Method and device for communication in a wireless communication network

Abstract

Method and device for communication in a wireless communication network This method of communication in a wireless communication network, implemented by a processor of a communication node, comprises, for a plurality of packets to be transmitted to a destination node, for a predetermined application: a reception of information (56, 58) relating to a quality of communication in the network and information relating to an application performance indicator, a determination (62, 64, 66) of a plurality of disjoint transmission windows, and calculation, for each transmission window, of an associated central time instant and an associated window width, a classification (68) of the transmission windows, in quality of service classes, according to a statistical distribution of the latency and the application performance indicator, and a distribution (70) of the packets to be transmitted in at least one of the transmission windows,the distribution taking into account at least one constraint associated with said predetermined application. Figure for the abstract: Figure 3,

Description

Title of the invention: Method and device for communication in a wireless communication network

[0001] The present invention relates to a method of communication in a wireless communication network.

[0002] The invention also relates to an associated device and computer program.

[0003] The invention relates more generally to the field of wireless telecommunications, and more particularly to the field of fifth generation (5G) telecommunications systems.

[0004] One of the main objectives of 5G wireless communication technologies is to increase the capacity and communications density of cellular networks, while improving reliability and reducing end-to-end communication latency between a transmitting device and a receiving device, each of the devices being equipped with suitable communication interfaces.

[0005] Ultra-reliable and low-latency communication, known by the acronym URLLC (for "Ultra-reliable and low-latency communication") represents a category of services covered by 5G systems, and has numerous applications in areas where reliability and latency are critical, for example industrial automation, remote surgery applications, augmented reality, intelligent transport systems.

[0006] The approaches previously proposed to achieve the performance levels required to perform URLLC communication, and in particular to control latency, are divided into two categories: proactive approaches and reactive approaches.

[0007] Proactive approaches reserve resources and select communication mechanisms, to be implemented by each communication node of a wireless communication network, so as to ensure the required quality of service (e.g. latency), including in the worst cases of possible degradation of transmission conditions, for example in the presence of a very noisy transmission channel or one with interference. Such approaches induce over-consumption of resources and communication mechanisms, the worst cases envisaged being able to occur very rarely in practice.

[0008] Reactive approaches propose the activation of additional resources or additional communication mechanisms (e.g., packet resending) in an adaptive manner, in the event of degradation of transmission conditions, for example the adaptation of transmission parameters based on a prediction of the transmission channel. Such approaches can induce high latency, and also require significant computational resources for calculating predictions.

[0009] The inventors have proposed, in patent application FR2103542, an improved method and device capable of jointly reducing latency and improving the reliability of a wireless communication network while maintaining appropriate efficiency.

[0010] All known approaches to latency control, for example minimizing jitter (i.e. the variation of latency over time), implement optimizations dependent on the communication channel.

[0011] The inventors have developed a different approach, which allows better control and reduction of the necessary computing resources, while satisfying a predetermined quality target.

[0012] To this end, the invention proposes, according to one aspect, a method of communication in a wireless communication network comprising a plurality of communication nodes, implemented by a processor of a communication node, comprising a reception of a plurality of packets to be transmitted to a destination node, for a predetermined application. This method comprises steps of:

[0013] - reception of information relating to a quality of communication in the network of wireless communication, including transmission latency data, and receiving information relating to an application performance indicator,

[0014] -determination of a plurality of disjoint transmission windows, and calculation, for each transmission window, of an associated central time instant and an associated window width,

[0015] - classification of transmission windows, into quality of service classes, in function of a statistical distribution of latency and the application performance indicator,

[0016] - distribution of the packets to be transmitted to carry out a sending of each packet in at least one of the transmission windows, the distribution taking into account at least one constraint associated with said predetermined application.

[0017] Advantageously, in the proposed method, the distribution of the packets to be transmitted in each of the determined communication windows takes into account at least one constraint associated with the application. Thus, the latency control is advantageously adapted to the envisaged application. Unlike the approaches of the state of the art, the optimization is no longer solely based on the state of the network or the communication channel, to the extent that certain latencies are acceptable by the envisaged application.

[0018] Advantageously, the latency of the communication is no longer dependent on the vagaries of the communication environment but is directly adapted to the entity that will use this data (for example the application or the control system)

[0019] The communication method according to the invention may also have one or more of the characteristics below, taken independently or in any technically conceivable combination.

[0020] The method further comprises an implementation of at least one jitter reduction mechanism for each transmission window.

[0021] The method further comprises sending the packets to be transmitted by implementing the determined mechanisms, according to the distribution of the packets per transmission window, and receiving feedback information relating to an application performance indicator.

[0022] The application performance indicator is a packet loss cost.

[0023] The method further comprises receiving a first target quality of service indicator value from one or more nodes of the communication network and determining an initial mechanism orchestration strategy for each node of the wireless communication network, enabling the first target quality of service indicator value to be achieved.

[0024] The first target quality of service indicator value is a maximum and / or minimum end-to-end latency value in the wireless communication network and / or a jitter value and / or a reliability value.

[0025] The classification of transmission windows is carried out into at least three quality of service classes, including an easy service class, a standard service class and a difficult service class.

[0026] The transmission windows are disjoint, at least two successive transmission windows being separated by a period of non-transmission.

[0027] According to another aspect, the invention provides a communication device forming a communication node in a wireless communication network comprising a plurality of communication nodes, comprising a processor configured to, following reception of a plurality of packets to be transmitted to a destination node, for a predetermined application, implement:

[0028] - a module for receiving information relating to a communication quality in the wireless communication network, including transmission latency data, and reception of information relating to an application performance indicator,

[0029] - a module for determining a plurality of disjoint transmission windows, and calculating, for each transmission window, an associated central time instant and an associated window width,

[0030] - a module for classifying transmission windows, into quality classes of service, based on a statistical distribution of latency and the application performance indicator,

[0031] - a module for distributing the packets to be transmitted to carry out a sending of each packet in at least one of the transmission windows, the distribution taking into account at least one constraint associated with said predetermined application.

[0032] Advantageously, the communication device is configured to implement a communication method as briefly described above, according to all the embodiments envisaged.

[0033] According to another aspect, the invention relates to a computer program comprising software instructions which, when executed by a programmable electronic device, implement a communication method as briefly described above.

[0034] Other characteristics and advantages of the invention will emerge from the description given below, for information purposes only and in no way limiting, with reference to the appended figures, among which:

[0035] [Fig-1] [Fig.l] is a block diagram of a wireless communication network comprising a communication device according to one embodiment of the invention;

[0036] [Fig.2] [Fig.2] is an example of transmission windows according to a mode of realization lization;

[0037] [Fig.3] [Fig.3] is a synopsis of the main stages of a communication process communication according to an embodiment;

[0038] [Fig.4] [Fig.4] is an example of transmission windows in a scenario of implementation of the automatic packet retransmission mechanism.

[0039] [Fig.l] schematically illustrates a wireless communication network 2, comprising a plurality of communication nodes 4, 6 and a radio communication channel 8.

[0040] A communication node is a communication device (or communication equipment), provided with a transmitter and a receiver and configured to operate in a wireless communication network, according to a chosen wireless communication protocol.

[0041] For example, the wireless communication device is provided with one or more transmitting / receiving antennas.

[0042] The communication channel 8 comprises in practice a plurality of interconnected wireless communication nodes, and configured to communicate in transmission and reception. The communication channel is variable over time depending on the usage scenario but also depending on the pair of nodes. This dynamic channel impacts the performance of the communication system (reliability, latency).

[0043] According to an optional embodiment, the wireless communication network 2 is a multi-hop network between communication nodes, the communication between a source node and a destination node being carried out through several relay nodes.

[0044] Alternatively, any other type of wireless communication network is conceivable.

[0045] The communication nodes of the network may be any type of device equipped with wireless communication interfaces, for example a relay station, a base station, a user device.

[0046] According to a variant or in addition, the communication network 2 is a network of sensors configured to capture physical quantities, or a transport network comprising vehicles carrying communication interfaces and road signaling devices also equipped with communication interfaces.

[0047] Of course, these examples are not exhaustive, any type of wireless communication network being capable of implementing a method as described below.

[0048] The communication device (or node) 4 of the communication network 2 is a programmable electronic device configured to implement a communication method as described below.

[0049] According to embodiments, this communication method is implemented in a centralized or distributed manner.

[0050] This programmable electronic device 4 comprises an electronic memory unit 10, a communication interface 12, and a processor 14

[0051] The processor 14 is configured to execute modules 20, 22, 24 to implement the communication method in a wireless communication network according to the invention.

[0052] Packets to be transmitted, P, each packet having an associated timestamp, are received and stored in the electronic memory unit 10, for example in a buffer memory structure.

[0053] The module 20 is a module for receiving packets to be transmitted and for estimating communication conditions, and in particular for estimating a statistical distribution of latency in the communication network. Furthermore, the receiving module 20 is configured to also receive information relating to an application performance indicator, from a destination node or from a control system associated with a destination node. For example, in the example illustrated in [Fig.l], the destination node is the device 6, which implements a chosen application.

[0054] The module 22 is a control module configured to implement a determination of a plurality of transmission windows and a calculation for each transmission window of an associated time instant and an associated window width, and a classification of transmission windows into quality of service classes based on the statistical distribution of latency and the application performance indicator, e.g. received from the destination node.

[0055] The module 24 is a module for distributing the packets to be transmitted in at least one of the determined transmission windows, taking into account at least one constraint associated with a predetermined application App, implemented by the destination node 6, or by a system 15 connected to the destination node.

[0056] For example, system 15 is a robot or receiver, connected to destination node 6.

[0057] Preferably, the module 24 takes into account information transmitted by the destination node, coming from the application App, and thus distributes the packets according to their use by the application App, for example in the appropriate order for the application App.

[0058] For example, in an application for capturing digital images by the communication device 4 and for processing the images, for example for a classification implementing an artificial intelligence algorithm, by the destination node 6, the state of the computational resources available in the device 6 for image processing can be taken into account for the distribution of the image data packets, to avoid overloading the buffer memory of the device 6 when the computational resources are not available.

[0059] In other words, communications are scheduled based on channel constraints and the needs of the user application or user system.

[0060] The communication device 4 is configured to receive feedback, at least from the destination node 6, which is intended to receive the packets Pjet to provide them to the given application App, as well as from the intermediate nodes of the communication network.

[0061] The feedback includes in particular information relating to transmission latency, in particular end-to-end latency.

[0062] End-to-end latency includes, for example, one or more parameters among the propagation, processing and queuing delays of packets and associated data, for transmission between a source node and a destination node of the wireless communication network.

[0063] In one embodiment, the modules 20, 22, 24 are each produced in the form of communication software.

[0064] This software is capable of being recorded, in the form of a computer program comprising software instructions which implement a communication method as described below. The software instructions are capable of being recorded on a medium, not shown, readable by a computer. The computer-readable medium is, for example, a medium capable of storing electronic instructions and of being coupled to a bus of a computer system. For example, the readable medium is an optical disk, a magneto-optical disk, a ROM memory, a RAM memory, any type of non-volatile memory (for example EPROM, EEPROM, FLASH, NVRAM), a magnetic card or an optical card.

[0065] In a variant not shown, the modules 20, 22, 24 are each produced in the form of a programmable logic component, such as an FPGA (Field Programmable Gate Array), a GPU (graphics processor) or a GPGPU (General-purpose graphics processing), or in the form of a dedicated integrated circuit, such as an ASIC (Application Specific Integrated Circuit).

[0066] [Fig.2] illustrates a graph representing the probability density of reception of packets (on the ordinate, axis referenced “P”) as a function of time (on the abscissa, axis referenced “T”) for a number K of calculated transmission windows, K being an integer greater than 1.

[0067] In the example of [Fig.2], three transmission windows, denoted respectively Wi, w2 and wk are illustrated.

[0068] Each window Wi is defined by a central time instant f (respectively tb t2 and tk in the example) and by a window width D;, defined by the time difference between a transmission window end instant ti2 and a transmission window start instant G: D^GG.

[0069] In the illustrated example, the packets transmitted during the first window Wi will have a lower latency than the packets transmitted during the second window w2 and so on.

[0070] Between two transmission windows, transmission is not activated, which is represented by a non-transmission range “OFF”. Thus, advantageously, the proposed communication method saves resources.

[0071] According to embodiments, the resources comprise a hardware resource for storage, calculation or transmission, and / or a virtual resource such as a virtual machine, and / or a time or frequency resource. Indeed, the communication spectrum being divided into time / frequency cells, it is possible to manage the allocation of the time / frequency cells to the recipients by time allocation, or by frequency allocation or by a division of the time / frequency spectrum.

[0072] [Fig.3] is a synopsis of the main stages of a communication process implemented by a programmable electronic device as described in [Fig.l].

[0073] The method is executed during a predetermined execution time, then its execution is repeated. The method comprises receiving and storing packets to be transmitted (step 50) to a destination node for a predetermined application, which runs in parallel with the other steps described below. The packets to be transmitted are, for example, packets formatted according to the chosen communication protocol. The arrival of the packets is ordered in time, each packet having an associated timestamp.

[0074] It should be noted that the method described can be used for one or more applications. For a given application, processing of the packets in the order of arrival is envisaged.

[0075] For a plurality of applications, it is possible to assign a priority to a flow or a time-to-live to each packet (or TTL for “Time-to-Live”).

[0076] The method further comprises a step 52 of receiving a first target value of expected quality of service indicator, obtained from one or more nodes of the wireless communication network. The target value of quality of service indicator is for example a maximum end-to-end latency value and / or a minimum end-to-end latency value, and / or a jitter value, and / or a reliability value (success rate of packet reception).

[0077] In step 54, an initial strategy for orchestrating mechanisms for each node of the wireless communication network, making it possible to achieve the first target value of the quality of service indicator, is developed and applied. This step 54 can be implemented by any method known for this purpose, for example the method described in patent application FR2103542.

[0078] According to embodiments, a mechanism is chosen from: mechanisms for deploying and controlling the wireless communication network, mechanisms for adapting the physical layer of the communication network and its configurations, frequency agility mechanisms, mechanisms related to access and resource planning, mechanisms for controlling and monitoring communication resources, mechanisms for routing and managing connectivity, redundancy mechanisms (e.g. the HARQ automatic retransmission procedure), mechanisms for selecting multiple paths, mechanisms for managing resources (e.g., adding resources such as adding a relay, communication antennas or computing units).

[0079] The selection of one or more mechanisms is performed based on the first target quality of service indicator value, taking into account the execution time.

[0080] A mechanism orchestration strategy designates the selection of a set of mechanisms, the mechanisms being executed by nodes of the network, to achieve a quality level, i.e. a target quality indicator value.

[0081] Following the application of step 54 of implementing the mechanisms of the initial orchestration strategy, the method implements a step 56 of evaluating the performance formance of the communication network. For example, a statistical distribution of latency is estimated. For example, the statistical distribution of end-to-end latency is calculated.

[0082] Next, information relating to a quality of communication in the wireless communication network, including transmission latency data, is received at the statistics receiving step 58. For example, information relating to latency is received from all nodes in the communication network.

[0083] In addition, in step 58, information relating to an application performance indicator is received. In embodiments, the application performance indicator is a packet loss cost. Thus, during the subsequent iteration of the steps, the performance will be improved for adaptation to the application, i.e., for example, making it possible to prioritize the transmission of packets that minimizes the loss cost.

[0084] The method then comprises a step 60 of determining a second target value of expected quality of service indicator, dependent on the application, and the information received in step 58.

[0085] The method then comprises a determination of a plurality of transmission windows 62, comprising a step 64 of obtaining a number K of transmission windows, K being an integer greater than 1, a step 66 of calculating for each transmission window, an associated central time instant and an associated window width, and a step 68 of classifying the transmission windows, into quality of service classes, as a function of a statistical distribution of the latency and the application performance indicator.

[0086] According to one embodiment, the number K of windows obtained in step 64 is a predetermined number, for example K=3.

[0087] According to a variant, the number K of windows is calculated according to the statistical distribution of the end-to-end latency calculated previously.

[0088] An example of window calculation will be explained below with reference to [Fig.4].

[0089] Transmission windows are classified into a plurality of quality classes of service at step 68, based on the statistical distribution of the end-to-end latency calculated previously and the application performance indicator.

[0090] For example, in one embodiment, the classification involves associating a level of ease of service, ranging from "easy-to-serve," "standard-to-serve," and "difficult-to-serve."

[0091] A subset of packets to be transmitted is distributed (step 70) to one or more of the transmission windows, taking into account at least one constraint associated with the application. Thus, the latency of the packets is controlled in a de terministic, the time of sending certain packets can be brought forward or delayed according to application needs.

[0092] In addition, a weight is calculated and associated with each transmission window, for example by integrating the latency distribution for each window.

[0093] Preferably, the transmission windows thus calculated are disjoint.

[0094] In one embodiment, at least for a portion of the transmission windows, two successive transmission windows are separated by a transmission stop range (for example, the “OFF” ranges in [Fig.2]).

[0095] The method then comprises, according to embodiments, a step 72 of controlling the variation of the latency, which implements a strategy of orchestration of mechanisms, in order to reduce the jitter for each transmission window.

[0096] For example, in step 72 a minimization of the sum of the variations of the latency sk of the transmission windows is implemented.

[0097] In step 74 the strategy for orchestrating jitter reduction mechanisms is implemented, and the packets are transmitted to the destination node via the wireless communication network, each node implementing the determined mechanisms.

[0098] Following the transmission, a reception of feedback information (step 76) takes place, during which information relating to an application performance indicator is received. In embodiments, the application performance indicator is a packet loss cost. Step 76 is then followed by step 58 of receiving feedback information previously described.

[0099] [Fig.4] illustrates transmission windows in a scenario implementing the automatic packet retransmission mechanism, known as HARQ (from the English “Hybrid Automatic Repeat reQuest”).

[0100] In a conventional implementation, the retransmission is performed reactively, following the reception of a non-acknowledgment (NAQ) from the receiver. Such a strategy induces a high latency, due to the waiting for acknowledgment or non-acknowledgment, but allows for a saving of implemented resources. To ensure a predetermined latency, a proactive retransmission is necessary, but this induces a potential overconsumption of transmission resources. With the proposed method, and as shown in [Fig.4], the RTX;retransmission of packets (denoted RTX; for the ith retransmission) is advanced or delayed in a controlled manner. This makes it possible to reduce the jitter while ensuring a latency according to the application needs.

[0101] In more detail, in [Fig.4] four transmission windows are illustrated in a graph representing the probability density (reference "P"), abscissa axis) as a function of time (reference "T", ordinate axis), in a manner analogous to the representation of [Fig.2], the number of windows determined by the method being K=4 in this example, while a number of windows equal to 8 is provided using the dis- statistical contribution of latency (classical method).

[0102] In addition to the transmission windows, the transmission or retransmission of packets is represented. Thus, the first window Wi corresponds to the initial transmission, the second window w2 to two retransmissions in parallel, RTXi and RTX2, the third window w3 to three retransmissions in parallel, RTX3, RTX4 and RTX5, the fourth window w4 to two retransmissions in parallel, RTX6 and RTX7.

[0103] The way of parallelizing transmissions, and the choices of retransmissions depend on the application performance indicator, and therefore allow adaptation to the application.

[0104] The method described is further adaptable to the implementation of a plurality of distinct applications.

[0105] Advantageously, the invention allows quasi-deterministic control of latency, and a reduction in latency variation.

[0106] Advantageously, the implementation of mechanisms for minimizing the variation in latency per transmission window is facilitated, because the transmission windows are classified into quality of service classes, and it is easier to optimize the transmission for the same quality of service class.

[0107] Advantageously, the proposed method makes it possible to save resources, in particular during non-transmission periods.

Claims

Claims

1. Method of communication in a wireless communication network comprising a plurality of communication nodes, implemented by a processor of a communication node (4), comprising a reception of a plurality of packets to be transmitted to a destination node, for a predetermined application, and being characterized in that it comprises steps of: - reception of information (56, 58) relating to a quality of communication in the wireless communication network, comprising transmission latency data, and reception of information relating to an application performance indicator, - determination (62, 64, 66) of a plurality of disjoint transmission windows, and calculation, for each transmission window, of an associated central time instant and an associated window width, - classification (68) of the transmission windows, in quality of service classes,based on a statistical distribution of the latency and the application performance indicator, - distribution (70) of the packets to be transmitted to carry out a sending of each packet in at least one of the transmission windows, the distribution taking into account at least one constraint associated with said predetermined application.,

2. The method of claim 1, further comprising implementing (72, 74) at least one jitter reduction mechanism for each transmission window.

3. Method according to claim 2, further comprising a (74) sending of the packets to be transmitted by implementing the determined mechanisms, according to the distribution of the packets per transmission window, and a reception (76) of feedback information relating to an application performance indicator.

4. Method according to one of claims 1 to 3, in which the application performance indicator is a packet loss cost.

5. A method according to any one of claims 1 to 4, further comprising receiving (52) a first target quality of service indicator value from one or more nodes of the communication network and determining (54) an initial mechanism orchestration strategy for each node of the wireless communication network, enabling the first target quality of service indicator value to be achieved. service.

6. The method of claim 5, wherein the first target quality of service indicator value is a maximum and / or minimum end-to-end latency value in the wireless communication network and / or a jitter value and / or a reliability value.

7. A method according to any one of claims 1 to 6, wherein the classification (68) of the transmission windows is carried out into at least three quality of service classes, comprising an easy service class, a standard service class and a difficult service class.

8. A method according to any one of claims 1 to 7, wherein said transmission windows are disjoint, at least two successive transmission windows being separated by a period of non-transmission.

9. Communication device forming a communication node (4) in a wireless communication network comprising a plurality of communication nodes, comprising a processor configured to, following reception of a plurality of packets to be transmitted to a destination node, for a predetermined application, implement: - a module (20) for receiving information relating to a quality of communication in the wireless communication network, comprising transmission latency data, and for receiving information relating to an application performance indicator, - a module (22) for determining a plurality of disjoint transmission windows, and calculating, for each transmission window, an associated central time instant and an associated window width, - a module (22) for classifying the transmission windows, into quality of service classes,based on a statistical distribution of the latency and the application performance indicator, - a module (24) for distributing the packets to be transmitted to carry out a sending of each packet in at least one of the transmission windows, the distribution taking into account at least one constraint associated with said predetermined application.,

10. A computer program comprising software instructions which, when executed by a programmable electronic device, implement a method of communicating in a wireless communication network according to claims 1 to 8.

Images