Communication device, access point, and communication method

Abstract

The present invention provides a communication apparatus, an access point and a communication method. To avoid blocking a node AC queue in a degraded MU EDCA mode due to periodic OFDMA transmission of data from another AC queue in a resource unit provided by the AP, the present invention proposes using a dedicated HEMU EDCA Timer for each AC queue so that the AC queue can exit the degraded MU EDCA mode independently of other AC queues. In this regard, upon successful transmission of data stored by two or more traffic queues in each of one or more accessed resource units provided by the AP within one or more transmission opportunities, the node sets each traffic queue that transmitted in the accessed resource unit into the degraded MU EDCA mode and for a predetermined duration of degradation that is derived from a respective timer associated with the transmitting traffic queue counting down. Then, upon expiration of any timer, the node switches the associated traffic queue back to the legacy EDCA mode.

Description

[0001] (This application is a divisional application of application No. 201780065054.8, filed on October 16, 2017, with the title "Enhanced management of ACs in multi-user EDCA transmission mode in wireless networks". TECHNICAL FIELD

[0002] The present invention generally relates to communication networks, and more particularly to communication networks providing channel access to nodes by contention and providing secondary access to these nodes to sub-channels (or resource units) partitioning a transmission opportunity TXOP granted to an access point, for transmitting data.

[0003] The present invention is applicable in wireless communication networks, in particular in 802.11ax networks, for providing access to 802.11ax composite channels and / or to OFDMA resource units forming, for example, 802.11ax composite channels granted to an access point, to nodes, and enabling uplink communication. BACKGROUND

[0004] The IEEE 802.11 MAC standard defines the way in which a wireless local area network (WLAN) must work at the physical layer level and at the medium access control (MAC) layer level. Typically, the 802.11 MAC (Medium Access Control) operating mode implements the well-known Distributed Coordination Function (DCF), which relies on a contention-based mechanism based on the so-called "Carrier Sense Multiple Access / Collision Avoidance" (CSMA / CA) technique.

[0005] The 802.11 medium access protocol standard or operating mode mainly concerns the management of communicating nodes waiting for the wireless medium to become free in order to try to access it.

[0006] The network operating mode defined by the IEEE 802.11ac standard provides very high throughput (VHT) by moving from the 2.4 GHz band, which is considered to be extremely vulnerable to interference, to the 5 GHz band, thus making it possible to use wider frequency contiguous channels of 80 MHz, two of which can optionally be combined to obtain a 160 MHz channel as the operating band of the wireless network.

[0007] The 802.11ac standard also modifies control frames such as Request to Send (RTS) and Allow to Send (CTS) frames to allow composite channels with different predefined bandwidths of 20MHz, 40MHz, or 80MHz, where these composite channels consist of one or more consecutive communication channels within the operating frequency band. A 160MHz composite channel can be a combination of two 80MHz composite channels within the 160MHz operating frequency band. The control frame specifies the channel width (bandwidth) of the target composite channel.

[0008] Therefore, a composite channel includes a primary channel for a given node to perform an EDCA backoff procedure to access the medium, and at least one secondary channel, each for example, 20 MHz.

[0009] EDCA (Enhanced Distributed Channel Access) defines service categories and four corresponding access categories that allow high-priority services to be processed differently from low-priority services.

[0010] The implementation of EDCA in a node can be achieved using multiple service queues (known as "access categories") for serving data services with different priorities, where each service queue is associated with a corresponding queue backoff value. The queue backoff value is calculated based on individual queue contention parameters (e.g., EDCA parameters) and is used to compete for access to the communication channel in order to transmit the data stored in the service queue.

[0011] Traditional EDCA parameters include CW for each service queue. min CW max And AIFSN, of which CW min and CW max It represents the lower and upper boundaries of the selection range of the EDCA contention window CW for a given traffic queue. AIFSN represents the number of arbitration inter-frame gaps and defines the number of time slots (typically 9 μs) that a node must listen for as idle media before decrementing the queue backoff value associated with the traffic queue under consideration, excluding the DIFS interval (which defines the sum of AIFS time periods).

[0012] EDCA parameters can be defined in beacon frames sent by specific nodes in the network to broadcast network information.

[0013] The contention window (CW) and queue backoff value are EDCA variables.

[0014] The traditional EDCA backoff process involves a node selecting a queue backoff value for the service queue from each contention window (CW), and then decrementing that queue backoff value when the medium is idle after the AIFS period. Once the backoff value reaches zero, the node is allowed to access the medium.

[0015] Thus, EDCA queue backoff values or counters serve two purposes. First, they drive nodes to access the medium efficiently by reducing the risk of collisions; second, they provide management of quality of service, i.e. QoS, by reflecting the aging of the data contained in the traffic queue (the older the data, the lower the backoff value), and thus by providing different priorities to traffic queues through different values of EDCA parameters, in particular the AIFSN parameter which starts the decrease of the EDCA queue backoff value.

[0016] Thanks to the EDCA backoff procedure, nodes can access the communication network using a contention-based access mechanism based on the queue contention parameters, typically based on the computed queue backoff counter or value.

[0017] A communication node uses the primary channel to listen whether the channel is free, and can use the secondary channel to extend the primary channel to form a composite channel. The primary channel can also be used alone.

[0018] In view of the tree-like decomposition of the operating frequency band into basic 20MHz channels, some secondary channels are named tertiary or quaternary channels.

[0019] In 802.11 ac, all transmissions and thus possible composite channels include the primary channel. This is because nodes only perform full Carrier Sense Multiple Access / Collision Avoidance (CSMA / CA) and Network Allocation Vector (NAV) tracking on the primary channel. Other channels are assigned as secondary channels, where on a secondary channel, a node only has the capability of CCA (Clear Channel Assessment), i.e. to detect the free or busy state / status of said secondary channel.

[0020] The problem with the use of composite channels defined in 802.11 n or 802.11 ac (or 802.11 ax) is that nodes compatible with the use of composite channels, i.e. 802.11 n and 802.11 ac compatible nodes or "HT nodes" (standing for High Throughput nodes), must coexist in the same wireless network with legacy nodes that cannot use composite channels but rely only on legacy 20MHz channels, i.e. non-HT nodes compatible only with e.g. 802.11 a / b / g, and thus must share the same 20MHz channels.

[0021] To solve this problem, 802.11 n, 802.11 ac and 802.11 ax standards provide the possibility to duplicate control frames (e.g. RTS / CTS frames or CTS-to-Self frames or ACK frames to acknowledge the correct or incorrect reception of transmitted data) on each 20MHz channel in 802.11 a legacy format, called "non-HT", to establish protection of the requested TXOP on the whole composite channel.

[0022] This is targeted to any legacy 802.11a node, which uses any 20MHz channel contained in the composite channel to learn about the communication ongoing on the 20MHz channel. As a result, the legacy node is prevented from initiating a new transmission before the end of the current composite channel TXOP granted to the 802.11n / ac / ax node.

[0023] As originally proposed by 802.11n, a duplication of the legacy 802.11a or "non-HT" transmission is provided to enable the simultaneous transmission of two identical 20MHz non-HT control frames on both the primary and secondary channels making up the used composite channel.

[0024] This approach has been extended for 802.11ac to allow duplication on the channels making up the 80MHz or 160MHz composite channel. In the rest of this document, "duplicated non-HT frame" or "duplicated non-HT control frame" or "duplicated control frame" means that the node device duplicates the legacy or "non-HT" transmission of a given control frame on the secondary 20MHz channel of the (40MHz, 80MHz or 160MHz) operating band.

[0025] In practice, to request a new TXOP to the composite channel (equal to or greater than 40MHz), the 802.11n / ac node performs an EDCA backoff procedure in the primary 20MHz channel as described above. In parallel, the 802.11n / ac node performs a channel sensing mechanism such as Clear Channel Assessment (CCA) signal detection on the secondary channel to detect the secondary channel as idle (channel state / status "idle") during the PIFS interval before the start of the new TXOP (i.e. before any queue backoff counter expires).

[0026] Recently, the Institute of Electrical and Electronics Engineers (IEEE) officially approved the 802.11ax Task Group as a successor to 802.11ac. The main goal of the 802.11ax Task Group relates to trying to improve the data speed of wireless communication devices used in dense deployment scenarios.

[0027] Recent developments of the 802.11ax standard try to optimize the use of composite channels by a plurality of nodes in a wireless network with an Access Point (AP). Indeed, typical content has important data volumes related to, for example, high-definition audiovisual real-time interactive content. Moreover, it is well known that the performance of the CSMA / CA protocol used in the IEEE 802.11 standard rapidly deteriorates as the number of nodes and the traffic volume increase (i.e. in dense WLAN scenarios).

[0028] In this context, Multi-User (MU) transmission has been considered to allow multiple simultaneous transmissions to different users in both downlink (DL) and uplink (UL) directions with respect to the AP and during transmission opportunities granted to the AP. In uplink, Multi-User transmission can be used to reduce the probability of collisions by allowing multiple non-AP stations or nodes to transmit simultaneously.

[0029] To actually perform such Multi-User transmission, it has been proposed to split the licensed communication channel into sub-channels (also referred to as Resource Units (RUs)) which are shared by multiple users (non-AP stations / nodes) in the frequency domain, e.g. based on Orthogonal Frequency Division Multiple Access (OFDMA) techniques. Each RU can be defined by a number of tones, with an 80MHz channel containing up to 996 available tones.

[0030] OFDMA is a multi-user variant of OFDM that emerged as a new key technology to improve the efficiency of wireless networks based on the advanced architecture. OFDMA combines OFDM on the physical layer with Frequency Division Multiple Access (FDMA) on the MAC layer, enabling different subcarriers to be assigned to different stations / nodes to increase concurrency. Adjacent subcarriers often experience the same channel conditions and are thus grouped into sub-channels: OFDMA sub-channels or RUs are thus a collection of subcarriers.

[0031] As currently conceived, the granularity of such OFDMA sub-channels is finer than the original 20MHz channel band. Typically, 2MHz or 5MHz sub-channels can be considered as the minimum width, so that e.g. 9 sub-channels or Resource Units are defined within a single 20MHz channel.

[0032] The multi-user feature of OFDMA allows the AP to assign or provide different RUs to different non-AP stations / nodes to increase contention. This can help to reduce contention and collisions within the 802.11 network.

[0033] In contrast to downlink OFDMA where the AP can send multiple data directly to multiple stations (supported by a specific indication within the PLCP header), a trigger mechanism has been adopted for the AP to trigger Multi-User Uplink (MU UL) OFDMA communication from various nodes.

[0034] To support Multi-User Uplink (i.e. uplink transmission to the 802.11ax Access Point (AP) during a pre-emptive TXOP), the 802.11ax AP must provide signaling information for legacy nodes (non-802.11ax nodes) to set their NAV and for 802.11ax nodes to determine the allocation of Resource Units RUs provided by the AP.

[0035] The 802.11ax standard defines a Trigger Frame (TF) sent by the AP to 802.11ax stations to trigger a multi-user uplink communication.

[0036] The document IEEE 802.11-15 / 0365 proposes to send a "trigger" frame (TF) by the AP to request the transmission of uplink (UL) multi-user (OFDMA) PPDU from multiple stations. The TF defines the resource units provided by the AP to these stations. In response, the stations send UL MU (OFDMA) PPDU as an immediate response to the trigger frame. All transmitters can send data simultaneously but using disjoint sets of RUs (i.e. frequencies in OFDMA scheme) resulting in less interfered transmissions.

[0037] The bandwidth or width of the target composite channel is announced in the TF frame, which means that a value of 20MHz, 40MHz, 80MHz or 160MHz is added. In appropriate cases, the TF frame is sent via the primary channel of 20MHz and the TF frame is copied (repeated) on each other channel of 20MHz, forming the target composite channel. As mentioned above for the duplication of control frames, it is expected that each nearby legacy (non-HT or 802.11ac) station receiving the TF on the primary channel will then set the station's NAV to the value specified in the TF. This prevents these legacy stations from accessing the channels in the target composite channel during the TXOP.

[0038] A resource unit RU can be reserved for a specific station, in which case the AP indicates in the TF that the RU is reserved for the station. Such a RU is called a scheduled RU. The indicated station does not need to contend when accessing the scheduled RU reserved for it.

[0039] The type of data allowed to be sent by a station in a scheduled RU can be specified by the AP in the TF. For example, the TF includes a "preferred AC" field of 2 bits in which the AP indicates one of the four EDCA access categories. On the other hand, the AP can leave the scheduled RU open to any type of data. To activate or not activate the "preferred AC", the TF includes another 1-bit field, the "AC preference level".

[0040] To better improve the efficiency of the system with respect to unmanaged traffic for the AP (e.g. uplink management frames from associated stations, non-associated stations intending to reach the AP, or simply unmanaged data traffic), the AP can offer resource units to 802.11ax stations through contention-based access. In other words, a resource unit RU can be randomly accessed by more than one station (of a group of stations registered to the AP). Such a RU is called a random RU and is so indicated in the TF. The random RU can be used as a basis for contention between stations intending to access the communication medium to send data.

[0041] A typical random resource selection procedure is defined in document IEEE 802.11-15 / 1105. According to this procedure, each 802.11ax node maintains a dedicated backoff engine, called OFDMA or RU (short for Resource Unit) backoff engine, using a RU contention parameter including a RU backoff value, to contend for access to one of the random RUs. Once its OFDMA or RU backoff value reaches 0 (the OFDMA or RU backoff value is decremented by the number of random RUs for example at each new TF-R frame as defined here), the node becomes eligible for RU access and thus randomly selects one RU among all the random RUs defined in the received trigger frame. The node then transmits data for at least one of the traffic queues using the selected RU.

[0042] From the above, it is readily seen that the multi-user uplink medium access scheme (or OFDMA or RU access scheme) allows to reduce the number of collisions generated by simultaneous medium access attempts while also reducing the overhead due to medium access (as the medium access cost is shared among several nodes). Thus, the OFDMA or RU access scheme appears to be more efficient (in terms of medium usage) than the traditional EDCA contention-based medium access scheme in the context of high density 802.11 cells.

[0043] Although the OFDMA or RU access scheme appears to be more efficient, the EDCA access scheme must also survive and thus coexist with the OFDMA or RU access scheme.

[0044] This is mainly due to the presence of legacy 802.11 nodes that must still have a chance to access the medium while they are not aware of the OFDMA or RU access scheme. And global fairness with respect to medium access must be ensured.

[0045] It is also more necessary that 802.11ax nodes should have a chance to achieve access to the medium through the traditional EDCA contention-based medium access, for example to transmit data to another node (i.e. for traffic different from uplink traffic to the AP).

[0046] Thus, both medium access schemes, EDCA and OFDMA / RU access scheme, must coexist.

[0047] This coexistence has drawbacks.

[0048] For example, 802.11ax nodes and legacy nodes have the same medium access probability using the EDCA access scheme. However, 802.11ax nodes have additional medium access opportunities using the MU uplink or OFDMA or RU access scheme.

[0049] This leads to an access to the medium that is not completely fair between 802.11ax nodes and legacy nodes.

[0050] To restore some fairness between nodes, a solution is proposed as follows: upon successful transmission of data via an accessed resource unit (i.e. by UL OFDMA transmission), the current value of at least one queue contention parameter is modified to a penalized or degraded value to reduce the probability for the node to access the communication channel by the (EDCA) contention access scheme. For example, the penalized or degraded value is more restrictive than the original (or legacy) value.

[0051] For example, the document IEEE 802.11-16 / 1180 entitled “Proposed text changes for MU EDCA parameters” proposes that upon successful transmission of data in an AP reserved resource unit (RU) (MU UL OFDMA), the node is set in MU EDCA mode for a predetermined duration counted down by a timer (denoted hereafter as HEMU EDCA Timer (acronym for High Efficiency Multi-User EDCA Timer)) during which the EDCA parameters are set to values called MU EDCA parameter values or MU values that are different from the legacy values used in legacy EDCA mode. The MU parameter values are set to values that are more restrictive than the legacy values: a more restrictive value for an EDCA parameter means that the probability for the node to access the communication channel by the EDCA access scheme using the MU value is reduced with respect to the access using the legacy value.

[0052] In other words, upon transmission of some data by a node from one or more traffic queues using a scheduled RU assigned to the node by the AP, the node should immediately modify the EDCA parameters associated with the transmitted traffic queues (hereafter called “degraded”, “penalized” or “blocked” traffic queues) with some special more restrictive (“MU” or “degraded”) values that can be provided by the AP in a dedicated information element of the beacon frame, the EDCA parameters also including the values to be used by the node for their HEMU EDCA Timer.

[0053] Thus, it can be noted that the AP can send to the node a message to drive the node to modify their current values of EDCA parameters to the more restrictive values of the MU values upon successful transmission of data by the node via an accessed resource unit. This is also to reduce the probability for the node to access the communication channel by the EDCA access scheme.

[0054] In addition, the AP can determine the more restrictive values based on the history of data received from the node (e.g. via the RU).

[0055] The disclosed method proposes to increase the value of the AIFSN for each transmission traffic queue only, while keeping the CW min and the CW max unchanged. With the extension of the respective AIFS time period, especially in high-density environments where the medium is kept idle for a long time, it is prevented (or at least substantially delayed) that the traffic queues in MU EDCA mode decrement their queue backoff value or counter when the medium is sensed to be idle. New accesses to the medium using the EDCA access scheme are statistically significantly reduced, or even no longer exist.

[0056] Upon switching to the MU EDCA mode, the node starts its HEMUEDCATimer countdown. Each time the node successfully transmits data in the newly reserved RU (MU UL OFDMA), the HEMUEDCATimer is reinitialized. It is proposed that the HEMUEDCATimer has a high initialization value (e.g. tens of milliseconds) in order to encompass several new opportunities for MU UL transmission.

[0057] Upon expiration of the HEMUEDCATimer, the traffic queues in MU EDCA mode switch back to the legacy EDCA mode with legacy EDCA parameters, thereby exiting the queues from the MU EDCA mode.

[0058] Thus, this mechanism of dual working modes (legacy EDCA mode and MU EDCA mode) facilitates the use of the MU UL mechanism by reducing the probability that the node uses the EDCA mechanism to access the medium through MU UL transmission.

[0059] The HEMUEDCATimer mechanism that reinitializes the HEMUEDCATimer each time the node successfully transmits new data in the accessed reserved RU means that the node remains in the MU EDCA state as long as the AP provides (scheduled or random) RUs to the node.

[0060] This method has major drawbacks as now explained.

[0061] If the node transmits data from two or more traffic queues in one or more resource units provided by the AP (e.g. if the dedicated traffic queue becomes empty, the node selects other data to be sent from the traffic queue with higher priority), the two or more traffic queues become in the MU and more restrictive EDCA mode. Since the AIFSN of these traffic queues, for example, is very restrictive, it is mainly prevented to access the medium via the EDCA access scheme.

[0062] It can happen that the AP periodically provides the node (in this situation) with resource units indicating the preferred traffic queue from which data is to be selected.

[0063] As long as this polling to the preferred traffic queue continues, the node empties the corresponding traffic queue when accessing the offered resource unit, while keeping all two or more traffic queues in MU EDCA mode. This means that the other traffic queues remain locked in MU and more restrictive EDCA mode and cannot be emptied by accessing the medium.

[0064] Thus, the QoS in the network is severely degraded. SUMMARY

[0065] The present invention seeks to overcome the above limitations. In particular, the present invention seeks to overcome the loss of QoS handling caused by the introduction of MU UL OFDMA transmission.

[0066] The inventors have noticed that locking the other traffic queues in MU contention mode, such as MU EDCA mode, is caused by the fact that the HEMUEDCATimer is reinitialized each time data from the same preferred traffic queue is (periodically) transmitted in a resource unit. Thus, the unicity of the HEMUEDCATimer managing all traffic queues simultaneously in MU contention mode is substantially detrimental to QoS.

[0067] Thus, the present invention aims at restoring some QoS by breaking the unicity of the HEMUEDCATimer.

[0068] In this context, the present invention provides a communication method in a communication network comprising a plurality of nodes, at least one node comprising a plurality of traffic queues for serving data traffic at different priorities, each traffic queue being associated with a corresponding queue backoff value computed from a corresponding queue contention parameter having a legacy value in a legacy contention mode and used to contend for access to a communication channel for transmitting data stored in the traffic queue,

[0069] The communication method comprises:

[0070] at the node

[0071] transmitting data stored in two or more traffic queues in one or more accessed resource units offered by another node within a grant of one or more transmission opportunities on the communication channel to the another node;

[0072] when transmitting data in each of the visited resource units, setting each of the transmission traffic queues (i.e. transmitted in the visited unit) in a MU contention mode different from the legacy contention mode and for a predetermined duration resulting from the countdown of the corresponding timer associated with this transmission traffic queue, wherein the corresponding queue contention parameters of the MU contention mode are set to MU values different from the legacy values; and upon expiration of any timer, switching back the associated (degraded) traffic queue to the legacy contention mode in which the corresponding queue contention parameters are set back to the legacy values.

[0073] Thus, the application proposes to use a dedicated HEMU EDCA Timer for each AC queue so that each AC queue can exit from the MU contention mode independently from the other AC queues.

[0074] The timer can be implemented using hardware and software.

[0075] Thus, fairness between the two contention modes is restored for 802.11ax nodes.

[0076] The MU values different from the legacy values of the traffic queues means that at least one of the MU values and the legacy values of the same contention parameter are different from each other, regardless of whether the MU values and the legacy values of the other contention parameters are equal or different.

[0077] Correspondingly, the application also relates to a communication device in a communication network comprising a plurality of nodes, the communication device comprising:

[0078] a plurality of traffic queues for serving data traffic with different priorities, each traffic queue being associated with a corresponding queue backoff value computed from corresponding queue contention parameters having legacy values in a legacy contention mode and used for contending for access to a communication channel for transmitting data stored in this traffic queue,

[0079] a plurality of timers, each timer being associated with one of the traffic queues; and

[0080] at least one microprocessor configured to perform the following steps:

[0081] transmitting data stored in two or more traffic queues in one or more visited resource units provided by another node within one or more transmission opportunities granted to this another node on the communication channel;

[0082] when transmitting data in the visited resource units, setting each of the transmission traffic queues into a MU contention mode different from the legacy contention mode and for a predetermined duration resulting from the countdown of the associated timer, wherein the respective queue contention parameters of the MU contention mode are set to MU values different from the legacy values; and

[0083] upon expiration of any of the timers, switching back the associated (degraded) traffic queue to the legacy contention mode in which the respective queue contention parameters are set back to the legacy values.

[0084] The apparatus node has the same advantages as the above defined method.

[0085] Some of the features are explained below with reference to the method, while they can be translated into system features specific to any apparatus node according to the application.

[0086] In embodiments, the predetermined (preferably degraded) durations for initializing the timers associated with the two respective traffic queues are different from each other. This configuration improves the management of the QoS.

[0087] In other embodiments, the predetermined duration for initializing the timer associated with the respective traffic queue is computed from a common initialization value received from the other node and from an adjustment parameter specific to the respective traffic queue. Thus, the combination of the common initialization value and the adjustment parameter of the AC queue makes it possible to easily adjust the duration for which each AC queue should remain in the MU contention mode. Thus, this configuration also improves the management of the QoS.

[0088] In variants, the predetermined duration for initializing the timer associated with the respective traffic queue is set to a respective initialization value received directly from the other node. In other words, the other node, such as the AP, directly drives the duration for which the AC queues should respectively remain in the MU contention mode.

[0089] This makes it possible to efficiently manage the QoS, in particular because the other node, such as the AP, can have a global view of the network (such as statistics related to collisions, number of nodes, etc.). This makes it easier to prioritize the adjustment of such AC queues, which can result in a more efficient network.

[0090] In some embodiments, the timer associated with the respective traffic queue is reinitialized to the respective predetermined duration each time data from this associated traffic queue is transmitted in the visited resource units by the other node within any subsequent transmission opportunity granted to the other node on the communication channel.

[0091] This means that the timer used by the AC queue will expire only if, within the predetermined duration of the initialization of the timer, there is no transmission of data from the AC queue in any of the resource units provided by another node. Otherwise, the timer is reinitialized again.

[0092] As a result, the traffic queue can exit the MU contention mode by restoring the corresponding queue contention parameter to the legacy value only if there is no OFDMA transmission of data from the considered AC queue within the specified duration.

[0093] In an embodiment, the method further comprises, at the node, contending for access to the communication channel using the queue contention parameter in the MU contention mode. This means that the MU contention mode follows the same contention scheme as the legacy contention mode (e.g. legacy EDCA) but with MU (preferably degraded (i.e. more restrictive)) contention parameters in order to penalize the AC queues polled by the AP for transmission in the RUs.

[0094] In an embodiment, the method further comprises, at the node, periodically receiving beacon frames from an access point, each beacon frame broadcasting network information related to the communication network to the plurality of nodes,

[0095] wherein at least one of the received beacon frames comprises the legacy and MU values of the queue contention parameter of the plurality of traffic queues and at least one initialization value to initialize the timer to the predetermined duration associated with the traffic queue.

[0096] In an embodiment, one or more of the transmitting traffic queues are set to the MU contention mode only if they successfully transmit data in the corresponding accessed resource unit. This configuration guarantees fairness. Indeed, in the idea of contention mode switching, the MU mode should be implemented only as a (here by RU) compensation of the existence of other transmission opportunities, which means a successful transmission of data.

[0097] In some embodiments, the MU value comprises a degraded arbitration inter-frame space number, AIFSN, compared to the legacy value. This configuration is easy to implement in order to directly reduce the opportunity of a particular traffic queue to access the medium by EDCA (to a desired level).

[0098] In particular, each queue backoff value can be initially selected from a corresponding contention window, decreased over time by the node to access the communication channel when reaching zero, and

[0099] The MU value of the queue contention parameter can comprise the same lower bound CWminas the legacy value minand / or the upper bound CW max , the lower bound CW min and the upper bound CW max both define a selection range for selecting the size of the contention window.

[0100] This configuration simplifies entering and exiting the MU contention mode (MU EDCA mode) because the contention window can be left unchanged. However, variants can consider having different bounds between the legacy and MU values.

[0101] In some embodiments, the method further comprises, at the node, when accessing the resource units provided by the other node within a subsequent transmission opportunity granted to the other node,

[0102] selecting data from the traffic queues of both the MU contention mode and the legacy contention mode based on the current queue backoff values associated therewith, and

[0103] transmitting the selected data in the accessed resource units within the new transmission opportunity.

[0104] Thus, in implementing the invention, a fair management of QoS is maintained.

[0105] According to a particular feature, selecting data comprises selecting data from the traffic queue associated with the lowest current queue backoff value. This preserves the EDCA-like behavior of the AC queues.

[0106] In alternative embodiments, the method further comprises, at the node, when accessing the resource units provided by the other node within a subsequent transmission opportunity granted to the other node,

[0107] selecting data from a preferred traffic queue indicated by the other node, and

[0108] transmitting the selected data in the accessed resource units within the new transmission opportunity.

[0109] According to a particular feature, the preferred traffic queue indication is comprised in a trigger frame received from the other node, the trigger frame reserving a transmission opportunity on the communication channel granted to the other node and defining resource units, RUs, thereby forming the communication channel comprising the accessed resource units.

[0110] This method enables the other node, typically an AP, to drive the QoS management.

[0111] In some embodiments of the invention, the accessed resource unit through which the data is transmitted is a random resource unit, wherein access to the resource unit is performed by competition using a separate RU contention parameter (separate from the queue contention parameter described above).

[0112] In other embodiments, the accessed resource unit through which the data is transmitted is a scheduled resource unit, wherein the scheduled resource is assigned to the node by another node.

[0113] Of course, some nodes can access the scheduled RU, while other nodes can access the random RU at the same time, which allows for a variety of nodes in MU contention mode at the same time (for use by one or more AC queues).

[0114] In some embodiments, the other node is an access point to which the node is registered in the communication network. This provision advantageously utilizes the central location of the access point.

[0115] From the perspective of the access point, this invention proposes a communication method in a communication network, the communication network including an access point and multiple nodes. Each node includes multiple service queues for serving data services according to different priorities. Each service queue is associated with a corresponding queue backoff value, which is calculated based on corresponding queue contention parameters with traditional values ​​in a conventional contention mode, and is used to compete for access to the communication channel to transmit the data stored in the service queue.

[0116] The communication method includes:

[0117] At the access point,

[0118] Access the communication channel to send a trigger frame, the trigger frame reserving transmission opportunities on the communication channel and defining a resource unit (RU) used by the node to transmit data to the access point for forming the communication channel; and

[0119] The node is sent a set of traditional values ​​for the queue contention parameters and a set of MU values ​​for the queue contention parameters that are different from the set of traditional values, as well as a set of initialization values ​​for the node timer associated with the service queue, to configure the node when two or more service queues of each node switch between MU contention mode and traditional contention mode, in which the corresponding queue contention parameter is set to the MU value, the MU contention mode being initialized based on the associated initialization value and maintained for a predetermined duration obtained by counting down from the associated timer, wherein the corresponding queue contention parameter in the MU contention mode is set to the corresponding MU value.

[0120] Accordingly, an access point in a communication network, the communication network further comprising a plurality of nodes, each node comprising a plurality of traffic queues for serving data traffic at different priorities, each traffic queue being associated with a respective queue contention parameter having a legacy value in a legacy contention mode, the respective queue contention parameter being computed for contending for access to a communication channel for transmitting data stored in the traffic queue,

[0121] The access point comprises at least one microprocessor configured to perform the following steps:

[0122] contending for access to the communication channel for transmitting data stored in the traffic queue,

[0123] transmitting to the nodes a set of legacy values of the queue contention parameter and a set of MU values of the queue contention parameter different from the set of legacy values, and a set of initialization values of a node timer associated with the traffic queues, for configuring the nodes when two or more traffic queues of each node are respectively switched between a legacy contention mode in which the respective queue contention parameter is set to a legacy value and a MU contention mode to be maintained for a predetermined duration initialized based on the associated initialization value and counted down by the associated timer, wherein the respective queue contention parameter in the MU contention mode is set to a respective MU value.

[0124] Thus, the access point can efficiently control the fairness in the network. Indeed, by means of the MU values and the values of the timers, the access point can drive the nodes to adapt their EDCA access scheme in the MU contention mode different from the legacy mode.

[0125] Preferably, the legacy values and the MU values, and the timer values can be evaluated based on a history of past transmissions from the nodes, in particular in the RUs (random or scheduled RUs) provided by the access point.

[0126] Optional features of the application are defined in the appended claims. Some of these features are explained below with reference to the method, while these features can be transposed into system features specific to any apparatus node according to the application.

[0127] In some embodiments, the set of legacy values, the set of MU values and the set of initialization values are transmitted within one or more beacon frames periodically transmitted by the access point for broadcasting network information related to the communication network to the plurality of nodes.

[0128] In yet other embodiments, the set of legacy values and the set of MU values differ for different arbitration inter-frame space numbers, AIFSNs.

[0129] In particular, the respective queue backoff value of a node can initially be selected from a corresponding contention window, which the node reduces over time by causing the queue backoff value to decrease until it reaches zero, at which time the communication channel is accessed, and

[0130] The set of legacy values and the set of MU values comprise the same lower boundary CWmin min and / or upper boundary CWmax max The lower boundary CWmin min and the upper boundary CWmax max together define a selection range from which the size of the contention window associated with the traffic queue is selected.

[0131] Another aspect of the application relates to a non-transitory computer readable medium storing a program which, when executed by a microprocessor or computer system in an apparatus, causes the apparatus to perform any of the methods as defined above.

[0132] The non-transitory computer readable medium can have similar features and advantages as those stated above and below in relation to the methods and apparatus.

[0133] Another aspect of the application relates to a method substantially as described herein with reference to the accompanying drawings Figure 5b , or Figure 11 , or Figure 11 and 12 , or Figure 11 , 12 and 14b, or Figure 11 , 12 and 14c, as shown in these figures.

[0134] At least part of the method according to the application can be implemented by a computer. The application can therefore take the form of a fully hardware embodiment, a fully software embodiment (including firmware, resident software, micro-code, etc.), or an embodiment combining software and hardware aspects that can all generally be referred to herein as a "circuit", "module" or "system". Furthermore, the application can take the form of a computer program product which can take the form of a computer program product embodied in any tangible medium having expressed in it a set of computer-usable program code.

[0135] As the application can be implemented in software, the application can be embodied as computer readable code for provision to a programmable apparatus on any suitable carrier medium. A tangible carrier medium can comprise a storage medium such as a hard disk drive, a magnetic tape device or a solid state memory device. A transient carrier medium can comprise a signal such as an electrical signal, an electronic signal, an optical signal, a sonic signal, a magnetic signal or an electromagnetic signal (e.g. a microwave or RF signal). BRIEF DESCRIPTION OF DRAWINGS

[0136] Other advantages of the present application will become apparent to those skilled in the art upon inspection of the drawings and detailed description thereof. Embodiments of the application will now be described, by way of example only, with reference to the following drawings.

[0137] Figure 1 A typical wireless communication system in which embodiments of the application can be implemented is shown;

[0138] Figure 2a , 2b An IEEE 802.11e EDCA relating to access categories is shown;

[0139] Figure 2c An example of degrading the values of the EDCA parameter set is shown;

[0140] Figure 3a An 802.11ac mechanism in which a backoff counter is decremented is shown;

[0141] Figure 3b An example of the mapping between eight priorities of traffic classes and four EDCA ACs is shown;

[0142] Figure 4a An 802.11ac channel allocation supporting channel bandwidths of 20MHz, 40MHz, 80MHz or 160MHz known in the prior art is shown;

[0143] Figure 4b An example of an 802.11ax uplink OFDMA transmission scheme in which an AP issues a trigger frame for a transmission opportunity to reserve OFDMA sub-channels (resource units) on an 80MHz channel known in the prior art is shown;

[0144] Figure 5a The status of a transmission traffic queue switching in MU EDCA mode known in the prior art is shown;

[0145] Figure 5b The status of a transmission traffic queue switching in MU EDCA mode according to embodiments of the application is shown;

[0146] Figure 6 A schematic representation of a communication apparatus or station according to embodiments of the application is shown;

[0147] Figure 7 a schematic representation of a wireless communication device according to an embodiment of the application is shown;

[0148] Figure 8 a typical transmission block of a communication node according to an embodiment of the application is shown;

[0149] Figure 9 the main steps performed by the MAC layer of a node in an embodiment of the application upon reception of new data to transmit are illustrated using a flowchart;

[0150] Figure 10 the steps for accessing the medium based on the EDCA medium access scheme in the two cases of using non-degraded EDCA parameters or using degraded EDCA parameters according to an embodiment of the application are illustrated using a flowchart;

[0151] Figure 11 the steps for accessing the resource units based on the RU or OFDMA access scheme upon reception of a trigger frame defining the RUs according to an embodiment of the application are illustrated using a flowchart;

[0152] Figure 12 the node management to switch back to the non-degraded mode according to an embodiment of the application is illustrated using a flowchart;

[0153] Figure 13 the structure of a trigger frame as defined in the 802.11ax standard is shown;

[0154] Figure 14a the structure of a standardized information element for describing the parameters of EDCA in a beacon frame is shown; and

[0155] Figure 14b and 14c an exemplary structure of a dedicated information element to transmit the degraded EDCA parameter values, and the HEMU EDCA Timer value according to an embodiment of the application is shown. DETAILED DESCRIPTION

[0156] The application will now be explained with the help of specific non-limitative typical embodiments and by referring to the attached drawings.

[0157] Figure 1 a communication system is shown in which a plurality of communication nodes (or stations) 101-107 exchange data frames via a wireless transmission channel 100 of a wireless local area network (WLAN) under the management of a central station or access point (AP) 110 to which the nodes have registered. The wireless transmission channel 100 is defined by an operating frequency band which consists of a single channel or a plurality of channels forming a composite channel.

[0158] Access to the shared wireless medium to transmit data frames is based on the CSMA / CA technique to listen to the carrier and avoid collisions by separating concurrent transmissions in space and time.

[0159] Carrier sensing in CSMA / CA is performed by both physical and virtual mechanisms. Virtual carrier sensing is achieved by transmitting control frames to reserve the medium before transmitting data frames.

[0160] Then, the source node or transmitting node, including the AP, first tries to listen to the medium being idle for at least a DIFS (for DCF Inter-Frame Space) time period before transmitting data frames by the physical mechanism.

[0161] However, if the shared wireless medium is sensed busy for the DIFS time period, the source node continues to wait until the wireless medium becomes idle.

[0162] To access the medium, the node starts a backoff counter that is designed to expire after a number of time slots chosen randomly in a so-called contention window [0, CW], where CW is an integer. This backoff mechanism or process, also called channel access scheme, is the basis of the collision avoidance mechanism that defers the transmission time by a random interval, thereby reducing the probability of collisions on the shared channel. After the backoff time period, i.e. when the backoff counter reaches zero, the source node can transmit data or control frames if the medium is idle.

[0163] One problem of wireless data communication is that the source node cannot listen while transmitting, thereby preventing the source node from detecting data corruption due to channel fading or interference or collision phenomena. The source node remains unaware of the corruption of the transmitted data frame and continues to transmit the frame unnecessarily, thereby wasting access time.

[0164] Thus, the collision avoidance mechanism of CSMA / CA provides a positive acknowledgement (ACK) of the data frame transmitted by the receiving node in case of successful reception of the frame to inform the source node that no corruption of the transmitted data frame occurred.

[0165] The ACK is transmitted at the end of the reception of the data frame, immediately after a time period called short inter-frame space (SIFS).

[0166] If the source node does not receive the ACK within a specified ACK timeout or detects a different frame being transmitted on the channel, the source node can conclude that the data frame is lost. In this case, the source node usually reschedules the frame transmission according to the backoff procedure described above.

[0167] To improve the collision avoidance efficiency of CSMA / CA, a four-way handshake mechanism is optionally implemented. One implementation is known as the RTS / CTS exchange defined in the 802.11 standard.

[0168] The RTS / CTS exchange involves the exchange of control frames to reserve the wireless medium before transmitting data frames during a transmission opportunity, called TXOP in the 802.11 standard, thus protecting the data transmission from any further collisions. The four-way CTS / RTS handshake is well known and will not be further explained here. For more details, please refer to the standard.

[0169] The RTS / CTS four-way handshake is very efficient in terms of system performance, especially for large frames, because it reduces the length of the messages involved in the contention process.

[0170] In detail, assuming perfect channel sensing by the communicating nodes, a collision can only occur if two (or more) frames are transmitted in the same slot after DIFS (DCF interframe space), or if the backoff counters of two (or more) source nodes reach zero almost at the same time. If the two source nodes use the RTS / CTS mechanism, only the RTS frame can collide. Fortunately, this collision is detected early by the source nodes when no CTS response is received.

[0171] Management of Quality of Service (QoS) has been introduced at the node level in such wireless networks through the well-known EDCA mechanism defined in the IEEE 802.11e standard.

[0172] In fact, in the original DCF standard, a communicating node comprises only one transmission queue / buffer. However, the delay in transmitting / retransmitting a previous frame prevents the communication from having QoS, since a subsequent data frame cannot be transmitted before the end of the transmission / retransmission of the previous frame.

[0173] Figure 2a and 2b The IEEE 802.11e EDCA mechanism is shown, which involves access categories in order to improve Quality of Service (QoS).

[0174] The 802.11e standard relies on a coordination function, called Hybrid Coordination Function (HCF), which has two operating modes: Enhanced Distributed Channel Access (EDCA) and HCF Controlled Channel Access (HCCA).

[0175] EDCA enhances or extends the functionality of the original access DCF method: EDCA is designed to support prioritized traffic similar to DiffServ, which is a protocol for specifying and controlling network traffic on a per-class basis so that certain types of traffic are prioritized.

[0176] EDCA is an important channel access scheme or mechanism in WLANs due to its distributed and easily deployable characteristics. This scheme uses contention parameters to compete for access to at least one communication channel in the communication network, allowing nodes to transmit locally stored data on the accessed channel.

[0177] The aforementioned deficiency of unsatisfactory QoS due to frame retransmission delays has been addressed by utilizing multiple transmission queues / buffers.

[0178] QoS support in EDCA is achieved by introducing four Access Classes (ACs) and thereby four corresponding transport / service queues or buffers (210). Typically, the four ACs are arranged in descending order of priority as follows: Voice (or "AC_VO"), Video (or "AC_VI"), Best-effort (or "AC_BE"), and Backstage (or "AC_BG").

[0179] Of course, another number of business queues can also be considered.

[0180] Each AC has its own service queue / buffer for storing the corresponding data frames to be transmitted over the network. Data frames (i.e., MSDUs) coming from the upper layers of the protocol stack are mapped to one of the four AC queues / buffers and are thus input into the mapped AC buffer.

[0181] Each AC also has its own set of queue contention parameters, which are associated with priority values, thus defining services with higher or lower priorities in the MSDU. Therefore, there are multiple service queues used to serve data services according to different priorities. Queue contention parameters typically include the Control Warrant (CW) for each service queue. min CW max The AIFSN and TXOP_Limit parameters. CW min and CW max This defines the lower and upper boundaries of the selection range for the EDCA contention window (CW) for a given traffic queue. AIFSN represents the Arbitration Inter-Frame Spacing Number and defines the number of time slots (typically 9 μs) that a node must listen for as idle media before decrementing the queue backoff value / counter associated with the considered traffic queue, excluding the DIFS interval (which defines the sum of AIFS time periods). TXOP_Limit defines the maximum size of TXOPs a node can request.

[0182] This means that each AC (and the corresponding buffer) acts as an independent DCF contention entity including the respective queue backoff engine 211. Thus, each queue backoff engine 211 is associated with each traffic queue to use the queue contention parameters and to set the respective queue backoff value / counter (selected from the CW) for contending for access to the at least one communication channel for transmitting the data stored in the respective traffic queue on the accessed communication channel.

[0183] The contention window CW and the queue backoff value / counter are referred to as EDCA variables.

[0184] This is such that the ACs within the same communication node compete with each other to access the wireless medium and to obtain a transmission opportunity using, for example, the conventional EDCA access scheme described above.

[0185] The service differentiation between the ACs is achieved by setting different queue backoff parameters (such as different CW min , CW max , AIFS and different transmission opportunity duration limits (TXOP_Limit) between the ACs. This helps to adjust the QoS.

[0186] The use of the AIFSN parameters and the queue backoff values for accessing the medium in the EDCA mechanism is explained below with reference to Figure 3a

[0187] Figure 2b The default values of the CW min , CW max and AIFSN parameters are shown.

[0188] In this table, the typical respective values of aCWmin and aCWmax are defined in the above-mentioned standard as 15 and 1023, respectively. Other values can be set by the nodes in the network (typically the access point) and shared between the nodes. This information can be broadcast in a beacon frame.

[0189] To determine the delay AIFS[i] between the detection of the medium being idle and the start of the decrement of the queue backoff value for the traffic queue "i", the node multiplies the value indicated in the AIFSN parameter of the traffic queue "i" (i.e. AIFSN[i]) by the slot duration (typically 9 microseconds) and adds this value to the DIFS duration.

[0190] As Figure 3a ​As shown, this is such that each traffic queue waits for a period of AIFS[i] (which includes a DIFS period that defers access to the medium) before decrementing its associated queue backoff value / counter. The figure shows two AIFS[i] corresponding to two different ACs. It can be seen that one priority traffic queue starts decrementing its backoff value earlier than the other less priority traffic queue. This is repeated after each new medium access by any node in the network.

[0191] This decrement deferral mechanism, in addition to using an average lower CW, makes the high priority traffic in EDCA have a higher chance of being transmitted compared to low priority traffic: a node with high priority traffic, on average, waits statistically less before sending a packet than a node with low priority traffic.

[0192] Thus, the EDCA queue backoff values or counters serve two purposes. First, they drive nodes to access the medium efficiently by reducing the risk of collision; second, they provide management of Quality of Service, QoS, by reflecting the aging of the data contained in the traffic queue (the older the data, the lower the backoff value), and thereby providing different priorities to traffic queues via different values of EDCA parameters, in particular the AIFSN parameter that delays the start of the decrement of the EDCA queue backoff value.

[0193] Reference Figure 2a The buffers AC3 and AC2 are typically reserved for real-time applications (e.g. voice AC_VO or video AC_VI transmission). These buffers have the highest and second highest priority, respectively.

[0194] The buffers AC1 and AC0 are reserved for best effort (AC_BE) traffic and background (AC_BG) traffic. The buffers AC1 and AC0 have the second lowest and lowest priority, respectively.

[0195] Each data unit MSDU arriving at the MAC layer from an upper layer (e.g. the link layer) with a priority is mapped into an AC according to a mapping rule. Figure 3b An example of mapping between eight priority traffic classes (User Priority or UP 0-7 according to IEEE 802.1d) and four ACs is shown. The data frame is then stored in the buffer corresponding to the mapped AC.

[0196] At the end of the backoff procedure for a traffic queue (or AC), the MAC controller of the transmitting node (hereafter Figure 7 transmits a data frame from this traffic queue to the physical layer for transmission onto the wireless communication network.

[0197] Since the ACs operate concurrently when accessing the wireless medium, it is possible that two ACs of the same communication node end their backoff at the same time. In this case, the virtual collision handling procedure (212) of the MAC controller operates the selection of the AC with the highest priority among the colliding ACs (as shown in Figure 3b and discards the transmission of data frames from the ACs with lower priority.

[0198] The virtual collision handling procedure then commands the ACs with lower priority to start the backoff operation again with an increased CW value.

[0199] The QoS obtained by using these ACs can be signaled in the MAC data frames (e.g. in the QoS control field included in the header of IEEE 802.1 Ie MAC frames).

[0200] To meet the growing demand for faster wireless networks to support bandwidth-intensive applications, 802.1 1 ac targets larger bandwidth transmissions via multi-channel operation. Figure 4a An 802.1 1 ac channel allocation is shown that supports 20 MHz, 40 MHz, 80 MHz or 160 MHz composite channel bandwidth.

[0201] IEEE 802.1 1 ac introduces support for a limited number of predefined subsets of 20 MHz channels to form dedicated predefined composite channel configurations that can be reserved for transmission of data by any 802.1 1 ac node on the wireless network.

[0202] The predefined subsets are shown in the diagram and correspond to 20 MHz, 40 MHz, 80 MHz and 160 MHz channel bandwidths compared to only 20 MHz and 40 MHz supported by 802.1 1 n. In fact, the 20 MHz constituent channels 300-1 to 300-8 are concatenated to form wider communication composite channels.

[0203] In the 802.1 1 ac standard, the channels in each predefined 40 MHz, 80 MHz or 160 MHz subset are contiguous within the operating band, i.e. no holes (missing channels) are allowed in the composite channel ordered in the operating band.

[0204] The 160 MHz channel bandwidth comprises two 80 MHz channels, which can or can not be frequency contiguous. The 80 MHz and 40 MHz channels comprise two 40 MHz channels and two 20 MHz channels, respectively, which are frequency adjacent or contiguous. However, the present application can include embodiments with any composition of channel bandwidths (i.e. comprising only contiguous channels within the operating band or comprising non-contiguous channels within the operating band).

[0205] The TXOP is granted to the nodes over the "primary channel" (400-3) by means of an enhanced distributed channel access (EDCA) mechanism. In practice, for each composite channel having a bandwidth, 802.11 ac designates one channel as "primary", which means that this channel is used to contend for access to the composite channel. The primary channel of 20 MHz is common to all nodes (STAs) belonging to the same basic set, i.e. managed by the same local access point (AP) or registered to the same local access point (AP).

[0206] However, in order to ensure that other legacy nodes, i.e. legacy nodes not belonging to the same set, do not use the secondary channels, it is proposed to duplicate the control frames reserving such composite channels (e.g. RTS frames / CTS frames) on each 20 MHz channel of the composite channel.

[0207] As previously addressed, the IEEE 802.11 ac standard enables to bind up to 4 or even 8 20 MHz channels. Since the number of channels is limited (19 in the 5 GHz band in Europe), channel saturation becomes a problem. In practice, in densely populated areas, the 5 GHz band will tend to saturate even with a 20 MHz or 40 MHz bandwidth usage per wireless LAN cell.

[0208] The development of the 802.11 ax standard attempts to enhance the efficiency and usage of the wireless channel in dense environments.

[0209] From this perspective, a multi-user (MU) transmission feature can be considered, allowing to take advantage of a master node, typically an AP, of multiple simultaneous transmissions in both downlink (DL) and uplink (UL) directions with respect to different users. In uplink, multi-user transmission can be used to reduce the probability of collision by allowing multiple nodes to transmit to the AP simultaneously.

[0210] In order to actually perform such multi-user transmission, it has been proposed to split the granted 20 MHz channel (400-1 to 400-4) into sub-channels 410 (basic sub-channels, also referred to as subcarriers or resource units (RUs)), wherein multiple users share these sub-channels 410 in the frequency domain, e.g. based on an orthogonal frequency division multiple access (OFDMA) technique.

[0211] Reference is made to Figure 4b This situation is illustrated.

[0212] The multi-user feature of OFDMA enables a node, typically an access point, i.e. an AP, to assign different RUs to different nodes to increase contention. This can help to reduce contention and collisions within the 802.11 network.

[0213] In contrast to MU downlink OFDMA (where the AP can directly send multiple data to multiple nodes (supported by specific indications within the PLCP header)), a triggering mechanism has been adopted for the AP to trigger MU uplink communication from each node.

[0214] To support MU uplink transmission (during AP-preempted TXOP), the 802.11ax AP must provide two legacy nodes (non-802.11ax nodes) to set their NAV and the 802.11ax nodes to determine the signaling information used for resource unit allocation.

[0215] In the following description, the term "traditional" refers to a non-802.11ax node, which means an 802.11 node that does not support prior technology for OFDMA communication.

[0216] like Figure 4b As shown in the example, the AP sends a trigger frame (TF) 430 to the target 802.11ax node. The TF frame informs the bandwidth or width of the target composite channel, meaning a value of 20MHz, 40MHz, 80MHz, or 160MHz. The TF frame is transmitted on the primary 20MHz channel and copied (repeated) on each of the other 20MHz channels, thus forming the target composite channel. As described above regarding the copying of control frames, it can be expected that the nearby legacy nodes (non-HT or 802.11ac nodes) that receive the TF frame (or its copy) on the primary channel will then set their NAV to the value specified in the TF frame. This prevents these legacy nodes from accessing the channels in the target composite channel during the TXOP.

[0217] Based on the AP's decision, the trigger frame TF can define multiple resource units (RUs) 410 or "random RUs" that can be randomly accessed by nodes in the network. In other words, the random RUs in the TF, specified or assigned by the AP, can serve as the basis for competition between nodes that intend to access the communication medium to send data. A collision occurs when two or more nodes attempt to transmit simultaneously on the same RU.

[0218] In that case, the trigger frame is called the Trigger Frame for Random Access (TF-R). The TF-R can be transmitted by the AP to allow multiple nodes to perform MU UL (Multi-User Uplink) random access to obtain the RU used by these nodes for UL transmission.

[0219] In addition to, or instead of, random RUs, the trigger frame TF can also specify scheduled resource units. Scheduled RUs can be reserved by the AP for certain nodes, in which case no contention is needed for accessing these RUs for these nodes. These RUs and their respective scheduled nodes are indicated in the trigger frame. For example, a node identifier (such as an association ID (AID) assigned to each node at registration) is added in the TF frame in association with each scheduled RU to explicitly indicate the nodes allowed to use each scheduled RU.

[0220] An AID equal to 0 can be used to identify a random RU.

[0221] The multi-user feature of OFDMA enables the AP to assign different RUs to different nodes to increase contention. This can help to reduce contention and collisions within the 802.11 network.

[0222] In the example of Figure 4b Each 20 MHz channel (400-1, 400-2, 400-3 or 400-4) is sub-divided in the frequency domain into four sub-channels or RUs 410 (typically 5 MHz in size).

[0223] Of course, the number of RUs that sub-divide a 20 MHz channel can be different from four. For example, 2 to 9 RUs can be provided (each of 10 MHz to about 2 MHz in size).

[0224] Once a node has used a RU to transmit data to the AP, the AP replies with an acknowledgement ACK (not shown in the figures) to acknowledge the data on each RU, so that each node can know when its data transmission was successful (ACK received) or not (no ACK after a timeout expires).

[0225] The document IEEE 802.11-15 / 1105 provides a typical random allocation procedure that can be used by a node to access a random RU indicated in a TF. This random allocation procedure (called RU contention scheme) is managed by a dedicated RU access module separate from the channel access module described above, and is configured to manage access to at least one resource unit provided by another node (typically the AP) within a transmission opportunity granted to the other node on a communication channel, in order to transmit locally stored data on the accessed resource unit. Preferably, the RU access module comprises a RU backoff engine separate from the queue backoff engine, which uses RU contention parameters (including a computed RU backoff value) to contend for access to a random RU.

[0226] In other words, the RU contention scheme is based on a new backoff counter within 802.11ax nodes, called OFDMA or RU backoff counter / value (or OBO), to allow dedicated contention when accessing a random RU to transmit data.

[0227] Each node STA1-STAn is a transmission node with respect to the reception of the AP, and as a result, each node has an active RU backoff engine separate from the queue backoff engine, for calculating the RU backoff value (OBO) to be used for contention for accessing at least one random resource unit partitioning the transmission opportunity granted on the communication channel, to transmit data stored in any traffic queue AC.

[0228] The random allocation procedure in this document, for a node having an active RU backoff value OBO among a plurality of nodes, comprises the following steps: a first step for determining, from the trigger frame, random sub-channels or RUs of the communication medium available for contention; a second step for verifying whether the value of the active RU backoff value OBO local to the node under consideration is not greater than the number of random RUs detected as available; and then, in the case of successful verification, a third step for randomly selecting a random RU among the random RUs detected as available to transmit data. In the absence of verification of the second step, a fourth step is performed (instead of the third step) to decrement the RU backoff value OBO by the number of RUs detected as available.

[0229] As illustrated, some resource units can not be used (410u), because the nodes having a RU backoff value OBO smaller than the number of available random RUs do not randomly select one of these random RUs, while some other nodes collide (as illustrated 410c) (because two of these nodes have randomly selected the same RU).

[0230] The MU uplink (UL) medium access scheme, which includes both scheduled RUs and random RUs, proves to be very efficient compared to the legacy EDCA access scheme. This is because both the number of collisions generated by simultaneous medium access attempts and the overhead due to medium access are reduced.

[0231] However, the EDCA access scheme and the MU UL OFDMA / RU access scheme must coexist, in particular to allow legacy 802.11 nodes to access the medium and even to allow 802.11ax nodes to initiate communications with nodes other than the AP.

[0232] While the EDCA access scheme taken alone provides a fair access to the medium in all nodes, its association with the MU UL OFDMA / RU access scheme introduces a drift in fairness. This is because, compared to legacy nodes, 802.11ax nodes have additional opportunities to transmit data by using the resource units provided in the transmission opportunities granted to another node, in particular the AP.

[0233] To restore some fairness between nodes, solutions have been proposed.

[0234] For example, in copending UK application 1612151.9, filed on July 13, 2016, the current value of at least one EDCA parameter is modified to a different value (MU EDCA parameter) upon successful transmission of data via the accessed resource unit, i.e. via UL OFDMA transmission. This is to reduce the probability of the node to access the communication channel via (legacy EDCA) contention.

[0235] In this framework, it is proposed to reduce the probability of the node to transmit based on EDCA (i.e. using the EDCA medium access scheme) immediately upon successful transmission of its data using the MU UL mechanism. This reduction is done by modifying well-known EDCA parameters.

[0236] The proposed mechanism as described in the document entitled "Proposed text changes for MU EDCA parameters" sets each transmission traffic queue in MU EDCA mode in response to successful transmission of data in the accessed MU UL OFDMA resource unit. This setting is done for a predetermined duration known as HEMUEDCATimer. The MU EDCA mode is the mode where each EDCA parameter is set to a MU value different from the legacy value used in the different legacy EDCA mode.

[0237] To switch from the legacy EDCA contention access mode to the MU EDCA mode, the node can modify its EDCA parameters (AIFSN, CW min and / or CW max ) for all traffic queues that successfully transmitted some data in the accessed resource unit. The switch back to the legacy EDCA mode can occur upon expiration of the HEMUEDCATimer, noting that this timer is reset to its initial value each time the node transmits new data (from any AC) during a new accessed resource unit provided by the AP. It is suggested that the HEMUEDCATimer is initialized with a high value (e.g. tens of milliseconds) so as to encompass several new opportunities for MU UL transmission.

[0238] The MU value of the EDCA parameter can be transmitted by the AP in a dedicated information element (typically sent within a beacon frame that broadcasts network information to the nodes).

[0239] The disclosed method proposes to increase the value of the AIFSN for each transmission traffic queue only, while keeping the CW min and CW max unchanged. With the increase of the respective AIFS time period, it is prevented (or at least substantially delayed) that each traffic queue in the MU EDCA mode decrements its queue backoff value or counter when the medium is sensed to be idle. During the predetermined duration, new accesses to the medium using the EDCA access scheme are statistically significantly reduced, or even not possible anymore.

[0240] The MU mode AIFSN value can be very restrictive. Thus, in a high density environment where the medium is busy most of the time (and thus idle for very short time periods), a node in MU EDCA mode has to wait for the respective very restrictive AIFS time period and thus does not decrement the backoff value of the AC queue in MU EDCA mode very often. The result is that the node cannot compete for EDCA access to the medium very often.

[0241] It is noted that the specific configuration in the publication tends to completely prevent the transmission traffic queue from EDCA access to the medium when in MU EDCA mode (except for the case that the network is not used at all). The AP specifies this specific mode of operation by indicating a specific value (typically 0) for the AIFSN parameter in the set of MU EDCA parameters. This specific value means for the node that it should use a very high value for its AIFSN, which is equal to the HEMUEDCATimer (hint: the value of the HEMUEDCATimer should be high, i.e. in the order of tens of milliseconds, compared to the worst AIFS[i] in the legacy EDCA mode which is less than 0.1 milliseconds).

[0242] Unfortunately, as long as the node accesses the OFDMA RU periodically for transmission of data, the traffic queue of the node in MU EDCA mode remains in the same MU mode. This is particularly true for those traffic queues in MU mode that do not even send any data in the accessed OFDMA RU during the potentially lengthy time period of the periodic OFDMA access. This is contrary to the QoS principles as described in the 802.1 le standard.

[0243] Reference is now made to Figure 5a to illustrate the case where Figure 5a Examples of the application using the MU EDCA parameters as described in the above publication are illustrated.

[0244] In the scenario of this figure, the AP 501 polls the node 502 by sending a standardized trigger frame 1300, requesting the node to transmit some QoS data from the AC_VI access category. This can be done by providing the node with one or more scheduled RUs. This category can be indicated in the "preferred AC" field 1330 shown. Figure 13

[0245] After a SIFS time, the node 502 initiates a MU UL OFDMA transmission 510 by picking up some QoS data (511) from the requested traffic queue AC_VI. In this exemplary scenario, there is not enough QoS data ready to be sent in the requested traffic queue AC_VI. In this context, the node 502 is allowed to retrieve other QoS data (512) from a higher priority traffic queue, e.g. the AC_VO access category in this example. This data retrieval rule makes it possible to maximize the bandwidth usage as specified in the 802.11 standard.

[0246] Thus, the node 502 transmits the AC_VI data 511 and the AC_VO data 512 to the AP using the scheduled RUs. The two respective transmission traffic queues AC_VI and AC_VO are thereby switched to the MU EDCA mode (indicated by the white figure in the black box), in which the node 502 now uses the MU EDCA parameters for each of these transmission traffic queues. In particular, a higher value of the AIFSN parameter can be used, and optionally a higher value of the CW min and CW min parameters.

[0247] In parallel, the HEMU EDCA Timer 590 is started to count down while the node 502 will be allowed to switch back to the legacy EDCA mode with the legacy EDCA parameters. The "switch back" can occur after the expiration of a predetermined duration, i.e. when the HEMU EDCA Timer reaches 0.

[0248] However, each time the node 502 transmits data in a granted resource unit provided by the AP within any subsequent transmission opportunity on the communication channel, the HEMU EDCA Timer is reinitialized to its initial value (the predetermined duration). In other words, each time the node 502 is polled again by the AP, the timer is reinitialized.

[0249] This occurs in the example of Figure 5a where the AP 501 sends a new trigger frame 1300-2 with new RUs to the node 502, while the HEMU EDCA Timer 590 has not yet expired. The AP polls the node 502 again to send QoS data from the AC_VI access category. ​

[0250] The node 502 transmits again QoS data 520 from the AC_VI access category, and the HEMU EDCA Timer 590 is reinitialized to its initial value, i.e. the predetermined duration. The same happens when the AP 501 polls again the node 502 for a new QoS data from the AC_VI access category by sending a new trigger frame 1300-3.

[0251] In this scenario, the node 502 is periodically polled by the AP for OFDMA transmission of QoS data from the AC_VI. Eventually, the AC_VO category does not participate in new OFDMA transmissions as long as the AC_VI category provides enough data, and remains blocked in the MU EDCA mode.

[0252] In addition, the MU mode AIFSN value of the traffic queue AC_VO (typically a more restrictive value, i.e. a high value) prevents (or heavily delays) the traffic queue AC_VO from decrementing the associated backoff value for EDCA contention of the medium.

[0253] This makes the AC_VO category, which essentially has the highest QoS priority, remain locked in the MU EDCA mode without new EDCA opportunities to transmit data. Thus, the QoS requirements of 802.1 lax are still heavily degraded.

[0254] Within this framework, the present invention proposes to restore the QoS fairness by breaking the singularity of the HEMU EDCA Timer used by the traffic queue locked in the MU mode upon periodic node polling by the AP.

[0255] In particular, upon transmission (preferably successful) of data stored in two or more traffic queues in one or more granted transmission opportunities on the communication channel by another node in one or more accessed resource units provided by the other node, the node 502 can set each transmission traffic queue (i.e. transmitted in an accessed resource unit) into the MU EDCA mode different from the legacy EDCA mode and for a predetermined duration resulting from the countdown of each timer associated with the transmission traffic queue. Next, upon expiration of any timer, the node 502 can switch the associated traffic queue back to the legacy EDCA mode with each EDCA parameter set back to the legacy value.

[0256] Thus, the present invention provides to the node a plurality of timers each associated with one of the traffic queues. When a specific HEMU EDCA Timer is dedicated to each AC queue, the AC queue can exit from the MU EDCA mode independently from the other AC queues. The QoS at the AC queue level is thus restored.

[0257] Reference will now be made to Figure 5b the results of an example implementation of the application, in which Figure 5b the same sequence is illustrated for the recovery of QoS via the processing of independent HEMU EDCA Timers. Figure 5a

[0258] After the first TF 1300, both transmission traffic queues AC_VI and AC_VO are in the MU EDCA mode. When each respective traffic queue is allowed to switch back to the legacy EDCA mode with legacy EDCA parameters, each of their respective HEMU EDCA Timers (591 for AC_VI and 592 for AC_VO) are simultaneously started to count down.

[0259] According to the application, the evolution of these individual timers is independent of each other.

[0260] As described below, different predetermined durations can be used to initialize the two timers associated with AC_VI and AC_VO. This is in order to improve the QoS management.

[0261] Thus, upon reception of the next TF 1300-2 in the accessed OFDMA RU and transmission of AC_VI data as required by the AP, the HEMU EDCA Timer 591 associated with AC_VI is reinitialized with its respective initial predetermined duration, while the HEMU EDCA Timer 592 associated with AC_VO continues to elapse (as no VO data is transmitted in the accessed RU after TF 1300-2).

[0262] As a result, the HEMU EDCA Timer 592 associated with AC_VO expires before the HEMU EDCA Timer 591 associated with AC_VI, thus relaxing the MU EDCA constraints for the traffic queue AC_VO. Indeed, the traffic queue AC_VO switches back to the legacy EDCA mode using legacy EDCA parameters. Thus, the backoff value of the AC_VO traffic queue can be decreased in the normal way, thus allowing the AC_VO queue to efficiently contend for the medium.

[0263] Figure 6 A communication device 600 of the wireless network 100 is schematically illustrated, wherein the communication device 600 is configured to implement at least one embodiment of the application. The communication device 600 can preferably be a device such as a microcomputer, a workstation or a light portable device. The communication device 600 comprises a communication bus 613 to which the following components are preferably connected:

[0264] • a central processing unit 611, denoted CPU, such as a microprocessor;​

[0265] • a read-only memory 607, denoted ROM, for storing computer programs used to implement the present application;

[0266] • a random access memory 612, denoted RAM, for storing executable code of the method according to an embodiment of the application and registers configured to record variables and parameters required to implement the method according to an embodiment of the application; and

[0267] • at least one communication interface 602 connected to a wireless communication network 100, for example a wireless communication network according to the 802.1 lax protocol, via which digital data packets or frames or control frames are transmitted. Under the control of a software application running in the CPU 611, frames are written from the FIFO send memory in the RAM 612 to the transmission network interface, or frames are read from the reception network interface and written to the FIFO receive memory in the RAM 612.

[0268] Optionally, the communication device 600 can also comprise the following components:

[0269] • a data storage component 604, such as a hard disk, for storing computer programs used to implement the method according to one or more embodiments of the application;

[0270] • a disk drive 605 for a disk 606, configured to read data from the disk 606 or write data to said disk;

[0271] • a screen 609 for displaying decoded data and / or acting as a graphical interface with the user through a keyboard 610 or any other pointing component.

[0272] The communication device 600 can optionally be connected to various peripheral devices, such as a digital camera 608, each of which is connected to an input / output card (not shown) to supply data to the communication device 600.

[0273] Preferably, a communication bus provides communication and interoperability between the elements included in or connected to the communication device 600. The representation of the bus is not limiting, in particular the central processing unit can operate to communicate instructions to any element of the communication device 600, directly or through another element of the communication device 600.

[0274] The disc 606 can optionally be replaced by any information medium such as a compact disc (CD-ROM) (rewritable or not), a ZIP disc, a USB key or a memory card, and generally by an information storage component which can be read by a microcomputer or microprocessor, integrated or not into the device, possibly removable and configured to store one or more programs whose execution makes it possible to implement the method according to the application.

[0275] As mentioned previously, the executable code can optionally be stored in the read-only memory 607, on the hard disk 604 or on a removable digital medium such as the disc 606. According to an optional variant, the executable code of the program can be received via the interface 602 over the communication network 603, to be stored in one of the storage components of the communication device 600, such as the hard disk 604, before being executed.

[0276] The central processing unit 611 is preferably configured to control and direct the execution of the instructions or part of the software code of the program according to the application, stored in one of the above-mentioned storage components. At power-up, the program stored in non-volatile memory, for example on the hard disk 604 or in the read-only memory 607, is transferred to the random access memory 612 containing the executable code of the program and the registers for storing the variables and parameters necessary for implementing the application.

[0277] In the preferred embodiment, the device is a programmable device which implements the application using software. However, alternatively, the application can be implemented in hardware, for example in the form of an application-specific integrated circuit or ASIC.

[0278] Figure 7 is a block diagram schematically showing the architecture of a communication device or node 600 (in particular one of the nodes 100-107) configured to execute at least in part the application. As shown, the node 600 comprises a physical (PHY) layer block 703, a MAC layer block 702 and an application layer block 701.

[0279] The PHY layer block 703 (here the 802.11 standardized PHY layer) has the task of formatting frames, modulating frames on or demodulating frames from any 20 MHz channel or composite channel, and thus transmitting or receiving frames via the wireless medium 100 used. These frames can be 802.11 frames, for example medium access trigger frames TF 430 to define resource units in the transmission opportunities granted, MAC data and management frames based on 20 MHz width to interact with legacy 802.11 stations, and MAC data frames of OFDMA type with a width smaller than 20 MHz legacy, typically 2 MHz or 5 MHz, with respect to this wireless medium.

[0280] The MAC layer block or controller 702 preferably comprises a MAC 802.11 layer 704 implementing the conventional 802.1 lax MAC operations and an additional block 705 for performing at least partly the present application. The MAC layer block 702 can optionally be implemented in software, which is loaded into the RAM 512 and executed by the CPU 511.

[0281] Preferably, this additional block, called MU EDCA mode management module 705, implements the part of the present application relating to the node 600, i.e. manages the switching between the two conventional modes and the MU EDCA mode and handles the timers used to control the traffic queues in the MU EDCA mode.

[0282] From the AP point of view, this MU EDCA mode management module 705 can be provided to transmit to the nodes a set of conventional values of the EDCA parameters, a set of MU values of the EDCA parameters different from the set of conventional values, and a set of initialization values of the HEMU EDCA Timer, to drive these nodes into the MU EDCA mode to remain in this mode at least for a corresponding duration. Thus, these values drive the node to configure itself when one of its traffic queues switches between the conventional EDCA mode, in which each EDCA parameter is set to a conventional value, and the MU EDCA mode, in which each EDCA parameter is set to a MU value, and the predetermined duration is initialized based on the associated initialization value and decremented by the associated counter.

[0283] The MAC 802.11 layer 704 and the MU EDCA mode management module 705 interact with each other to provide the management of the channel access module handling the queue backoff engine and the RU access module handling the RU backoff engine as described below.

[0284] In the upper part of the figure, the application layer block 701 runs an application for generating and receiving data packets, e.g. of a video stream. The application layer block 701 represents all the stacked layers above the MAC layer according to the ISO standardization.

[0285] Embodiments of the present application are now illustrated using various typical embodiments. Although the proposed examples use the trigger frame 430 (see Figure 4b ) transmitted by the AP for multi-user uplink transmission, equivalent mechanisms can be used in a centralized or ad-hoc environment (i.e. without AP). This means that the operations described below with reference to the AP can be performed by any node in an ad-hoc environment.

[0286] These embodiments are illustrated mainly by considering OFDMA resource units in the context of IEEE 802.1 lax. However, the application of the present invention is not limited to the IEEE 802.1 lax context.

[0287] Furthermore, the present invention does not necessarily rely on the use of the MU access scheme as described in 802.1 lax. Any other RU access scheme defining an alternative medium access scheme allowing nodes to access the same medium simultaneously can also be used.

[0288] A set of MU values can be more restrictive than a set of legacy values, which results in traffic queues in MU EDCA mode using the EDCA contention access scheme to access the medium less often.

[0289] However, in some embodiments, the set of MU values can be more lenient.

[0290] For the sake of clarity, the following description focuses on a more restrictive set of MU values. In this context, the MU EDCA mode is referred to as a "degraded" mode, while the legacy EDCA mode is referred to as a "non-degraded" mode.

[0291] Figure 8 A typical transmission block of a communication node 600 according to an embodiment of the present invention is illustrated.

[0292] As mentioned above, the node comprises a channel access module and possibly a RU access module, both implemented in the MAC layer block 702. The channel access module comprises:

[0293] a plurality of traffic queues 210 for serving data traffic at different priorities;

[0294] a plurality of queue backoff engines 211 each associated with a respective traffic queue for using EDCA parameters, in particular for computing a respective queue backoff value to be used for contending for access to at least one communication channel, for transmitting data stored in the respective traffic queue. This is the EDCA access scheme.

[0295] According to an embodiment of the present invention, each queue backoff engine 211 has its own HEMU EDCA Timer 2110. This means that the node comprises a plurality of timers each associated with one of the traffic queues.

[0296] Furthermore, by updating the EDCA parameters according to the teachings of the present invention, an EDCA mode switch 213 for handling switching between the degraded MU EDCA mode and the legacy EDCA mode is provided in the node. The EDCA mode switch operates in response to each OFDMA transmission of the node in a RU.

[0297] The RU access module comprises a RU backoff engine 800 separate from the queue backoff engine, for using the RU contention parameters, in particular for calculating the RU backoff value to be used for contending for access to the OFDMA random resource units defined in the received TPs (for example transmitted by the AP), for transmitting the data stored in the arbitrary traffic queue in the OFDMA RUs. The RU backoff engine 800 is associated with a transmission module called OFDMA multiplexer 801. For example, when the RU backoff value OBO described below reaches zero, the OFDMA multiplexer 801 is responsible for selecting the data to be transmitted from the AC queue 210.

[0298] The conventional AC queue backoff register 211 drives the medium access request along the EDCA protocol (channel contention access scheme), while in parallel the RU backoff engine 800 drives the medium access request onto the OFDMA multi-user protocol (RU contention access scheme).

[0299] In the coexistence of the two contention access schemes, the source node implements a medium access mechanism with collision avoidance based on the calculation of the backoff value:

[0300] - the queue backoff counter value, which corresponds to the number of time slots (in addition to the DIFS period) that the node waits after the communication medium has been detected idle before accessing the medium. This is EDCA, whether in the downgraded state or in the non-downgraded state;

[0301] - the RU backoff counter value (OBO), which corresponds to the number of idle random RUs detected by the node after the TXOP has been granted to the AP or to any other node on the composite channel made up of RUs before accessing the medium. This is OFDMA. A variant to count down OBO based on the number of idle random RUs can be based on a time-dependent countdown.

[0302] Figure 9 The main steps performed by the MAC layer 702 of the node 600 when receiving new data to be transmitted are illustrated using a flowchart. Figure 9 The conventional FIFO feed in the 802.11 context is illustrated.

[0303] At the outset, the traffic queue 210 stores no data to be transmitted. As a result, no queue backoff value 211 is calculated. The corresponding queue backoff engine or the corresponding AC (Access Category) is considered inactive. When data is stored in the traffic queue, the queue backoff value is calculated immediately (according to the corresponding queue backoff parameters) and the associated queue backoff engine or AC is considered active.

[0304] When the node has data ready to be transmitted on the medium, the data is stored in one of the AC queues 210 and the associated backoff 211 should be updated.

[0305] In step 901, new data is received from an application running on the device (e.g. from the application layer 701), from another network interface or from any other data source. The new data is ready to be transmitted by the node.

[0306] In step 902, the node determines in which AC queue 210 the data should be stored. This operation is typically performed by checking the TID (Traffic Identifier) value attached to the data (according to the matching shown). Figure 3b

[0307] Then, step 903 stores the data in the determined AC queue. This means that the data is stored in the AC queue having the same data type as the data.

[0308] In step 904, a conventional 802.11 AC backoff calculation is performed with the queue backoff engine associated with the determined AC queue.

[0309] If the determined AC queue is empty (i.e. the AC is not active) immediately before the storage of step 903, a new queue backoff value for the corresponding backoff counter needs to be calculated.

[0310] Thus, the node calculates the queue backoff value as equal to a random value selected in the range [0, CW], where: CW is the current value of the CW of the considered access category (as defined in the 802.11 standard). It should be reiterated that the queue backoff value will be added to the AIFSN (possibly degraded in MU EDCA mode) in order to implement the relative priority of the different access categories. CW is a congestion window value selected from the range [CW min ,CW max ], where the two bounds CW min and CW max (possibly degraded) depend on the considered access category.

[0311] As a result, the AC becomes active.

[0312] The above parameters CW, CW min , CW max , AIFSN and backoff value form the EDCA parameters and the variables associated with each AC. These values are used to set the relative priority of accessing the medium for data of different categories.

[0313] The EDCA parameters typically have fixed values (e.g. CW min , CW max ​and AIFSN), while the EDCA variables (CW and backoff values) evolve over time and medium availability. As will be readily apparent from the above, the present application provides for the evolution of the EDCA parameters through the switching between the degrading and non-degrading parameter values.

[0314] Furthermore, if needed, step 904 can comprise computing a RU backoff value OBO. If the RU backoff engine 800 is not active (e.g. because there is no data in the traffic queue prior to the previous step 903), and if new data to be addressed to the AP has been received, then the RU backoff value OBO needs to be computed.

[0315] The RU backoff value OBO can be computed in a similar way as the EDCA backoff value, i.e. using dedicated RU contention parameters such as dedicated contention window [0, CWO] and selection range [CWO min ,CWO max , etc.

[0316] Note that some embodiments can provide a distinction between data that can be sent via resource units (i.e. compatible with MU UL OFDMA transmission) and data that cannot be sent via resource units. Such a decision can be made during step 902 and a corresponding flag item can be added to the stored data.

[0317] In this case, the RU backoff value OBO is computed only if the newly stored data is flagged as compatible with MU UL OFDMA transmission.

[0318] After step 904, Figure 9 the process ends.

[0319] Once data is stored in the AC queue, the node can access the medium through the EDCA access scheme as illustrated below with reference to Figure 10 in legacy EDCA mode or in degraded MU EDCA mode, or directly through the resource units provided by the AP via one or more trigger frames as illustrated below with reference to Figure 11 .

[0320] Figure 10 The steps for accessing the medium based on the (legacy or degraded MU) EDCA medium access scheme are illustrated using a flowchart.

[0321] Steps 1000-1020 describe the legacy wait introduced in the EDCA mechanism to reduce collisions on the shared wireless medium. In step 1000, the node 600 listens to the medium to wait for the medium to become available (i.e. the detected energy is below a given threshold on the primary channel).

[0322] During the AIFS[i] time interval (which includes the DIFS time interval and the AIFSN[i] time interval - see Figure 3a When the medium becomes idle, step 1010 is executed, in which node 600 decrements the backoff counter 211 of all active (non-zero) AC[] queues by 1. In other words, the node decrements the queue backoff value for each basic time unit in which the communication channel is detected as idle.

[0323] Next, in step 1020, node 600 determines whether at least one of the AC backoff counters has reached 0.

[0324] If no AC queue backoff reaches zero, node 600 waits for another backoff slot (typically 9 μs) and thus loops back to step 1000 to listen to the medium again during the next backoff slot. This allows the AC backoff counters to be decremented at each new backoff slot as soon as the medium is listened to as idle, once the respective AIFS[i] of the AC backoff counters has expired.

[0325] If at least one AC queue backoff reaches 0, then step 1030 is executed, in which node 600 (more precisely, virtual conflict handler 212) selects the active AC queue with a zero queue backoff counter and the highest priority.

[0326] In step 1040, an appropriate amount of data is selected from the chosen AC for transmission to match the bandwidth of the TXOP.

[0327] Next, in step 1050, if, for example, an RTS / CTS exchange has been successfully performed to grant a TXOP, node 600 initiates an EDCA transmission. Thus, node 600 transmits the selected data on the medium during the granted TXOP.

[0328] Next, at step 1060, node 600 determines whether the EDCA transmission has ended. If it has ended, step 1070 is executed.

[0329] At step 1070, node 600 updates the contention window CW of the selected traffic queue based on the transmission status (positive or negative ACK, or no ACK received). Typically, in the event of a transmission failure, node 600 doubles the value of CW until CW reaches the maximum value CW of the data-dependent AC type. max (Degraded or non-degraded) until then. On the other hand, if the EDCA transmission is successful, the contention window CW is set to the minimum CW of the AC type, which also depends on the data. min (Downgraded or not downgraded).

[0330] Next, if the selected traffic queue is not empty after the EDCA data transmission, a new associated queue backoff counter is randomly selected from [0, CW] similar to step 904.

[0331] This ends the processing of Figure 10 .

[0332] Figure 11 The steps of accessing a resource unit based on a RU or OFDMA access scheme upon reception of a trigger frame defining the RUs are illustrated using a flow chart. This illustrates for example the behavior of the node 502 in Figure 5b .

[0333] In step 1110, the node judges whether a trigger frame is received from an access point in the communication network, the trigger frame reserving a transmission opportunity on the communication channel granted to the access point and defining resource units RUs forming the communication channel. If a trigger frame is received, the node analyzes the content of the received trigger frame.

[0334] In step 1120, the node judges whether the node can transmit data on one of the RUs defined in the received trigger frame. The judgment can involve one or both of two conditions, in particular regarding the type of RUs.

[0335] By analyzing the content of the received TF, the node judges whether the defined RUs are scheduled resource units assigned to the node by the access point. This can be done by looking for its AID in the received TF, the AID being associated to a specific scheduled RU to be used for the MU UL OFDMA transmission.

[0336] Furthermore, by analyzing the content of the received TF, the node judges whether one or more random RUs are defined in the TF, i.e. RUs accessed by contention using dedicated RU contention parameters (including the above-mentioned OBO value 800). In this case, the node further judges whether its current OBO value 800 allows to select a random RU (e.g. in case the OBO 800 is smaller than the number of random RUs in the TF).

[0337] If one scheduled RU is assigned to the node or the node is allowed to access (after contention) one random RU, the node determines the size of the random / scheduled RU to be used and performs step 1130. Otherwise, the node decrements the RU backoff value OBO 800 based on the number of random resource units defined in the received trigger frame and the processing ends when the node cannot access any RU defined by the received TF.

[0338] In step 1130, the node selects at least one of the traffic queues 210 from which data is to be transmitted and adds data of the selected queue to the transmission buffer until the amount of data reaches the size of the selected resource unit to be used.

[0339] Various criteria to select the current traffic queue can be involved.

[0340] For example, this can be done by:

[0341] Selecting the traffic queue 210 with the lowest associated queue backoff value. Thus, the selection of the traffic queue depends on the value of the EDCA backoff 211, guaranteeing that the node complies with the EDCA principle and that the correct QoS is achieved for the data of this node;

[0342] Randomly selecting one non-empty traffic queue from the traffic queues;

[0343] Selecting the traffic queue storing the largest amount of data, i.e. the most heavily loaded;

[0344] Selecting the non-empty traffic queue with the highest associated traffic priority (taking into account the Figure 3b indicated AC categories);

[0345] Selecting the non-empty traffic queue associated with a data type matching the data type associated with the resource unit on which the data to be transmitted is to be transmitted. Such a specified data type can be the traffic queue indicated in the trigger frame by the AP using the Figure 13 preferred AC field 1340 set to 1, for example. This is the selection criterion used in the example of Figure 5b

[0346] After step 1130, step 1140 provides that the node sets or updates the list of transmit / transmission queues by inserting the current traffic queue from which the data selected in step 1130 comes. This list maintains the insertion order of the transmit / transmission queues, so that, for example, the primary transmit / transmission queue (the first queue selected in step 1030) and the subsequent transmit / transmission queues can be easily identified.

[0347] In addition, the node can store, during step 1140, an information item representative of the amount of data thus selected from the current traffic queue, in order to be transmitted in the RU. For example, the node then updates the list of transmit queues by also inserting the amount of data selected from the current traffic queue.

[0348] This list of transmit / transmission queues can be implemented by a table containing, for each traffic queue, the rank of the transmission queue (which can be simplified to "primary" or "secondary" queue) and the amount of data put in the transmission buffer. ​

[0349] In step 1150, the node judges whether the amount of data stored in the transmission buffer is sufficient to fill the selected resource unit.

[0350] If not, there is still space in the resource unit for additional data. Thus, the process loops back to step 1130, during which another traffic queue can be selected using the same selection criteria. In this way, the transmission buffer is filled step by step to reach the size of the selected resource unit.

[0351] Thus, it can be noted that multiple transmission traffic queues of the same node can be involved during a MU UL OFDMA transmission, thus causing multiple queues to enter the MU EDCA mode.

[0352] In a variant where mixing data from two or more traffic queues is avoided (i.e. data of the selected RU is selected from a single traffic queue), padding data can be added to completely fill the selected RU. This is to ensure that the entire RU duration has energy that can be detected by legacy nodes.

[0353] In another variant implementing a specific data aggregation rule, if the first selected traffic queue does not have enough data to completely fill the accessed resource unit, data from a higher priority traffic queue can be selected.

[0354] Once the transmission buffer is full for the selected RU, step 1160 initiates the MU UL OFDMA transmission of the data stored in the transmission buffer to the AP. The OFDMA transmission is based on the OFDMA sub-channels and modulation defined in the received trigger frame and in particular in the RU definition.

[0355] Then, once the transmission has taken place, and preferably upon a successful transmission (i.e. an acknowledgement is received from the AP), step 1170 determines new values of the EDCA parameter(s) to be applied to the traffic queue(s) in order to modify the value(s) to a penalizing value.

[0356] Thus, the transmission queues added in the list in step 1140 enter the MU EDCA mode, which means that the EDCA or "queue contention" parameter set of these transmission queues should be modified, in particular to the determined degrading parameter values. One or more transmission queues can already be in the MU EDCA mode. However, the degrading parameter values are also determined (which can be modified by the most recently received beacon frame with new degrading values).

[0357] During step 1170, the degrading parameter values are determined.

[0358] In an embodiment, the degraded value of the EDCA parameter comprises a degraded arbitration interframe space number AIFSN compared to the non-degraded value of the EDCA parameter used by the traffic queue not set in the MU EDCA mode. In other words, the AIFSN of the transmission queue is set to a degraded value. Figure 2c A degraded EDCA parameter set is shown with a degraded AIFSN value.

[0359] In some embodiments, the AIFSN is the only parameter modified when switching to the MU EDCA mode. This means that the degraded value of the EDCA parameter comprises the same lower boundary CWmin min and / or upper boundary CWmax max as the non-degraded value used in the legacy EDCA mode. min and CWmax max both define a selection range for the size of the selection contention window.

[0360] The degraded value used in this step is preferably selected in a dedicated information element last received which usually forms part of a beacon frame transmitted by the AP. Thus, for a node receiving periodically a beacon frame from the access point (each beacon frame broadcasting network information related to the communication network to a plurality of nodes), the received beacon frame thus usually comprises, in addition to the non-degraded (or legacy EDCA) values, the degraded values of the EDCA parameters for the plurality of traffic queues switching to the MU EDCA mode.

[0361] If no such degraded values are received from the AP, the default values as described in the standard can be used.

[0362] Step 1170 also comprises determining, for each transmission traffic queue AC, a predetermined degraded duration HEMUEDCATimer[AC] value. This duration defines the period of time for which the node has to remain in the MU EDCA mode for the associated degraded traffic queue. This information can also be obtained from the AP (for example, from a specific dedicated information element of the received beacon frame as shown in Figure 14b or 14c below).

[0363] After step 1170, step 1180 actually replaces the current values of the EDCA parameters associated with the transmission traffic queues with the degraded values determined in step 1170.

[0364] In case the parameters CW min and / or CW max have new values, the current CW of one or more traffic queues can be outdated. In this case, a new CW can be selected from the newly defined range [CW min , CW max ].

[0365] Next, in step 1190, the timer 2110 associated with each transmission traffic queue 210 is initialized with the corresponding predetermined degrading duration HEMUEDCATimer[AC] as determined in step 1170. The timer 2110 then starts and gradually elapses over time.

[0366] Note that if the timer has elapsed when step 1180 is performed (which means that the associated traffic queue is already in the MU EDCA mode), the timer is again initialized (i.e. reset) to the HEMUEDCATimer[AC] value to keep the node in the MU EDCA mode for the next HEMUEDCATimer[AC] period. This is the case of the timer 591 in the example of Fig. 5. Figure 5b

[0367] Figure 12 The node management at the queue level to switch back to the non-degraded legacy mode in the above example is illustrated using a flowchart. This management is based on the HEMUEDCATimer[AC] dedicated to the traffic queue AC of interest. Indeed, as long as this timer HEMUEDCATimer[AC] has not expired, the traffic queue AC can remain in the MU EDCA mode.

[0368] Thus, in step 1210, it is checked whether the HEMUEDCATimer[AC] has expired, i.e. reached the value 0.

[0369] In the affirmative, in step 1220, the traffic queue AC is switched back to the EDCA mode. This can include resetting the EDCA parameters to the non-degraded values (e.g. the values provided by the AP to the node using the beacon frame of the following Figure 14a ).

[0370] Note that since the timer is reinitialized in each new step 1190, the expiration of the HEMUEDCATimer[AC] only occurs in the case where no data from the traffic queue AC is transmitted from the node in any OFDMA resource unit provided by the AP in the subsequent TXOP granted by the AP for the predetermined degrading duration.

[0371] Next, the process ends in step 1230.

[0372] The process of Figure 12 is performed in parallel and independently for each traffic queue in the degraded MU EDCA mode (i.e. for which the timer is elapsing). This is because the timers 2110 are separate according to the teachings of the present application.

[0373] Figure 13 ​The structure of a trigger frame as defined in the 802.11ax draft standard is shown.

[0374] The trigger frame 1300 comprises a dedicated field 1310 called User Info field. This field contains a "Trigger Dependent Common Info" field 1320 which contains an "AC

[0375] The Preferred AC field 1340 is a 2-bit field indicating the AC queue (values from 0 to 3) from which data should be sent by the node on the RUs allocated to the node in the trigger frame.

[0376] The AC Preferred Level field 1330 is a bit indicating whether the value of the Preferred AC field 1340 is meaningful or not. If the field 1340 is set to 1, the node should take into account the Preferred AC field 1340 when selecting data at step 1130. If the field 1330 is set to 0, the node is allowed to send data from any AC queue regardless of the value of the Preferred AC field 1340.

[0377] Other fields of the trigger frame are defined in the 802.11ax standard.

[0378] The AP can also be in charge of broadcasting the EDCA parameters for both EDCA and MU EDCA modes, and one or more initialization values to be used to initialize or reset the timers 2110 associated with the traffic queues 210. The AP preferably uses the well-known beacon frame dedicated to configure all the nodes in the 802.11 cell for the broadcast. Note that if the AP cannot broadcast the EDCA parameters, the nodes are configured to fall back to the default values as defined in the 802.11ax standard.

[0379] Figure 14a The structure of a standardized information element 1410 used to describe the parameters of the EDCA in the beacon frame is shown.

[0380] The fields 1411, 1412, 1413, 1414 describe the parameters associated with each traffic queue 210. For each traffic queue, the subfield 1415 includes the EDCA parameters: AIFSN as the delay before starting to decrease the associated backoff value, CWmin and CWmax as the values of the contention window, and finally TXOP limit as the maximum transmission data time of the 802.11 device. min max

[0381] All the other fields of the information element are fields described in the 802.11 standard.

[0382] Figure 14b ​​An exemplary structure of a dedicated information element 1420 for transmitting the degraded EDCA parameter values according to the application, and a common initialization value of the timers HEMUEDCATimer[AC] for all traffic queues is shown. The dedicated information element 1420 can be included in the beacon frame transmitted by the AP.

[0383] The dedicated information element 1420 includes, for each AC queue, the degraded EDCA parameters (1421, 1422, 1423, 1424) to be used by the nodes in MU EDCA mode. The dedicated information element 1420 also includes a subfield 1425 for specifying a common initialization value of the HEMUEDCATimer.

[0384] Each subfield 1421, 1422, 1423, 1424 includes the degraded AIFSN value for the corresponding traffic queue, and the degraded ECWmin value and the degraded ECWmax value (which can be the same as the conventional EDCA values).

[0385] In the present embodiment, the predetermined degraded duration for initializing the timers HEMUEDCATimer[AC] associated with each traffic queue AC is computed according to the common initialization value 1425 received from the AP and according to an adjustment parameter specific to each corresponding traffic queue.

[0386] By using different adjustment parameters, different predetermined degraded durations for initializing the timers associated with two corresponding traffic queues can be obtained.

[0387] In one embodiment, the common initialization value as provided by the AP can be multiplied by a constant value (adjustment parameter) based on the priority of each traffic queue AC. For example, the constant value can be equal to 1 for the AC_VO and AC_VI access categories, and equal to 3 for the AC_BE and AC_BG access categories.

[0388] Figure 14c Another exemplary structure of a dedicated information element 1430 for transmitting the degraded EDCA parameter values according to the application, and one initialization value of each timer HEMUEDCATimer[AC] implemented by the nodes is shown. The dedicated information element 1430 can be included in the beacon frame transmitted by the AP.

[0389] The dedicated information element 1430 includes, for each AC queue, a set of degraded parameters (1431, 1432, 1433, 1434) to be used by the nodes in MU EDCA mode. The dedicated information element 1430 also includes a subfield 1425 for specifying a common initialization value of the HEMUEDCATimer.

[0390] Each subfield 1431, 1432, 1433, 1434 comprises the downgraded AIFSN value of the corresponding traffic queue, as well as the downgraded ECWmin value and the downgraded ECWmax value (which can be the same as the conventional EDCA values), and finally the initialization value of the HEMU EDCA Timer to be used for the traffic queue of interest.

[0391] This means that the AP is responsible for calculating and then transmitting dedicated initialization values for each traffic queue. In the present embodiment, the predetermined downgraded duration to be used for initializing the timer HEMU EDCA Timer [AC] associated with each traffic queue is set to the corresponding initialization value received directly from the AP.

[0392] In order to improve the QoS management, the initialization values calculated by the AP are preferably based on the priority of each AC.

[0393] Although the present application has been described above with reference to particular embodiments, the present application is not limited to these specific embodiments and modifications will be apparent to those skilled in the art within the scope of the present application.

[0394] For example, although in the above explanation the EDCA parameters and the downgraded MU EDCA parameters are broadcast in a dedicated information element of the same beacon frame, a variant can provide for alternating between a beacon frame transmitting the EDCA parameters and another beacon frame broadcasting the downgraded MU EDCA parameters.

[0395] Many other modifications and variations of the above-described illustrative embodiments of the application will be apparent to those skilled in the art from consideration of the foregoing description of the application that is presented only by way of example and not of limitation. It is intended that the present application include all such modifications and variations as fall within the scope of the appended claims. In particular, it is intended that only those features of the specific embodiments that are absolutely necessary for the practice of the present application be described as essential. Particularly, different features from different embodiments can be interchanged and still be within the scope of the application. Accordingly, the application is limited only by the following claims.

[0396] In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite articles "a" or "an" do not exclude a plurality. The mere fact that different features are recited in mutually different dependent claims does not indicate that a combination of these features cannot be used to advantage.

Claims

1. A communication apparatus comprising: a transmitter to transmit data classified in any of a plurality of different ACs using a contention parameter in an EDCATXOP, wherein each of the different ACs has its own contention parameter, EDCA is enhanced distributed channel access, TXOP is transmission opportunity, and AC is access category; and a plurality of timers, each timer associated with one of the ACs, wherein, in a case where data classified in a particular AC of the plurality of different ACs is successfully transmitted in a multi-user resource unit provided by the access point within a transmission opportunity granted to the access point, a particular timer of the plurality of timers associated with the particular AC is updated to maintain the contention parameter set to a MU-EDCA parameter value for a predetermined duration before being set back to an EDCA parameter value different from the MU-EDCA parameter value for the particular AC.

2. The communication apparatus of claim 1, configured to re-initialize the particular AC to the predetermined duration each time data from the particular AC is transmitted in a multi-user resource unit provided by the access point within a subsequent transmission opportunity granted to the access point.

3. The communication apparatus of claim 1, further comprising a receiver that periodically receives a beacon frame from the access point, wherein, the received beacon frame includes the EDCA parameter value, the MU-EDCA parameter value, and a value to set the timer associated with the particular AC to the predetermined duration.

4. The communication apparatus according to claim 1, wherein the contention parameter for the particular AC is set to the MU-EDCA parameter value only when data from the particular AC is successfully transmitted in a multi-user resource unit provided by the access point within a transmission opportunity granted to the access point.

5. The communication apparatus according to claim 1, wherein the MU-EDCA parameter value includes a degraded AIFSN compared to an AIFSN of the EDCA parameter value, AIFSN is arbitration inter-frame space number.

6. An access point in a communication network, the communication network comprising communication devices, the communication devices having a transmitter for transmitting data classified in any of a plurality of different ACs in an EDCA TXOP using a contention parameter, wherein, each of the different ACs has its own contention parameter, EDCA is enhanced distributed channel access, TXOP is transmission opportunity, and AC is access category, the access point comprising: a transmitter to transmit, to the communication apparatus, an EDCA parameter value, a MU-EDCA parameter value different from the EDCA parameter value, and a value indicating a predetermined duration, wherein, in a case where data classified in a particular AC of the plurality of different ACs is successfully transmitted from the communication apparatus in a multi-user resource unit provided by the access point within a transmission opportunity granted to the access point, a particular timer of the communication apparatus associated with the particular AC is updated to maintain the contention parameter set to the MU-EDCA parameter value for a predetermined duration indicated by the value before being set back to an EDCA parameter value different from the MU-EDCA parameter value for the particular AC.

7. The access point of claim 6, wherein, the transmitter periodically transmits a beacon frame, wherein the transmitted beacon frame includes the EDCA parameter value, the MU-EDCA parameter value, and a value to set the timer associated with the particular AC to the predetermined duration.

8. The access point of claim 6, wherein, The MU-EDCA parameter value comprises a degraded AIFSN compared to an AIFSN of the EDCA parameter value, AIFSN being an arbitration inter-frame space number.

9. A communication method in a communication apparatus, the communication apparatus having a transmitter for transmitting data classified in any of a plurality of different ACs in an EDCA TXOP using a contention parameter, and the communication apparatus having a plurality of timers, each timer being associated with one of the ACs, wherein, Each of the different ACs has its own contention parameter, EDCA being enhanced distributed channel access, TXOP being transmission opportunity, AC being access category, the communication method comprising: In case data classified in a particular AC of the plurality of different ACs is successfully transmitted from the communication device in a multi-user resource unit provided by the access point within a transmission opportunity granted to the access point, a particular timer of the plurality of timers associated with the particular AC is updated to keep the contention parameter maintained set to the MU-EDCA parameter value for a predetermined duration before being set back to an EDCA parameter value different from the MU-EDCA parameter value for the particular AC.

10. A method of communication in an access point of a communication network, the communication network comprising communication devices having a transmitter for transmitting data classified in any of a plurality of different ACs in an EDCA TXOP using a contention parameter, wherein, Each of the different ACs has its own contention parameter, EDCA being enhanced distributed channel access, TXOP being transmission opportunity, AC being access category, the communication method comprising: transmitting to the communication device an EDCA parameter value, a MU-EDCA parameter value different from the EDCA parameter value, and a value indicating a predetermined duration, In case data classified in a particular AC of the plurality of different ACs is successfully transmitted from the communication device in a multi-user resource unit provided by the access point within a transmission opportunity granted to the access point, a particular timer of the communication device associated with the particular AC is updated to keep the contention parameter maintained set to the MU-EDCA parameter value for a predetermined duration indicated by the value before being set back to an EDCA parameter value different from the MU-EDCA parameter value for the particular AC.

11. A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the communication method of claim 9 or 10.

12. A computer program product comprising a computer program which, when executed by a processor, implements the steps of the communication method of claim 9 or 10.

Images