immediate ACK so that the sender gets faster feedback, which can help with potential loss recovery. This is also true for cases other than QUIC (i.e., when every Nth packet should normally generate an ACK). In this case, there is also typically a counter reset, and after that counter reset, another N packets in the correct sequence are required to generate an ACK packet.
[0070] This can be used in the implementation of the embodiments herein and is exemplified next.
[0071] Figure 5An example is schematically illustrated to illustrate how a sample burst with reordered packets according to some embodiments herein may result in a particular pattern of acknowledgment packets. Figure 5 The example of can be compared with the example in FIG. 4 and can result from the manipulation or formation of burst 402b with a burst having a different data packet than that burst, while the other burst can remain the same as in FIG. 4. Thus, in Figure 5 4, bursts 502a, 502c may correspond to bursts 402a, 402c, data packets 504a-1, 2, 3, 4 may correspond to data packets 402a-1 ... 4, data packets 504c-1, 2, 3, 4 may correspond to data packets 402c-1 ... 4, and acknowledgment packets 506a, 506c may correspond to acknowledgment packets 406a, 406c. To illustrate that the data packets of the data bursts 502a, 502c are sent in a certain order of numbering as is conventional, their respective ordered sequences are written in the figure, i.e., 1, 2, 3, 4. Thus, similar to FIG. 4, the data packets 504a-4 of the sequence generate acknowledgment packets 506a, etc.
[0072] However, as mentioned, there is another burst, sample burst 503 or 502b', which can be considered as a manipulated version of burst 402b (i.e., a normal or regular burst). In the example shown, the manipulation can involve a purposeful reordering of the order of data packets of the burst that would otherwise be sent conventionally, and thus reorder or change the order in which they are sent in the example of Figure 4, and also compared with the order of data packets in the previous and subsequent bursts 502a, 502c. In other words, instead of sending the data packets in the correct order, the data packets are purposefully sent in an incorrect order. If it is assumed and as explained above that the correct order is 1, 2, 3, 4 of the data packets of the burst, the data packets of the sample burst 503 are instead in an incorrect order or sequence 2, 1, 4, 3, which is why these data packets are named data packets 504b-2, 1, 4, 3 in the figure. In this example, this has been used here to force the endpoint to send an acknowledgment packet for each data packet of the sample burst 503. Since each data packet (when sent in reordering 2, 1, 4, 3) will be viewed by the endpoint (e.g., any of 120, 220) as being received in an incorrect order, each data packet will generate an acknowledgment packet. In other words and as shown in the figure, instead of a single acknowledgment packet 406b as in FIG. 4, four acknowledgment packets 506b-1 ... 506b-4 are now sent. It will be appreciated that a distinguishable pattern of acknowledgment patterns is thereby formed compared to the acknowledgment patterns sent in response to previous and subsequent bursts. This makes it possible to identify these acknowledgment packets as being generated in response to the sample burst 503, and also to identify that the first of these acknowledgment packets is an acknowledgment packet sent in response to the first data packet of the burst.
[0073] Also marked in the figure are sampling windows 508 and matching windows 510. Such windows are used in some embodiments of the present invention to facilitate implementation. Below is a more detailed description, but in short, the idea is that the sampling window 508 corresponds to a time range or interval associated with the transmission of the sample burst 503, during which potential confirmation packets (if any) expected to be sent in response to the sample burst will be received. The matching window is one or more windows within the sampling window and starts when each potential confirmation packet is received. Their width (i.e., duration) is set based on the frequency and number of confirmation packets that should be sent in response to the sample burst, that is, this specifically depends on the reordering method of the data packets. The idea is that if a specific number of potential confirmation packets are received within the matching window, this can be used to identify the confirmation packets of the expected pattern in response to the sample burst, that is, the occurrence of the confirmation packets sent in response to the sample burst. Although many different ways can be used to identify the expected pattern, as recognized by those skilled in the art, it will be advantageous if counting a specific number of packets during the time interval is sufficient to identify the pattern. Thus, in the example shown, the matching window can be set so that if 4 potential acknowledgment packets are received during a certain time interval (e.g., based on the time used to send 4 data packets), this can be considered to identify these potential acknowledgment packets as acknowledgments sent in response to the sample burst.
[0074] Figure 6 A simplified example is schematically illustrated to further illustrate how sampling windows and matching windows may be related to each other and used in embodiments herein. Here, a sampling window 608 is shown, and two matching windows 610a-b of equal size are shown therein (i.e., within a time range corresponding to the sampling window). Each matching window begins when a potential confirmation packet is received or a received potential confirmation packet is identified. In the figure, a potential confirmation packet is indicated by an x (i.e., in correspondence with the sampling window). Figure 5). The sampling window and the matching window usually have a predefined or predetermined length and may be predefined or predetermined based on one or more burst transmission cycles used, the burst interval used, and the sample burst for generating the confirmation packet pattern to be identified. In the example shown, there may be burst data packets of a burst sequence, so that each normal burst should only generate a single confirmation packet within the sampling window 610, and the sample burst (e.g., formed by the reordered packets) should generate at least two confirmation packets within the matching window. Therefore, as shown in the example, only a single potential confirmation packet is received within the matching window 610a (i.e., until the matching window "expires"), so this is not enough to conclude that the pattern has appeared (i.e., for identifying the pattern). However, within the matching window 610b, it is identified that two potential confirmation packets are received, and the conclusion is that the pattern has appeared (i.e., the expected pattern has been identified). Since it is known from how the sampling burst is formed which data packet of the sampling burst generates the first confirmation packet of the pattern, for example, the sending / transmission time of the data packet and the receiving time of the first confirmation packet of the pattern can be used to estimate the delay. If it is the same network node that sent the sample burst (eg, a data shaping node, such as 111 or 211), it also receives the acknowledgment packet, and the difference between the sending time and the receiving time corresponds to the RTT.
[0075] Figure 7 Flowchart for schematically illustrating an exemplary method according to some embodiments of the present invention. The method is for delay estimation on a communication path (e.g., 160a-b or 260) in a communication network (e.g., 100 or 200). The communication path includes (i.e., in the path in which it is located) a data flow target node (e.g., 120 or 220), which is configured to receive a data flow (e.g., 150 or 250) from a data flow source node (e.g., 130 or 230) via the communication path. The data flow shaping node is configured to provide the data flow to the data flow target node in a burst.
[0076] Note that the actions discussed below may be performed in any suitable order and / or may be performed with full or partial overlap in time where possible and appropriate. It should also be noted that not all actions and features discussed below are necessarily present in all embodiments.
[0077] Action 701
[0078] The data flow shaping node sends bursts of samples to result in a particular pattern of acknowledgment packets identifiable by a particular number of acknowledgment packets received during a matching window.
[0079] The sample burst and matching window may be as discussed above. The sample burst is part of a sequence of bursts sent by the data shaping node and may be a nominal or regular burst of the sequence, but wherein the data shaping node has changed the way in which the sample burst is composed of data packets, for example, by changing the order of two or more data packets and / or removing packets of the burst. Thus, the sample burst differs from previous and subsequent bursts of the sequence of bursts in a predefined manner. The sample burst is formed so that the data flow target node sends a confirmation packet of the specific pattern in response to the sample burst, and therefore, the specific pattern is different from the pattern of confirmation packets sent by the data flow target node in response to the previous and subsequent bursts.
[0080] Action 702
[0081] Starting (e.g., opening) a sampling window (such as described above), which may involve starting a sampling window timer (i.e., a timer associated with the sampling window), which will run for a period of time corresponding to the length of the sampling window. The sampling window may correspond to such a sampling window as described above. This action is typically performed by the data shaping node.
[0082] Action 703
[0083] During the sampling window, i.e., while the window is open, such as before a timer associated with the sampling window has expired, at least the reception of a potential acknowledgment packet (e.g., as received by the data shaping node) is listened for. Preferably, such listening may be performed by the same node (e.g., the data shaping node) that receives the acknowledgment packet. At some point in time, a potential acknowledgment packet is received, and such reception is identified, e.g., by the data shaping node.
[0084] Action 704
[0085] After receiving the potential ACK packet in action 703 (i.e., in response to the receipt of the potential ACK packet), starting (e.g., opening) a matching window (such as described above) may involve starting another timer that will run for a period of time corresponding to the length of the matching window. The matching window may correspond to the matching window described above and therefore have a specific number of ACK packets associated therewith that may be predetermined and / or determined by the sample burst.
[0086] Action 705
[0087] During the matching window, i.e., while the window is open, such as before a timer associated with the matching window has expired, the number of potential acknowledgement packets received is counted and a check is made whether a specific number of potential acknowledgement packets are received within the matching window. The specific number is predefined or predetermined and is, for example, based on the specific pattern of acknowledgement patterns and also based on the sample burst. The specific number is typically predefined or predetermined together with the matching window so that the specific number of acknowledgement packets received within the matching window will be the case for the acknowledgement packet pattern, while fewer acknowledgement packets (e.g., only one) are received within the matching window when acknowledgement packets are sent in response to a "normal" burst of a burst sequence (e.g., resulting from the preceding and following bursts rather than the sample burst).
[0088] Action 706
[0089] If the specific number of acknowledgment packets are received during the matching window, then there is a pattern match and the delay can be estimated (e.g., RTT is calculated, as described above). The method can then end, but can be performed again for another later burst of samples. In this way, the delay can be repeatedly estimated.
[0090] If the specific number of acknowledgment packets is not reached during the matching window and the sampling window has not expired and a new potential acknowledgment packet is received, a new matching window is started, action 704 is performed again, and so on.
[0091] Action 707
[0092] If there is no pattern match when the sampling window expires, it can be concluded that the pattern match has failed and no delay estimation is completed.
[0093] In some embodiments, information from such a "failure" (especially if it occurs more than once with the same settings) can be used as input to change or adapt the sample burst, sampling window, and / or matching window. In this way, different settings can be tested and tried until a suitable sample burst, sampling window, and matching window (including the number of acknowledgment packets to be received within the matching window) are applied (i.e., settings that result in a pattern match and can be used for delay estimation).
[0094] Figure 8is a flow chart schematically illustrating an embodiment of a method according to an embodiment of the present invention. The method is used to support the estimation of delay on a communication path (e.g., 160a-b or 260) between a data flow shaping node (e.g., 111; 211) and a detection node in a communication network (e.g., 100 or 200). As already explained above, the detection node can in principle be any node in the communication path (e.g., the detection node 110), and in some embodiments can correspond to the radio network node 210, or the network node 211 and / or the data flow shaping node 111. The communication path (e.g., 160a-b; 260) includes a data flow target node (e.g., 120; 220), which is configured to receive a data flow (e.g., 150; 250) from a data flow source node (e.g., 130; 230) via the communication path. The data flow shaping node (e.g., 111 or network node 211) is configured to provide the data flow (e.g., 150; 250) to the data flow target node (e.g., 120; 220) in a burst. As should be appreciated, a burst is transmitted during a burst interval associated with the burst, the burst interval having a transmission period of data transmission and a subsequent silent period of no data transmission.
[0095] The method may be executed by one or more devices, for example, by one or more devices of the first communication system 10 or the second communication system 20 , and / or the data flow shaping node 111 or the network node 211 , and / or the remote device 141 and / or the remote computer network 140 .
[0096] Note that the following actions may be performed in any suitable order and / or may be performed completely or partially overlapping in time where possible and appropriate.
[0097] Action 801
[0098] The one or more devices initiate sending of a sample burst (e.g., 503) by a data flow shaping node (e.g., 111; 211), the sample burst being a burst among the bursts that differs from previous and subsequent bursts (e.g., 502a, 502c) in a predefined manner, so that a data flow target node (e.g., 120; 220) sends a confirmation packet of a specific pattern in response to the sample burst, the specific pattern being different from a pattern of confirmation packets sent by the data flow target node (e.g., 120; 220) in response to the previous and subsequent bursts.
[0099] As will be appreciated, the previous and subsequent bursts are bursts of the data stream and are sent before and after (usually immediately before and after) the sample burst, respectively. Existing burst generating data stream shaping nodes can, for example, be controlled so that one or more bursts are sent, each corresponding to such a sample burst as above. How many packets are needed to generate confirmation packets can be predetermined or learned (e.g., by exchanging the order of packets and / or changing the burst size). This knowledge can be used together with the assumption as described above (i.e., each data packet received in an incorrect order generates a confirmation packet) to form the specific pattern. However, an alternative (although not preferred) way to produce the specific pattern can be to remove some packets and / or exchange some packets with packets of other content (e.g., virtual packets). Removing packets means that the data stream source node may have to resend the packets, which is usually undesirable, although other solutions can be considered first, but this is still acceptable in some cases. Replacing packets with virtual packets is usually undesirable because in principle its effect may be the same as removing packets and has additional negative effects (also resulting in other and unnecessary processing at the data stream target node). Phantom packets may also not result in an "extra" acknowledgment packet being sent as would be the case in the reordering case, as the phantom packets may be dropped before this can happen (e.g., when managed by QUIC's transport layer).
[0100] In some embodiments, the difference of the sample burst is that at least two data packets of the sample burst are in an incorrect order. This can be achieved by changing the order of these data packets when the data packets are to be sent as part of the sample burst. Changing the order to an incorrect order has the ability to generate confirmation packets, and therefore can be used to form the pattern while still sending all packets of the burst so that they can be used by the data flow target node. Changing the order is also a relatively simple operation, and therefore is conducive to the implementation of the embodiments herein.
[0101] The incorrect order can be achieved by exchanging the positions of two packets. For implementation reasons, exchanging the order may be preferred and easy to extend, each exchange enabling the mandatory generation of two confirmation packets. The two packets that exchange positions may be packets that appear consecutively. This may be advantageous, for implementation reasons and because it can be used to generate a useful pattern even when confirmation packets are normally sent (as frequently as for every second received packet), i.e., this is a general and flexible solution leading to the specific pattern.
[0102] The at least two packets may comprise the first transmitted packet of the sample burst. Typically, the first packet of a burst does not generate an acknowledgement packet, which makes it advantageous if the specific pattern can thereby comprise an acknowledgement packet generated by the first packet. It will generally increase the likelihood of identifiable different patterns of the sample burst.
[0103] The at least two packets may include the last transmitted packet of the sample burst. Typically, each burst consists of an integer multiple of the number of packets required to generate an acknowledgement packet, and the acknowledgement counter may be "reset" by the last packet of the burst. However, when different modes of acknowledgement packets have been enforced by embodiments of the present invention (i.e., by incorrectly ordered packets of the sample burst), the acknowledgement packet counter may not be reset by the last packet, which in turn will affect the acknowledgement packet pattern of subsequent bursts. However, since the last packet of the sample burst is involved here, it should generate an acknowledgement packet in any case, which can be avoided, and thus it can also become easier to identify different modes of the sample burst.
[0104] Action 802
[0105] The one or more devices initiate reception of potential acknowledgment packets identified by a probing node (eg, 110; 211) according to the particular pattern.
[0106] By identifying the specific pattern in potential ACK packets, these packets are likely to be actual ACK packets, and the risk of mixing ACK packets between bursts can also be avoided. It is even possible to find out which exact data packet of a sample burst generated a certain ACK packet of this pattern. This means that the delay can be estimated. Therefore, the embodiments of the present invention support the estimation of delay, which also works when the data packet is a QUIC packet without a spin bit. Note that when the probing node is a data shaping node, the delay on the communication path corresponds to the RTT.
[0107] Another advantage of embodiments herein is that the sample burst itself (i.e. above the packet level, e.g. its data rate, length, etc.) need not be affected or changed compared to a regular burst. Thus, a sample burst may appear identical to previous and subsequent bursts, since only, e.g., the order of the data packets may have changed. This means that undesirable shaping may be avoided when implementing embodiments herein.
[0108] In some embodiments, identifying a potential acknowledgment packet according to the particular pattern is based on receiving a potential acknowledgment packet by the probing node during a sampling window, e.g., as discussed above. The sampling window may correspond to a first time interval that is predefined and covers a time period during which a potential acknowledgment packet from a sample burst can be expected. The sampling window may correspond to a burst interval associated with the sample burst, or a time interval that begins at the start of a sample burst (e.g., at a time when a first acknowledgment packet of the sample burst can be expected to be received by the probing node) and ends at a time before an acknowledgment packet of a nearest subsequent burst can be expected to be received by the probing node.
[0109] In addition, according to the specific pattern recognition confirmation packet can be based on identifying a specific number of confirmation packets in a matching window, for example, as discussed above. The matching window is a time window with a predefined duration and starts from receiving each confirmation packet in a sampling window. Since the sample burst is formed as described above (for example, by packets in an incorrect order, such as by switching packets), this can be used to ensure that compared with the confirmation packets generated by a normal burst (wherein data packets are sent in the correct order, such as previous and subsequent bursts), the specific pattern will include more confirmation packets per time unit. This situation is utilized here, because the matching window can therefore be set to a suitable time period, during which only the sample burst should result in the confirmation packets of the specific number. Therefore, the sample burst can be identified by receiving the confirmation packets of the specific number during the matching window. This enables the implementation of the embodiments of this article to be relatively simple and can be quickly executed, because for example a counter can be used to implement, rather than any more advanced and usually heavy pattern recognition algorithm in terms of calculation.
[0110] Action 803
[0111] The one or more devices may initiate determining an estimate of the delay based on timing information regarding transmission of the sample burst (eg, 503) and receipt of the identified potential acknowledgment packet by the probing node.
[0112] The estimate of latency may be based on when the probing node receives the one or more identified potential acknowledgment packets and when the data flow shaping node sends one or more packets that lead to a sample burst of the one or more identified potential acknowledgment packets.
[0113] As used herein, initiating one or more devices to perform an action (e.g., sending) refers to any pre-action and / or sub-action that causes or causes the action to be performed. Initiation (i.e., pre-action and / or sub-action) can be performed by the device itself that performs the action or by another entity. The device itself that changes from not performing an action to actively performing an action will include some initiation of performing the action, or in other words, in this case (i.e., when the device itself performs the initiated action) includes initiation to perform the action. Initiating an action performed by another entity rather than performing the action itself can correspond to some signal (e.g., trigger signal) provided by another entity or an action performed, which causes the device to perform the action. As mentioned, in some embodiments, the method and / or some or all of its actions are performed, for example, by a data stream shaping node 111 or a network node 211 and / or a remote device 141 and / or a remote computer network 140 (e.g., a computer cloud), and the embodiments herein can then be initiated or provided as a cloud service, for example, at least in part. If the method and / or some or all of its actions are performed by, for example, one or more node parts of the communication network 100 or the wireless communication network 200 (for example, a node corresponding to the data flow shaping node 111 or the network node 211), the node that performs the corresponding action typically also performs the initiated action, for example, the data flow shaping node 111 or the network node 211 can identify the reception of a potential confirmation packet in action 802 and / or determine an estimate of the delay in action 803.
[0114] Fig. 9 is used to illustrate one or more devices 900 (for example, the above combined Figure 8 How the one or more devices discussed above can be configured to perform the above combined Figure 8 Schematic block diagram of an embodiment of the discussed methods and actions.
[0115] Thus, one or more devices 900 are for supporting the estimation of delay on the communication path (e.g., 160a-b; 260) between a data flow shaping node (e.g., 111; 211) and a detection node (e.g., 110; 210; 111; 211) in a communication network (e.g., 100; 200). The communication path includes a data flow target node (e.g., 120; 220), which is configured to receive a data flow (e.g., 150; 250) from a data flow source node (e.g., 130; 230) via the communication path. The data flow shaping node is configured to provide the data flow to the data flow target node in bursts.
[0116] The one or more devices 900 may include one or more processing modules 901, such as components, one or more hardware modules (including, for example, one or more processors), and / or one or more software modules for performing the methods and / or actions.
[0117] One or more devices 900 may also include a memory 902, which may include (such as containing or storing) a computer program 903. The computer program 903 includes "instructions" or "codes" that can be directly or indirectly executed by one or more devices 900 to perform the methods and / or actions. The memory 902 may include one or more memory units and may be further configured to store data, such as configurations and / or applications related to or used to perform the functions and actions of the embodiments herein.
[0118] In addition, one or more devices 900 may include one or more processors 904, i.e., one or more processors, such as one or more exemplary hardware modules and may include or correspond to one or more processing circuits. In some embodiments, the processing module 901 may include the processor 904, for example, "embodied in the form of the processor 904" or "implemented by the processor 904". In these embodiments, the memory 902 may include a computer program 903 executable by the processor 904, whereby the one or more devices 900 may be operated or configured to perform the method and / or its actions.
[0119] Typically, one or more devices 900 (e.g., processing module 901) include one or more input / output (I / O) modules 905 that are configured to participate in (e.g., by executing) any communications to and / or from other units and / or devices, such as sending information to and / or receiving information from other devices, for example, receiving from server device 130 and sending to client device 120. If applicable, one or more I / O modules 905 can be instantiated by obtaining (e.g., receiving) modules and / or providing (e.g., sending) modules.
[0120] In addition, in some embodiments, one or more devices 900 (e.g., processing module 901) include one or more startup modules, one or more identification modules, and one or more determination modules, such as one or more exemplary hardware and / or software modules for performing the actions of the embodiments of this document. These modules can be implemented in whole or in part by processor 904.
[0121] therefore:
[0122] One or more devices 900, and / or one or more processing modules 901, and / or one or more processors 904, and / or one or more I / O modules 905, and / or one or more initiation modules, and / or one or more sending modules may be operable or configured to initiate sending of the sample burst by the data stream shaping node.
[0123] In addition, one or more devices 900, and / or one or more processing modules 901, and / or one or more processors 904, and / or one or more I / O modules 905, and / or one or more initiation modules, and / or one or more identification modules may be operable or configured to initiate the reception of the potential confirmation packet identified by the detection node according to the specific pattern.
[0124] In addition, one or more devices 900, and / or one or more processing modules 901, and / or one or more processors 904, and / or one or more I / O modules 905, and / or one or more startup modules, and / or one or more determination modules may also be operable or configured to initiate determining the estimate of the delay based on the timing information.
[0125] Fig.10It is a schematic diagram illustrating some embodiments related to one or more computer programs and carriers thereof that enable the device 900 discussed above to perform the method and related actions. The computer program may be a computer program 903 and includes instructions that enable the device 900 to be performed as described above when executed by a processor 904 and / or a processing module 901. In some embodiments, a carrier is provided, or more specifically a data carrier (e.g., a computer program product) including a computer program is provided. The carrier may be one of an electrical signal, an optical signal, a radio signal, and one or more computer-readable storage media (e.g., one or more computer-readable storage media 1001 schematically illustrated in the accompanying drawings). Therefore, the computer program 903 may be stored on a computer-readable storage medium 1001. The carrier may exclude a temporary propagation signal, and the data carrier may be named as a non-temporary data carrier accordingly. Non-limiting examples of data carriers as one or more computer-readable storage media are memory cards or memory sticks, one or more disk storage media such as CDs or DVDs, or mass storage devices such as hard disk drives or solid-state drives (SSDs). One or more computer-readable storage media 1001 can be used to store data that can be accessed through a computer network 1002 (e.g., the Internet or a local area network (LAN)). Computer program 903 can also be provided as a pure computer program or included in one or more files. One or more files can be stored on one or more computer-readable storage media 1001, and for example, can be obtained by downloading, for example, via a server on the computer network 1002 shown in the figure. The server can be, for example, a Web or file transfer protocol (FTP) server. One or more files can be, for example, executable files, for downloading directly or indirectly to the device 900 and executing thereon so that it is executed as described above (e.g., by being executed by processor 904). One or more files can also or alternatively be used for intermediate downloading and compiling involving the same or another one or more processors so that they can be executed before further downloading and execution, so that the device 900 is executed as described above.
[0126] Note that any of the processing modules and circuits mentioned above may be implemented as software and / or hardware modules, for example, in existing hardware and / or as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), etc. It should also be noted that any of the hardware modules and / or circuits mentioned above may be included in a single ASIC or FPGA, for example, or distributed among several separate hardware components, whether packaged individually or assembled into a system on a chip (SoC).
[0127] Those skilled in the art will also understand that the modules and circuit systems discussed herein may refer to a combination of hardware modules, software modules, analog and digital circuits, and / or one or more processors configured with software and / or firmware (e.g., stored in a memory), which when executed by one or more processors can cause nodes and devices to be configured as and / or perform the above-described methods and actions.
[0128] The identification / recognition by any identifier herein may be implicit or explicit. The identification / recognition may be unique in a specific context (eg, in a wireless communication network or at least in a relevant part or area thereof).
[0129] As used herein, the term "network node" or simply "node" may refer to any type of node that can communicate with another node in a communication network (e.g., an IP network or a wireless communication network) and is included in the communication network. In addition, such a node may be a radio network node (as described below) or any network node (which, for example, can communicate with a radio network node) or be included therein. Examples of such network nodes include any radio network node, a core network node, operations and maintenance (O&M), an operations support system (OSS), a self-organizing network (SON) node, etc.
[0130] The term "radio network node" as may be used herein may refer to any type of network node for serving wireless communication devices (e.g., so-called user equipment or UE), and / or any radio node that is connected to other network nodes or network elements or from which wireless communication devices receive signals. Examples of radio network nodes are Node B, Base Station (BS), Multi-Standard Radio (MSR) node (such as MSR BS), eNB, eNodeB, gNB, network controller, RNC, Base Station Controller (BSC), repeater, donor node controlled repeater, base transceiver station (BTS), access point (AP), new radio (NR) node, transmission point, transmission node, node in a distributed antenna system (DAS), etc.
[0131] As may be used herein, each of the terms "wireless communication device", "user equipment" and "UE" may itself refer to any type of wireless device that is arranged to communicate with a radio network node in a wireless, cellular and / or mobile communication system, and may therefore be referred to as a wireless communication device. Examples include: a target device, a device-to-device UE, a device for machine type communication (MTC), a machine type UE or a UE capable of machine-to-machine (M2M) communication, a personal digital assistant (PDA), a tablet computer, a mobile device, a terminal, a smart phone, a notebook embedded device (LEE), a notebook mounted device (LME), a universal serial bus (USB) dongle, etc.
[0132] Although some terms are frequently used herein for convenience or in the context of examples involving certain other (eg, 3GPP or other related standard) nomenclatures, it must be understood that such terminology itself is non-limiting.
[0133] It should also be noted that although the terminology used herein may be particularly associated with and / or exemplified by particular communication systems or networks, this should not be viewed as limiting the scope of the embodiments herein to only these particular systems or networks or the like.
[0134] As used herein, the term "memory" may refer to a data storage device for storing digital information, typically a hard disk, magnetic storage, medium, portable computer floppy disk or disk, flash memory, random access memory (RAM), etc. In addition, the memory may be an internal register of a processor.
[0135] It should also be noted that any enumeration terms such as first device or node, second device or node, first base station, second base station, etc. should be considered non-limiting in themselves, and the terms themselves do not imply a certain hierarchical relationship. In the absence of any explicit information to the contrary, the enumeration naming should be regarded as just a way to implement different names.
[0136] As used herein, the expression "configured to" may mean that the processing circuit is configured to or adapted to perform (by means of software or hardware configuration) one or more of the actions described herein.
[0137] As used herein, the term "number" or "value" may refer to any kind of number, for example, binary, real, imaginary, or rational numbers, etc. In addition, a "number" or "value" may be one or more characters, for example, a letter or a string of letters. In addition, a "number" or "value" may be represented by a bit string.
[0138] As used herein, the expressions "may" and "in some embodiments" are generally used to illustrate that the features described can be combined with any other embodiments disclosed herein.
[0139] In the drawings, features that may be present in only some embodiments are generally drawn using dotted or dashed lines.
[0140] As used herein, the expressions "transmit" and "send" are generally interchangeable. These expressions may include transmissions by broadcast, unicast, multicast, etc. In this context, a transmission by broadcast may be received and decoded by any authorized device within range. In the case of unicast, one specifically addressed device may receive the transmission and encode it. In the case of groupcast (e.g., multicast), a specifically addressed group of devices may receive the transmission and decode it.
[0141] When the words “comprising” or “including” are used, they should be interpreted in a non-limiting sense, ie, meaning “consisting at least of…”
[0142] The embodiments herein are not limited to the preferred embodiments described above. Various substitutions, modifications and equivalents may be used. Therefore, the embodiments described above should not be considered to limit the scope of the present disclosure defined by the appended claims.
Claims
1. A method for supporting the estimation of a delay on a communication path (160a-b; 260) between a data flow shaping node (111; 211) and a detection node (110; 210; 111; 211) in a communication network (100; 200) performed by one or more devices (111; 140; 141; 211; 600), the communication path (160a-b; 260) comprising a data flow target node (120; 220) configured to receive a data flow (150; 250) from a data flow source node (130; 230) via the communication path (160a-b; 260), the data flow shaping node (111; 211) being configured to send a data flow to the data flow target node (120; 220) in a burst. 220) provides the data stream (150; 250), wherein The method comprises: - initiating (801) sending of a sample burst (503) by the data flow shaping node (111; 211), the sample burst (503) being a burst among the bursts that is different from the preceding and following bursts (502a, 502c), so that the data flow target node (120; 220) sends a specific pattern of acknowledgment packets to the probing node (110; 210; 111; 211) in response to the sample burst (503), the specific pattern being different compared to the pattern of acknowledgment packets sent by the data flow target node (120; 220) in response to the preceding and following bursts, wherein the sample burst (503) is different in that at least two data packets of the sample burst (503) are in an incorrect sequence to form the specific pattern, or wherein the sample burst is different in that packets of the sample burst are removed to form the specific pattern, - initiating (802) the reception of a potential acknowledgment packet identified by said detection node (110; 210; 111; 211) according to said specific pattern, and - initiating (803) determining an estimate of said delay based on timing information regarding the transmission of said sample burst (503) and the reception of said identified potential acknowledgment packet by said probing node (110; 210; 111; 211), wherein the estimation of the delay is based on when the probing node (110; 210; 111; 211) receives one or more of the identified potential confirmation packets and when the data flow shaping node (111; 211) sends one or more packets of the sample burst (503) that result in the one or more identified potential confirmation packets.
2. The method according to claim 1, wherein: The incorrect sequence is achieved by swapping the positions of two packets.
3. The method according to claim 2, wherein: The two packets that exchange positions are packets that appear consecutively.
4. The method according to any one of claims 1 to 3, wherein: The at least two packets include a first transmitted packet of the sample burst (508; 502b').
5. The method according to any one of claims 1 to 4, wherein: The at least two packets include a last transmitted packet of the burst of samples (508; 502b').
6. One or more devices (111; 140; 141; 211; 600) for supporting the estimation of delay on a communication path (160a-b; 260) between a data flow shaping node (111; 211) and a detection node (110; 210; 111; 211) in a communication network (100; 200), wherein: The communication path (160a-b; 260) includes a data flow target node (120; 220) configured to receive a data flow (150; 250) from a data flow source node (130; 230) via the communication path (160a-b; 260), the data flow shaping node (111; 211) is configured to provide the data stream (150; 250) to the data stream destination node (120; 220) in a burst, wherein the one or more devices are configured to: initiating (801) sending of a sample burst (503) by the data flow shaping node (111; 211), the sample burst (503) being a burst among the bursts that is different from the preceding and following bursts (502a, 502c) so that the data flow target node (120; 220) sends a specific pattern of acknowledgment packets to the probing node (110; 210; 111; 211) in response to the sample burst (503), the specific pattern being different compared to the pattern of acknowledgment packets sent by the data flow target node (120; 220) in response to the preceding and following bursts, wherein the sample burst (503) is different in that at least two data packets of the sample burst (503) are in an incorrect sequence to form the specific pattern, or wherein the sample burst is different in that packets of the sample burst are removed to form the specific pattern, initiating (802) the receipt of a potential acknowledgment packet identified by the detection node (110; 210; 111; 211) according to the specific pattern, and initiating (803) determining an estimate of said delay based on timing information regarding transmission of said sample burst (503) and receipt of said identified potential acknowledgment packet by said probing node (110; 210; 111; 211), wherein the estimation of the delay is based on when the probing node (110; 210; 111; 211) receives one or more of the identified potential confirmation packets and when the data flow shaping node (111; 211) sends one or more packets of the sample burst (503) that result in the one or more identified potential confirmation packets.
7. One or more devices according to claim 6, wherein: The incorrect sequence is achieved by swapping the positions of two packets.
8. One or more devices according to claim 7, wherein: The two packets that exchange positions are packets that appear consecutively.
9. One or more devices according to any one of claims 6-8, wherein: The at least two packets include a first transmitted packet of the sample burst (508; 502b').
10. One or more devices according to any one of claims 6-9, wherein: The at least two packets include a last transmitted packet of the burst of samples (508; 502b').
