《信息网络协议基础》课程教学资源（学习资料）IEEE 802.11ax：High-Efficiency WLANs

ACCEPTED FROM OPEN CALL IEEE 802.11AX:HIGH-EFFICIENCY WLANS BORIS BELLALTA ABSTRACT um access control (MAC)layer enhancements to further improve the WLAN performance, IEEE 802.11ax-2019 will replace both IEEE with a focus on the throughput and battery 802.11n-2009 and IEEE 802.11ac-2013 as the duration.This article overviews some of those next high-throughput WLAN amendment.In this new enhancements,and describes the potential article,we review the expected future WLAN benefits and drawbacks of each one.We have scenarios and use cases that justify the push for a grouped these enhancements into four main cat- new PHY/MAC IEEE 802.11 amendment.After egories:spatial reuse,temporal efficiency,spec- that,we overview a set of new technical features trum sharing,and multiple-antenna technologies. that may be included in the IEEE 802.11ax-2019 Moreover,we also discuss several key system-lev- amendment.and describe both their advantages el improvements for next-generation WLANs,as and drawbacks.Finally,we discuss some of the in addition to the IEEE 802.11ax-2019 amend- network-level functionalities that are required to ment,they will likely implement other in-prog- fully improve the user experience in next-gener- ress amendments such as IEEE 802.11aq-2016 ation WLANs and note their relation with other (pre-association discovery of services),IEEE ongoing IEEE 802.11 amendments. 802.11ak-2017 (bridged networks),and IEEE 802.11ai-2016(fast initial link setup time)to sat- INTRODUCTION isfy the created expectations. IEEE 802.11 wireless local area networks (WLANs)[1]are a cost-efficient solution for SCENARIOS,USE CASES,AND wireless Internet access that can satisfy most cur- rent communication requirements in domestic, REQUIREMENTS public,and business scenarios. The forecast number of devices and networks. Similar to other wireless technologies.WLANs and traffic characteristics and user demands for have evolved by integrating the latest technolog- the 2020-2030 decade motivate the development ical advances in the field as soon as they have of a new PHY/MAC IEEE 802.11 amendment to become sufficiently mature,aiming to continuous- cope with the new challenges and usages WLANs ly improve spectrum utilization and raw WLAN will face [2]. performance.IEEE 802.11n-2009 adopted sin- One of the most representative characteris- gle-user multiple-input multiple-output (SU-MI- tics of WLANs is the use of carrier sense multiple MO),channel bonding,and packet aggregation. access with collision avoidance (CSMA/CA)as Those mechanisms were further extended in IEEE the MAC protocol.It offers a reasonable trade- 802.11ac-2013,which also introduced downlink off between performance,robustness,and imple- multi-user (MU)MIMO transmissions.In addi- mentation costs.Using CSMA/CA,when a node tion,new amendments such as IEEE 802.11af- has a packet ready for transmission,it listens to 2013 and IEEE 802.11ah-2016 are further the channel.Once the channel has been detect- expanding the application scenarios of WLANs, ed as free (i.e.,the energy level on the channel which include cognitive radio,long-range commu- is lower than the clear channel assessment.CCA nication,advanced power saving mechanisms,and threshold),the node starts the backoff procedure support for machine-to-machine (M2M)devices. by selecting a random initial value for the backoff Partly because of their own success,next-gen- counter.The node then starts decreasing the back- eration WLANs face two main challenges.First. off counter while sensing the channel.Whenever a they must address dense scenarios,which is transmission,either from other nodes within the motivated by the continuous deployment of new same WLAN or belonging to other WLANs,is access points(APs)to cover new areas and pro- detected on the channel,the backoff counter is vide higher transmission rates.Second,the cur- paused until the channel is detected to be free rent evolution of Internet usage toward real-time again,at which point the countdown is resumed. high-definition audio and video content will also When the backoff counter reaches zero,the node significantly increase users'throughput needs in starts transmitting.Figure la shows an example of the coming years. CSMA/CA operation. To address those challenges,the High-Effi- ciency WLAN(HEW)Task Group [2]is current- DENSE WLAN SCENARIOS The author is with Uni- ly working on a new high-throughput amendment Providing high data rates in scenarios where versitat Pompeu Fabra, named IEEE 802.11ax-2019.This new amend- the density of WLAN users is very high (e.g.,1 Barcelona ment will develop new physical(PHY)and medi- user/m2)requires the deployment of many APs 38 1536-1284/16/$25.00©2016IEEE IEEE Wireless Communications.February 2016

38 1536-1284/16/$25.00 © 2016 IEEE IEEE Wireless Communications • February 2016 The author is with Uni￾versitat Pompeu Fabra, Barcelona Abstract IEEE 802.11ax-2019 will replace both IEEE 802.11n-2009 and IEEE 802.11ac-2013 as the next high-throughput WLAN amendment. In this article, we review the expected future WLAN scenarios and use cases that justify the push for a new PHY/MAC IEEE 802.11 amendment. After that, we overview a set of new technical features that may be included in the IEEE 802.11ax-2019 amendment, and describe both their advantages and drawbacks. Finally, we discuss some of the network-level functionalities that are required to fully improve the user experience in next-gener￾ation WLANs and note their relation with other ongoing IEEE 802.11 amendments. Introduction IEEE 802.11 wireless local area networks (WLANs) [1] are a cost-efficient solution for wireless Internet access that can satisfy most cur￾rent communication requirements in domestic, public, and business scenarios. Similar to other wireless technologies, WLANs have evolved by integrating the latest technolog￾ical advances in the field as soon as they have become sufficiently mature, aiming to continuous￾ly improve spectrum utilization and raw WLAN performance. IEEE 802.11n-2009 adopted sin￾gle-user multiple-input multiple-output (SU-MI￾MO), channel bonding, and packet aggregation. Those mechanisms were further extended in IEEE 802.11ac-2013, which also introduced downlink multi-user (MU) MIMO transmissions. In addi￾tion, new amendments such as IEEE 802.11af- 2013 and IEEE 802.11ah-2016 are further expanding the application scenarios of WLANs, which include cognitive radio, long-range commu￾nication, advanced power saving mechanisms, and support for machine-to-machine (M2M) devices. Partly because of their own success, next-gen￾eration WLANs face two main challenges. First, they must address dense scenarios, which is motivated by the continuous deployment of new access points (APs) to cover new areas and pro￾vide higher transmission rates. Second, the cur￾rent evolution of Internet usage toward real-time high-definition audio and video content will also significantly increase users’ throughput needs in the coming years. To address those challenges, the High-Effi￾ciency WLAN (HEW) Task Group [2] is current￾ly working on a new high-throughput amendment named IEEE 802.11ax-2019. This new amend￾ment will develop new physical (PHY) and medi￾um access control (MAC) layer enhancements to further improve the WLAN performance, with a focus on the throughput and battery duration. This article overviews some of those new enhancements, and describes the potential benefits and drawbacks of each one. We have grouped these enhancements into four main cat￾egories: spatial reuse, temporal efficiency, spec￾trum sharing, and multiple-antenna technologies. Moreover, we also discuss several key system-lev￾el improvements for next-generation WLANs, as in addition to the IEEE 802.11ax-2019 amend￾ment, they will likely implement other in-prog￾ress amendments such as IEEE 802.11aq-2016 (pre-association discovery of services), IEEE 802.11ak-2017 (bridged networks), and IEEE 802.11ai-2016 (fast initial link setup time) to sat￾isfy the created expectations. Scenarios, Use Cases, and Requirements The forecast number of devices and networks, and traffic characteristics and user demands for the 2020–2030 decade motivate the development of a new PHY/MAC IEEE 802.11 amendment to cope with the new challenges and usages WLANs will face [2]. One of the most representative characteris￾tics of WLANs is the use of carrier sense multiple access with collision avoidance (CSMA/CA) as the MAC protocol. It offers a reasonable trade￾off between performance, robustness, and imple￾mentation costs. Using CSMA/CA, when a node has a packet ready for transmission, it listens to the channel. Once the channel has been detect￾ed as free (i.e., the energy level on the channel is lower than the clear channel assessment, CCA, threshold), the node starts the backoff procedure by selecting a random initial value for the backoff counter. The node then starts decreasing the back￾off counter while sensing the channel. Whenever a transmission, either from other nodes within the same WLAN or belonging to other WLANs, is detected on the channel, the backoff counter is paused until the channel is detected to be free again, at which point the countdown is resumed. When the backoff counter reaches zero, the node starts transmitting. Figure 1a shows an example of CSMA/CA operation. Dense WLAN Scenarios Providing high data rates in scenarios where the density of WLAN users is very high (e.g., 1 user/m2) requires the deployment of many APs Boris Bellalta IEEE 802.11ax: High-Efficiency WLANs Ac cep te d f r o m Ope n C a ll

Successful transmission Data ACK STA Backoff countdown Busy channel Collision between two transmissions (a) Scenario (area,m2) APs STAs Description Stadium(-12,500 m) >1000 >50,000 Large events that require many APs to provide a satisfactory connectivity service able to support video uploading/downloading. Train(-600 m2) 1000 Full coverage inside a train to provide both work and entertainment services. Apartment building- Several short-range APs deployed in each apartment,offering full coverage and high 4 floors,6 apartments/ >120 <480 data rates for bandwidth hungry entertainment applications,as well as connectivity floor (-2400 m for house appliances.Community APs may also be deployed in the corridors and 2 shared spaces. WLAN coverage area 8B品B部88B9品B部BBB9888 ● B$0品品盟出出B品盟 Short-range WLAN (b) (d) Figure 1.Key scenarios in next-generation WLANs:a)example of CSMA/CA temporal evolution with one AP and two STAs;b) stadium;c)train;d)floor of a building with several apartments. placed close to each other (e.g.,within 5-10 m concentrated in small areas because of a fair,a of one another).Figure 1 depicts and describes conference,or a sporting event.The presence of three of those scenarios:a stadium,a train,and many people results in a high density of STAs an apartment building.In these dense scenarios, and the necessity of deploying many APs to offer most relevant challenges are related to interfer- satisfactory service.A fundamental challenge in ence issues,which increase the packet error rate these scenarios is to deploy,optimize,and coor- and reduce the number of concurrent transmis- dinate such a large number of APs and STAs. sions in a given area by preventing neighboring Public transport is also a key scenario for WLANs from accessing the channel.Addition- next-generation WLANs because trains,buses, ally,the presence of many stations (STAs)in the and planes will offer broadband Internet access same area increases the chances that the backoff In these scenarios,the user density may be nota counters of two or more STAs reach zero simul- bly high,with several people per square meter. taneously,which results in a collision. Then smart AP coordination can help improve In the stadium scenario,many people are spatial reuse,and the use of an efficient medium IEEE Wireless Communications.February 2016 39

IEEE Wireless Communications • February 2016 39 placed close to each other (e.g., within 5–10 m of one another). Figure 1 depicts and describes three of those scenarios: a stadium, a train, and an apartment building. In these dense scenarios, most relevant challenges are related to interfer￾ence issues, which increase the packet error rate and reduce the number of concurrent transmis￾sions in a given area by preventing neighboring WLANs from accessing the channel. Addition￾ally, the presence of many stations (STAs) in the same area increases the chances that the backoff counters of two or more STAs reach zero simul￾taneously, which results in a collision. In the stadium scenario, many people are concentrated in small areas because of a fair, a conference, or a sporting event. The presence of many people results in a high density of STAs and the necessity of deploying many APs to offer satisfactory service. A fundamental challenge in these scenarios is to deploy, optimize, and coor￾dinate such a large number of APs and STAs. Public transport is also a key scenario for next-generation WLANs because trains, buses, and planes will offer broadband Internet access. In these scenarios, the user density may be nota￾bly high, with several people per square meter. Then smart AP coordination can help improve spatial reuse, and the use of an efficient medium Figure 1. Key scenarios in next-generation WLANs: a) example of CSMA/CA temporal evolution with one AP and two STAs; b) stadium; c) train; d) floor of a building with several apartments. t AP STA STA Data Backoff countdown Busy channel ACK Successful transmission Collision between two transmissions (a) (b) (c) (d) Short-range WLAN WLAN coverage area Scenario (area, m2) APs STAs Description Stadium (~12,500 m2) > 1000 > 50,000 Large events that require many APs to provide a satisfactory connectivity service able to support video uploading/downloading. Train (~600 m2) 1000 Full coverage inside a train to provide both work and entertainment services. Apartment building — 4 floors, 6 apartments/ floor (~2400 m 2) > 120 < 480 Several short-range APs deployed in each apartment, offering full coverage and high data rates for bandwidth hungry entertainment applications, as well as connectivity for house appliances. Community APs may also be deployed in the corridors and shared spaces

Interactive and access protocol may help support many simulta- Backward Compatibility:Because WLANs high-definition video neous contenders. implementing IEEE 802.11ax-2019 must also Finally,in an apartment building,we can support devices using any previous IEEE 802.11 applications are find multiple autonomous and heterogeneous PHY/MAC amendments,mechanisms must also predicted to dominate WLANs overlapping,including short-range be implemented to make it backward compatible WLANs that offer high transmission rates in (i.e.,common frame headers and transmission future Internet usage. small spaces [3].In this scenario,each WLAN rates),although it is a clear source of inefficien- Two examples of is primarily configured independent of the oth- cy. applications that require ers,where the channel selection,channel width. and transmission power are randomly set or are NEW FEATURES AND CONCEPTS throughputs of simply the preset values.Therefore,autonomous The IEEE 802.11ax-2019 amendment may several gigabits per sec WLANs must be able to implement smart decen- include some new technical features compared to the IEEE 802.11ac-2013 amendment.We ond are high-definition tralized self-configuration and self-adaptation mechanisms to minimize the interference among introduce them in this section,providing insight multi-party video them. into their potential performance gains and lim- WLANs must also coexist with other wireless conferences in business itations.All numerical results presented in this networks that operate in the industrial,scientific. section are obtained using the analytical model environments,and and medical (ISM)band,such as wireless sensor and parameters from [6],unless otherwise stated. virtual reality networks and personal area networks.In addi- tion,Long Term Evolution(LTE)operators are SPATIAL REUSE enterfainment currently considering deploying LTE networks in In dense scenarios,the combined use of CSMA/ applications af home, the ISM band [4],which is known as LTE-Un- CA,a conservative CCA,and a high transmit which indude culture, licensed,thus opening further coexistence chal- power level may result in scenarios with limit- lenges for WLANs. ed spatial reuse.A conservative configuration films,and games. of both the CCA and transmit power levels FUTURE WLAN USAGES minimizes the interference among the WLANs Interactive and high-definition video applica- which supports higher transmission rates.How- tions are predicted to dominate future Internet ever,the number of concurrent transmissions is usage.Two examples of applications that require reduced,which may decrease the achievable area throughputs of several gigabits per second are throughput.The alternatives that can be used to high-definition multi-party video conferences reach an optimal trade-off between individual in business environments,which can help avoid transmission rates and the number of concurrent unnecessary travel and meetings,and virtu transmissions that maximize the area throughput al reality entertainment applications at home, include dynamically adapting the transmit power which include culture,films,and games.Addi- level,the CCA level,and the use of direction- tionally,web surfing is moving further toward a al transmissions based on the observed network multimedia experience,where rich text,images, performance. audio,and video content interact.Furthermore, Figure 2a shows three neighboring WLANs. file storage,management,and synchronization The channels that each WLAN uses are shown in the cloud are becoming the standard in terms in Fig.2b.Because WLANs A and C.and B and of content management and generation.Those C partially share their channels,they overlap. applications are bandwidth-demanding,and The three APs are inside the carrier sense range require both reliability and limited delay. of the others,as shown in Fig.2a,which pauses their backoff if either of the other two transmits. REQUIREMENTS Although WLAN C uses the widest channel,it Based on the aforementioned scenarios and achieves the lowest throughput because it over- expected use cases,there are four key require- laps with WLANs A and B,which are indepen- ments for the IEEE 802.11ax-2019 amendment. dent of each other(Fig.2c). Coexistence:WLANs operate as unlicensed Dynamic Adaptation of the Transmit Power devices in the ISM bands.Therefore,the IEEE and CCA Levels:Reducing the used transmission 802.11ax-2019 amendment has to include the power in a WLAN reduces its influence area required mechanisms to coexist with both the which benefits the spatial reuse.However,it may other wireless networks that operate there and result in a larger number of packet errors and the licensed devices. lower transmission rates,as well as an increased Higher Throughput:Improving both the sys- number of hidden nodes. tem and user throughput requires the improved Alternatively,to reduce the area of influ- use of channel resources.IEEE 802.11ax-2019 ence of neighboring WLANs and increase each aims for a four-fold throughput increase com- WLAN's chances to transmit,the nodes in a pared to IEEE 802.11ac-2013.To achieve this WLAN may increase their CCA level,hence goal,some new wireless technologies such as requiring a higher energy level in the channel dynamic CCA,OFDMA (orthogonal frequen- to consider it as occupied and pause the backoff cy-division multiple access),and advanced multi- countdown.In [7],significant throughput gains ple-antenna techniques may be used. are achieved by tuning the CCA level in a multi- Energy Efficiency:The target in IEEE cell WLAN scenario.The downside of increasing 802.11ax-2019 is,at least,to not consume more the CCA level is again the higher interference energy than the previous amendments,consid- that a node may suffer,which could be detrimen ering the aforementioned four-fold throughput tal in some cases. increase,which requires both new low-power Beamforming:Omnidirectional transmissions hardware architectures [5]and new low-power homogeneously spread the transmitted energy in PHY/MAC functionalities. all directions,which fills the channel with energy IEEE Wireless Communications.February 2016

40 IEEE Wireless Communications • February 2016 access protocol may help support many simulta￾neous contenders. Finally, in an apartment building, we can find multiple autonomous and heterogeneous WLANs overlapping, including short-range WLANs that offer high transmission rates in small spaces [3]. In this scenario, each WLAN is primarily configured independent of the oth￾ers, where the channel selection, channel width, and transmission power are randomly set or are simply the preset values. Therefore, autonomous WLANs must be able to implement smart decen￾tralized self-configuration and self-adaptation mechanisms to minimize the interference among them. WLANs must also coexist with other wireless networks that operate in the industrial, scientific, and medical (ISM) band, such as wireless sensor networks and personal area networks. In addi￾tion, Long Term Evolution (LTE) operators are currently considering deploying LTE networks in the ISM band [4], which is known as LTE-Un￾licensed, thus opening further coexistence chal￾lenges for WLANs. Future WLAN Usages Interactive and high-definition video applica￾tions are predicted to dominate future Internet usage. Two examples of applications that require throughputs of several gigabits per second are high-definition multi-party video conferences in business environments, which can help avoid unnecessary travel and meetings, and virtu￾al reality entertainment applications at home, which include culture, films, and games. Addi￾tionally, web surfing is moving further toward a multimedia experience, where rich text, images, audio, and video content interact. Furthermore, file storage, management, and synchronization in the cloud are becoming the standard in terms of content management and generation. Those applications are bandwidth-demanding, and require both reliability and limited delay. Requirements Based on the aforementioned scenarios and expected use cases, there are four key require￾ments for the IEEE 802.11ax-2019 amendment. Coexistence: WLANs operate as unlicensed devices in the ISM bands. Therefore, the IEEE 802.11ax-2019 amendment has to include the required mechanisms to coexist with both the other wireless networks that operate there and the licensed devices. Higher Throughput: Improving both the sys￾tem and user throughput requires the improved use of channel resources. IEEE 802.11ax-2019 aims for a four-fold throughput increase com￾pared to IEEE 802.11ac-2013. To achieve this goal, some new wireless technologies such as dynamic CCA, OFDMA (orthogonal frequen￾cy-division multiple access), and advanced multi￾ple-antenna techniques may be used. Energy Efficiency: The target in IEEE 802.11ax-2019 is, at least, to not consume more energy than the previous amendments, consid￾ering the aforementioned four-fold throughput increase, which requires both new low-power hardware architectures [5] and new low-power PHY/MAC functionalities. Backward Compatibility: Because WLANs implementing IEEE 802.11ax-2019 must also support devices using any previous IEEE 802.11 PHY/MAC amendments, mechanisms must also be implemented to make it backward compatible (i.e., common frame headers and transmission rates), although it is a clear source of inefficien￾cy. New Features and Concepts The IEEE 802.11ax-2019 amendment may include some new technical features compared to the IEEE 802.11ac-2013 amendment. We introduce them in this section, providing insight into their potential performance gains and lim￾itations. All numerical results presented in this section are obtained using the analytical model and parameters from [6], unless otherwise stated. Spatial Reuse In dense scenarios, the combined use of CSMA/ CA, a conservative CCA, and a high transmit power level may result in scenarios with limit￾ed spatial reuse. A conservative configuration of both the CCA and transmit power levels minimizes the interference among the WLANs, which supports higher transmission rates. How￾ever, the number of concurrent transmissions is reduced, which may decrease the achievable area throughput. The alternatives that can be used to reach an optimal trade-off between individual transmission rates and the number of concurrent transmissions that maximize the area throughput include dynamically adapting the transmit power level, the CCA level, and the use of direction￾al transmissions based on the observed network performance. Figure 2a shows three neighboring WLANs. The channels that each WLAN uses are shown in Fig. 2b. Because WLANs A and C, and B and C partially share their channels, they overlap. The three APs are inside the carrier sense range of the others, as shown in Fig. 2a, which pauses their backoff if either of the other two transmits. Although WLAN C uses the widest channel, it achieves the lowest throughput because it over￾laps with WLANs A and B, which are indepen￾dent of each other (Fig. 2c). Dynamic Adaptation of the Transmit Power and CCA Levels: Reducing the used transmission power in a WLAN reduces its influence area, which benefits the spatial reuse. However, it may result in a larger number of packet errors and lower transmission rates, as well as an increased number of hidden nodes. Alternatively, to reduce the area of influ￾ence of neighboring WLANs and increase each WLAN’s chances to transmit, the nodes in a WLAN may increase their CCA level, hence requiring a higher energy level in the channel to consider it as occupied and pause the backoff countdown. In [7], significant throughput gains are achieved by tuning the CCA level in a multi￾cell WLAN scenario. The downside of increasing the CCA level is again the higher interference that a node may suffer, which could be detrimen￾tal in some cases. Beamforming: Omnidirectional transmissions homogeneously spread the transmitted energy in all directions, which fills the channel with energy Interactive and high-definition video applications are predicted to dominate future Internet usage. Two examples of applications that require throughputs of several gigabits per sec￾ond are high-definition multi-party video conferences in business environments, and virtual reality entertainment applications at home, which include culture, films, and games

in areas where it is not required.Concentrat- ing the energy toward the desired destination improves the spatial reuse because the devices that are placed in other directions will observe the channel as being empty and,therefore start their own transmissions concurrently.However, similar to the previous case,those nodes outside the energy beam may also become hidden nodes. TEMPORAL EFFICIENCY WLAN WLAN C The backoff countdown,packet headers,inter- Data range frame spaces,collisions,and retransmissions are an intrinsic part of the CSMA/CA channel Carrier sense range access scheme,but they significantly decrease the effective time that a node spends transmitting data every time it accesses the channel.IEEE 802.11ax-2019 may include several solutions to mitigate such overheads. Control Packets:The time consumed by the (a) exchange of control packets may result in large overheads,particularly because they are usually transmitted at a low rate.Common control pack- WLAN A et exchanges between the AP and STAs include the request to send/clear to send (RTS/CTS) exchange to avoid hidden nodes and ACKs to WLAN B acknowledge the reception of data packets. Additionally,some of the new technical fea- tures described in the next sections that enable WLAN C multi-user transmissions may require frequent exchange of control packets to synchronize all -Channels involved STAs,hence also increasing the control 3640444852566064 packets'overheads. (b) Packet Headers,Aggregation,and Pig- 500 gy-Backing:Packet aggregation was introduced in IEEE 802.11n-2009 to reduce temporal over- 450 heads by combining short packets into a longer one.Using packet aggregation,multiple packets 400 can be transmitted with a single backoff,distrib- 350 uted inter-frame space (DIFS).short inter-frame space (SIFS),PHY header,and ACK. 300 The packet header overheads can be reduced 250 by supporting variable size headers and using only the minimum required fields for every pack- 200 et.Additionally,the use of shorter identifiers instead of the full MAC address is considered. 150 Moreover,the piggybacking of ACKs with 100 DATA will improve the efficiency,although some changes in the current setting of the network 50 allocation vector (NAV)are required because the full transmission duration is unknown to the WLAN A WLAN B WLAN C transmission initiator (c) Efficient Retransmissions:Packet errors are also a source of overhead because they current- Figure 2.Throughput unfairness between overlapping WLANs:a)three ly require full retransmission of the data pack- overlapping WLANs;b)channel used by each WLAN;c)throughput et.Further work about the use of incremental achieved by each WLAN. redundancy-based automatic repeat requests (ARQs)can reduce the time spent in retransmis- sions,although it implies some extra complexity number of active STAs is low do they have bidi- in both transmitter and receiver firmware. rectional and saturated traffic flows,and use a Simultaneous Transmit and Receive:By small backoff contention window;here,the use allowing the AP and a STA to simultaneously of STR can result in significant gains.Therefore. transmit and receive (STR),which is commonly specific channel access mechanisms should be known as full-duplex communication,the chan- considered in the IEEE 802.11ax-2019 amend- nel capacity can be theoretically doubled [8]. ment in case the STR capability is finally consid- Using CSMA/CA,the only way they can have ered.Alternatively,the STR capability can allow full-duplex communication is if they finish their WLANs to replace the CA feature of CSMA/CA backoff countdown simultaneously.Otherwise, with collision detection (CD),because collisions one will start transmitting before the other, can be promptly detected and resolved. causing the latter to pause its backoff until the Collision-Free MAC Protocols:Collisions rep- former finishes its transmission.Only when the resent an important waste of channel resources IEEE Wireless Communications.February 2016

IEEE Wireless Communications • February 2016 41 in areas where it is not required. Concentrat￾ing the energy toward the desired destination improves the spatial reuse because the devices that are placed in other directions will observe the channel as being empty and, therefore start their own transmissions concurrently. However, similar to the previous case, those nodes outside the energy beam may also become hidden nodes. Temporal Efficiency The backoff countdown, packet headers, inter￾frame spaces, collisions, and retransmissions are an intrinsic part of the CSMA/CA channel access scheme, but they significantly decrease the effective time that a node spends transmitting data every time it accesses the channel. IEEE 802.11ax-2019 may include several solutions to mitigate such overheads. Control Packets: The time consumed by the exchange of control packets may result in large overheads, particularly because they are usually transmitted at a low rate. Common control pack￾et exchanges between the AP and STAs include the request to send/clear to send (RTS/CTS) exchange to avoid hidden nodes and ACKs to acknowledge the reception of data packets. Additionally, some of the new technical fea￾tures described in the next sections that enable multi-user transmissions may require frequent exchange of control packets to synchronize all involved STAs, hence also increasing the control packets’ overheads. Packet Headers, Aggregation, and Pig￾gy-Backing: Packet aggregation was introduced in IEEE 802.11n-2009 to reduce temporal over￾heads by combining short packets into a longer one. Using packet aggregation, multiple packets can be transmitted with a single backoff, distrib￾uted inter-frame space (DIFS), short inter-frame space (SIFS), PHY header, and ACK. The packet header overheads can be reduced by supporting variable size headers and using only the minimum required fields for every pack￾et. Additionally, the use of shorter identifiers instead of the full MAC address is considered. Moreover, the piggybacking of ACKs with DATA will improve the efficiency, although some changes in the current setting of the network allocation vector (NAV) are required because the full transmission duration is unknown to the transmission initiator. Efficient Retransmissions: Packet errors are also a source of overhead because they current￾ly require full retransmission of the data pack￾et. Further work about the use of incremental redundancy-based automatic repeat requests (ARQs) can reduce the time spent in retransmis￾sions, although it implies some extra complexity in both transmitter and receiver firmware. Simultaneous Transmit and Receive: By allowing the AP and a STA to simultaneously transmit and receive (STR), which is commonly known as full-duplex communication, the chan￾nel capacity can be theoretically doubled [8]. Using CSMA/CA, the only way they can have full-duplex communication is if they finish their backoff countdown simultaneously. Otherwise, one will start transmitting before the other, causing the latter to pause its backoff until the former finishes its transmission. Only when the number of active STAs is low do they have bidi￾rectional and saturated traffic flows, and use a small backoff contention window; here, the use of STR can result in significant gains. Therefore, specific channel access mechanisms should be considered in the IEEE 802.11ax-2019 amend￾ment in case the STR capability is finally consid￾ered. Alternatively, the STR capability can allow WLANs to replace the CA feature of CSMA/CA with collision detection (CD), because collisions can be promptly detected and resolved. Collision-Free MAC Protocols: Collisions rep￾resent an important waste of channel resources Figure 2. Throughput unfairness between overlapping WLANs: a) three overlapping WLANs; b) channel used by each WLAN; c) throughput achieved by each WLAN. WLAN A WLAN B WLAN C Data range Carrier sense range WLAN A WLAN B WLAN C 36 40 44 48 52 56 60 64 Channels WLAN A WLAN B WLAN C 50 100 150 200 250 300 350 400 450 500 AP throughput (Mb/s) (a) (b) (c)

An unplanned deployment of WLANs Deterministic backoff results in chaotic and fragmented spectrum occupancy,which causes DATA many inefficiencies and undesirable interactions among neighboring STAA:· WLANs.To improve 。 the spectrum usage 1 STA B efficiency,two main approaches can be ACK considered in IEEE STA C: Time 802.110x2019: Simultaneous transmission using the STR capability dynamic channel Figure 3.CSMA/ECA operation.It can be observed how the use of a deterministic backoff allows pre- bonding and OFDMA diction of when a node will transmit again after successful transmission. in WLANs.IEEE 802.11ax-2019 may consider resources by dividing the channel width into enhancing or changing the underlying CSMA/ multiple narrow channels.Then these narrow CA protocol to minimize collisions.There are channels can be used to transmit to multiple two possibilities:moving to a centralized solu- users in parallel [11].A basic implementation tion or enhancing the current CSMA/CA proto- of OFDMA in WLANs may simply consider the col.Because centralized options such as hybrid use of multiple independent 20 MHz channels. coordination function controlled channel access This approach is shown in Fig.4:when channel (HCCA)were never adopted in WLANs,a focus bonding is used,each 20 MHz subchannel can be on enhancing CSMA/CA appears more plausi- independently allocated to a different user.The ble.CSMA with enhanced CA(CSMA/ECA)is RTS packet has been extended to announce the a particularly good candidate to replace CSMA/ subchannels allocation to the STAs.Additionally. CA because it is backward compatible,is easily OFDMA may enable the use of non-contiguous implemented,and outperforms CSMA/CA in all channel bonding and remove the requirement to cases [9]. use only 20 MHz consecutive channels. Figure 3 shows the basic operation of CSMA/ Figure 4a shows an example where dynamic ECA with STR.Compared to CSMA/CA,the channel bonding and OFDMA operate together main difference observed for CSMA/ECA is its The upper part of Fig.4a shows a snapshot of use of a deterministic backoff after successful the spectrum occupancy for a group of neigh- transmissions.This deterministic backoff guaran- boring WLANs.The lower part of Fig.4a shows tees that after some time,a collision-free sched- two transmissions:a node in the target WLAN ule can be achieved.In addition.because the AP transmits to a single user via a bonded channel can learn when the STAs will transmit,the use of of 40 MHz (left),and a node uses a bonded STR can provide huge performance gains. channel of 80 MHz and OFDMA to transmit to three different users(right).Figure 4b shows the SPECTRUM SHARING AP throughput when OFDMA is used to split An unplanned deployment of WLANs results in a 160 MHz channel into multiple subchannels a chaotic and fragmented spectrum occupancy, The parallelization of temporal overheads clearly which causes many inefficiencies and undesirable improves the throughput. interactions among neighboring WLANs [6].To improve the spectrum usage efficiency,two main MULTIPLE ANTENNAS approaches can be considered in IEEE 802.1lax- Spatial multiplexing using multiple antennas at 2019:dynamic channel bonding and OFDMA. both APs and STAs remains one of the key tech- Dynamic Channel Bonding:To adapt to nologies to achieve high throughput in WLANs. the instantaneous channel occupancy,IEEE IEEE 802.11ax-2019 will continue implementing 802.11ax-2019 may consider extending the both SU-MIMO and downlink MU-MIMO.as dynamic bandwidth channel access (DBCA) in IEEE 802.11ac-2013.However,it may also scheme introduced in the IEEE 802.11ac-2013 include or provide support for uplink MU-MI- amendment [10].Using DBCA,only the avail- MO,massive MIMO,and network MIMO,as able channel width is used at each transmission, well as distributed antenna solutions. which allows WLANs to adapt to the instan- Multi-User MIMO:Multi-user MIMO enables taneous spectrum occupancy.This mechanism multiple simultaneous transmissions to different helps fill most spectrum gaps and share them STAs from the AP in the downlink,and from fairly among neighboring WLANs. multiple STAs to the AP in the uplink.A sur- OFDMA:The use of orthogonal frequen- vey of MU-MIMO MAC protocols for WLANs cy-division multiple access (OFDMA)adds a is presented in [12],where the challenges and new degree of flexibility to the use of spectrum requirements to design an MU-MIMO MAC 42 IEEE Wireless Communications.February 2016

42 IEEE Wireless Communications • February 2016 in WLANs. IEEE 802.11ax-2019 may consider enhancing or changing the underlying CSMA/ CA protocol to minimize collisions. There are two possibilities: moving to a centralized solu￾tion or enhancing the current CSMA/CA proto￾col. Because centralized options such as hybrid coordination function controlled channel access (HCCA) were never adopted in WLANs, a focus on enhancing CSMA/CA appears more plausi￾ble. CSMA with enhanced CA (CSMA/ECA) is a particularly good candidate to replace CSMA/ CA because it is backward compatible, is easily implemented, and outperforms CSMA/CA in all cases [9]. Figure 3 shows the basic operation of CSMA/ ECA with STR. Compared to CSMA/CA, the main difference observed for CSMA/ECA is its use of a deterministic backoff after successful transmissions. This deterministic backoff guaran￾tees that after some time, a collision-free sched￾ule can be achieved. In addition, because the AP can learn when the STAs will transmit, the use of STR can provide huge performance gains. Spectrum Sharing An unplanned deployment of WLANs results in a chaotic and fragmented spectrum occupancy, which causes many inefficiencies and undesirable interactions among neighboring WLANs [6]. To improve the spectrum usage efficiency, two main approaches can be considered in IEEE 802.11ax- 2019: dynamic channel bonding and OFDMA. Dynamic Channel Bonding: To adapt to the instantaneous channel occupancy, IEEE 802.11ax-2019 may consider extending the dynamic bandwidth channel access (DBCA) scheme introduced in the IEEE 802.11ac-2013 amendment [10]. Using DBCA, only the avail￾able channel width is used at each transmission, which allows WLANs to adapt to the instan￾taneous spectrum occupancy. This mechanism helps fill most spectrum gaps and share them fairly among neighboring WLANs. OFDMA: The use of orthogonal frequen￾cy-division multiple access (OFDMA) adds a new degree of flexibility to the use of spectrum resources by dividing the channel width into multiple narrow channels. Then these narrow channels can be used to transmit to multiple users in parallel [11]. A basic implementation of OFDMA in WLANs may simply consider the use of multiple independent 20 MHz channels. This approach is shown in Fig. 4: when channel bonding is used, each 20 MHz subchannel can be independently allocated to a different user. The RTS packet has been extended to announce the subchannels allocation to the STAs. Additionally, OFDMA may enable the use of non-contiguous channel bonding and remove the requirement to use only 20 MHz consecutive channels. Figure 4a shows an example where dynamic channel bonding and OFDMA operate together. The upper part of Fig. 4a shows a snapshot of the spectrum occupancy for a group of neigh￾boring WLANs. The lower part of Fig. 4a shows two transmissions: a node in the target WLAN transmits to a single user via a bonded channel of 40 MHz (left), and a node uses a bonded channel of 80 MHz and OFDMA to transmit to three different users (right). Figure 4b shows the AP throughput when OFDMA is used to split a 160 MHz channel into multiple subchannels. The parallelization of temporal overheads clearly improves the throughput. Multiple Antennas Spatial multiplexing using multiple antennas at both APs and STAs remains one of the key tech￾nologies to achieve high throughput in WLANs. IEEE 802.11ax-2019 will continue implementing both SU-MIMO and downlink MU-MIMO, as in IEEE 802.11ac-2013. However, it may also include or provide support for uplink MU-MI￾MO, massive MIMO, and network MIMO, as well as distributed antenna solutions. Multi-User MIMO: Multi-user MIMO enables multiple simultaneous transmissions to different STAs from the AP in the downlink, and from multiple STAs to the AP in the uplink. A sur￾vey of MU-MIMO MAC protocols for WLANs is presented in [12], where the challenges and requirements to design an MU-MIMO MAC Figure 3. CSMA/ECA operation. It can be observed how the use of a deterministic backoff allows pre￾diction of when a node will transmit again after successful transmission. Time AP STA A STA C STA B Simultaneous transmission using the STR capability ACK DATA Deterministic backoff An unplanned deployment of WLANs results in chaotic and fragmented spectrum occupancy, which causes many inefficiencies and undesirable interactions among neighboring WLANs. To improve the spectrum usage efficiency, two main approaches can be considered in IEEE 802.11ax-2019: dynamic channel bonding and OFDMA

Occupied spectrum Free spectrum 1000 900 800 700 600 500 400 TS DATA 300 ⊙-1Tx-160MHz 200 钟-2TX-80MH2z 100 -4Tx-40 MHz +—8Tx-20MHz 10 0 30 40 50 60 Number of aggregated packets Figure 4.Dynamic spectrum access with and without OFDMA:a)dynamic channel bonding and OFDMA;b)achievable down- Tink throughput using a channel width of 160 MHz and OFDMA.The RTS size is 120 +56.Nx bits,with Nux the number of OFDMA subchannels. protocol are introduced,and several uplink and backoff countdowns at the same time,uplink downlink MAC proposals are reviewed. MU-MIMO transmissions may be started by In the downlink,a challenge for IEEE the AP using a special RTS"packet containing 802.11ax-2019 is to reduce the channel sounding information about the STAs that can transmit overheads given the same explicit approach as in in parallel.Then the selected STAs will simply IEEE 802.11ac-2013 is considered.The overhead start transmitting at the same time and wait to caused by the explicit channel sounding protocol receive the corresponding ACKs.This approach implemented depends on the channel sounding allows all selected STAs to be synchronized,but rate and number of sounded STAs,which can requires knowledge of the STAs's buffer occu- result in an unacceptable overhead in scenari- pancy by the AP,which can be provided by the os with many STAs.Solutions to reduce such a same STAs as in previous transmissions.Figure large overhead,apart from replacing the current 5b shows the throughput achieved by the aP in channel sounding protocol with a more efficient two downlink MU-MIMO configurations(cases solution,will require the use of smart schedul- 16:4:4 and 16:16:1).The obtained throughput ers that consider the current traffic patterns and values are compared with the case where a single the quality of service (QoS)demands of users to spatial stream is transmitted to only one destina- decide when the channel state information (CSI) tion (case 1:1:1). has to be requested,and from which STAs. Massive MIMO:Massive MIMO refers to the To support MU-MIMO transmissions in the case where the aP has many more antennas than uplink,IEEE 802.11ax-2019 must also detail how STAs and uses them to create a nearly identical the uplink CSI from each STA is obtained by number of point-to-point links as the number the AP.introduce a mechanism to signal a group of active STAs [13].In addition to the cost of of STAs to simultaneously start a transmission APs,the extra processing complexity,and high- and include techniques to overcome channel cal- er energy consumption,other open challenges ibration and timing issues in order to efficiently for massive MIMO include obtaining the CSI decode all the simultaneously received packets information;WLANs may require switching to at the AP. an implicit channel feedback approach. In both cases,using the collected CSI,the AP Network MIMO:In coordinated WLAN has to select the specific STAs that will take part deployments,network MIMO can be used to in the next MU-MIMO transmission.Therefore, minimize the interference among simultane- the design of an efficient mechanism to create ous transmissions from different APs.The idea groups of STAs with low spatial channel correla- behind network MIMO is that different APs tion and similar channel quality is still an open can coordinate the transmissions as if they were challenge.Failing to properly create those groups a large array of antennas,which reduces the may prevent next-generation WLANs from fully inter-transmission interference and increases the benefiting from MU-MIMO technology. spatial reuse [14].However,effectively solving Figure 5a shows an AP with three STAs.On the tight synchronization requirements among the left,we have three downlink MU-MIMO the APs remains an open challenge. transmissions.The ones directed to STAs A.B and C contain four,two,and one SU-MIMO WLAN-LEVEL IMPROVEMENTS spatial streams,respectively.To start a downlink The user experience in next-generation WLANs transmission,the AP omnidirectionally sends the will not be simply enhanced by increasing the PHY header with information about the group of achievable network and user throughput as pre selected STAs and the number of spatial streams vious technical features do.To achieve that goal that are transmitted to each STA in SU-MIMO apart from the IEEE 802.11ax-2019 amendment, mode.On the right,we show an uplink MU-MI- there are other IEEE 802.11 amendments in prog- MO transmission.In IEEE 802.11ax-2019, ress.Figure 6 shows the most significant ones, because several STAs are unlikely to finish their including the new features they are targeting. IEEE Wireless Communications.February 2016 43

IEEE Wireless Communications • February 2016 43 protocol are introduced, and several uplink and downlink MAC proposals are reviewed. In the downlink, a challenge for IEEE 802.11ax-2019 is to reduce the channel sounding overheads given the same explicit approach as in IEEE 802.11ac-2013 is considered. The overhead caused by the explicit channel sounding protocol implemented depends on the channel sounding rate and number of sounded STAs, which can result in an unacceptable overhead in scenari￾os with many STAs. Solutions to reduce such a large overhead, apart from replacing the current channel sounding protocol with a more efficient solution, will require the use of smart schedul￾ers that consider the current traffic patterns and the quality of service (QoS) demands of users to decide when the channel state information (CSI) has to be requested, and from which STAs. To support MU-MIMO transmissions in the uplink, IEEE 802.11ax-2019 must also detail how the uplink CSI from each STA is obtained by the AP, introduce a mechanism to signal a group of STAs to simultaneously start a transmission, and include techniques to overcome channel cal￾ibration and timing issues in order to efficiently decode all the simultaneously received packets at the AP. In both cases, using the collected CSI, the AP has to select the specific STAs that will take part in the next MU-MIMO transmission. Therefore, the design of an efficient mechanism to create groups of STAs with low spatial channel correla￾tion and similar channel quality is still an open challenge. Failing to properly create those groups may prevent next-generation WLANs from fully benefiting from MU-MIMO technology. Figure 5a shows an AP with three STAs. On the left, we have three downlink MU-MIMO transmissions. The ones directed to STAs A, B, and C contain four, two, and one SU-MIMO spatial streams, respectively. To start a downlink transmission, the AP omnidirectionally sends the PHY header with information about the group of selected STAs and the number of spatial streams that are transmitted to each STA in SU-MIMO mode. On the right, we show an uplink MU-MI￾MO transmission. In IEEE 802.11ax-2019, because several STAs are unlikely to finish their backoff countdowns at the same time, uplink MU-MIMO transmissions may be started by the AP using a special RTS packet containing information about the STAs that can transmit in parallel. Then the selected STAs will simply start transmitting at the same time and wait to receive the corresponding ACKs. This approach allows all selected STAs to be synchronized, but requires knowledge of the STAs’s buffer occu￾pancy by the AP, which can be provided by the same STAs as in previous transmissions. Figure 5b shows the throughput achieved by the AP in two downlink MU-MIMO configurations (cases 16:4:4 and 16:16:1). The obtained throughput values are compared with the case where a single spatial stream is transmitted to only one destina￾tion (case 1:1:1). Massive MIMO: Massive MIMO refers to the case where the AP has many more antennas than STAs and uses them to create a nearly identical number of point-to-point links as the number of active STAs [13]. In addition to the cost of APs, the extra processing complexity, and high￾er energy consumption, other open challenges for massive MIMO include obtaining the CSI information; WLANs may require switching to an implicit channel feedback approach. Network MIMO: In coordinated WLAN deployments, network MIMO can be used to minimize the interference among simultane￾ous transmissions from different APs. The idea behind network MIMO is that different APs can coordinate the transmissions as if they were a large array of antennas, which reduces the inter-transmission interference and increases the spatial reuse [14]. However, effectively solving the tight synchronization requirements among the APs remains an open challenge. WLAN-Level Improvements The user experience in next-generation WLANs will not be simply enhanced by increasing the achievable network and user throughput as pre￾vious technical features do. To achieve that goal, apart from the IEEE 802.11ax-2019 amendment, there are other IEEE 802.11 amendments in prog￾ress. Figure 6 shows the most significant ones, including the new features they are targeting. Figure 4. Dynamic spectrum access with and without OFDMA: a) dynamic channel bonding and OFDMA; b) achievable down￾link throughput using a channel width of 160 MHz and OFDMA. The RTS size is 120 + 56 · Ntx bits, with Ntx the number of OFDMA subchannels. t f Data ACK RTS CTS DATA ACK Occupied spectrum Free spectrum t t Band f f 0 10 20 30 40 50 60 0 100 200 300 400 500 600 700 800 900 1000 Number of aggregated packets AP throughput (Mb/s) 1Tx – 160 MHz 2Tx – 80 MHz 4Tx – 40 MHz 8Tx – 20 MHz

The user experience in next-generation WLANs MU-MIMO transmission SU-MIMO transmission will not be simply Downlink Uplink enhanced by increasing the achievable network STA A TA A and user throughput as previous technical STA B STA B features do. To achieve that goal, apart from the STA C STA C IEE802.11ax2019 amendment,there are PHY header other IEEE 802.11 Data transmission duration amendments in Data RTS" progress. AP 四四口+ ACK STAA :口 STA B. ▣ STA C O (a) 2400 2400 2200 2200 2000 ZHW 09L 2000 1800 1800 1600 1400 s 1600 1400 1200 1200 1000 800 '([s/qW)indy6noy 1000 800 600 600 400 400 200 200 1:1:116:4:416:16:1 1:1:116:4:416:16:1 (b) Figure 5.Multi-user spatial multiplexing:a)downlink and uplink MU-MIMO transmissions in WLANs;b)downlink throughput.The AP is equipped with 16 antennas.Each configuration is identi- fied by the code x:y:z,where x refers to the total number of transmitted spatial streams,y the number of destinations,and z the number of streams transmitted to each destination. In scenarios where multiple APs are used to network virtualization and software defined net- offer a large coverage area and higher data rates works are also an interesting option as they are by deploying multiple WLANs nearby,a mecha- able to centralize all decisions,including the hand- nism that enables fast handoff among the APs is off between overlapping APs.Future IEEE amend- required because the current large delays when a ments will probably consider such an approach. STA switches to a new AP are unacceptable.The Device-to-device (D2D)communications. IEEE 802.11ai-2016 amendment targets this chal- which are also known as WiFi Direct in the lenge.It aims to provide a handoff duration below IEEE WLAN context,will also be a key element 100 ms by implementing preemptive channel sens- in next-generation WLANs [15].Examples of ing and user authentication.Alternatives based on D2D communications are file synchronization 44 IEEE Wireless Communications.February 2016

44 IEEE Wireless Communications • February 2016 In scenarios where multiple APs are used to offer a large coverage area and higher data rates by deploying multiple WLANs nearby, a mecha￾nism that enables fast handoff among the APs is required because the current large delays when a STA switches to a new AP are unacceptable. The IEEE 802.11ai-2016 amendment targets this chal￾lenge. It aims to provide a handoff duration below 100 ms by implementing preemptive channel sens￾ing and user authentication. Alternatives based on network virtualization and software defined net￾works are also an interesting option as they are able to centralize all decisions, including the hand￾off between overlapping APs. Future IEEE amend￾ments will probably consider such an approach. Device-to-device (D2D) communications, which are also known as WiFi Direct in the IEEE WLAN context, will also be a key element in next-generation WLANs [15]. Examples of D2D communications are file synchronization Figure 5. Multi-user spatial multiplexing: a) downlink and uplink MU-MIMO transmissions in WLANs; b) downlink throughput. The AP is equipped with 16 antennas. Each configuration is identi￾fied by the code x:y:z, where x refers to the total number of transmitted spatial streams, y the number of destinations, and z the number of streams transmitted to each destination. AP STA C STA B STA A STA A STA B STA C STA C STA B STA A AP Downlink Uplink Data PHY header ACK RTS’’ t SU-MIMO transmission MU-MIMO transmission Data transmission duration 1:1:1 16:4:4 16:16:1 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 AP throughput (Mb/s), ch. width = 20 MHz 1:1:1 16:4:4 16:16:1 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 AP throughput (Mb/s]), ch. width = 160 MHz (a) (b) The user experience in next-generation WLANs will not be simply enhanced by increasing the achievable network and user throughput as previous technical features do. To achieve that goal, apart from the IEEE 802.11ax-2019 amendment, there are other IEEE 802.11 amendments in progress

The use of D2D IEEE 802.11aa communications will reduce the amount of IEEE 802.11ak airtime required for Extended IEEE 802.11ai each fransmission by functionalities allowing higher data IEEE 802.11aq rates and avoiding the IEEE 802.11ay use of the AP as a : relay.Beamforming is IEEE 802.11ad an interesting feature for D2D communication ::: IEEE 802.11ax because it may allow PHY/MAC IEEE 802.11ac concurrent transmissions inside the same WLAN IEEE 802.11n by different groups of nodes. IEEE 802.11ah >t 20092010201120122013201420152016201720182019 Feature Amendment Objective Handoff duration below 100 ms by Fast handoff between APs IEEE 802.11ai-2016 implementing preemptive channel sensing and user authentication. STAs with D2D capability can decide to D2D/WiFi Direct WiFi Alliance feature create their own WLAN to communicate directly. Multi-technology APs and 1EEE802.11ak-2017 Interconnection of multiple WLANs. bridging capabilities Video traffic 1EEE8v02.11aa-2012 Stream prioritization and groupcast differentiation mechanisms. APs provide pre-association information AP/service selection lEEE802.11aq-2018 about the services offered to help STAs to choose between multiple APs Figure 6.Upcoming IEEE 802.11 amendments. with an external hard disk and instantaneous file very fast millimeter-wave communications.In exchange between a mobile phone and a pro- this situation,the STAs of any of those networks jector/smart television.The use of D2D com- should be able to communicate with the STAs in munications will reduce the amount of airtime any of the other WLANs.This is the aim of the required for each transmission by allowing higher IEEE 802.11ak-2017 amendment,which focuses data rates and avoiding the use of the aP as a on adding bridging capabilities to WLANs. relay.Beamforming is an interesting feature for Finally,next-generation WLANs will use traf- D2D communication because it may allow con- fic differentiation,flow admission control,and current transmissions inside the same WLAN by groupcast mechanisms from the IEEE 802.11e- different groups of nodes. 2007,IEEE802.11ae-2012 and IEEE802.11aa- IEEE 802.11ax-2019 aims to operate in the 2012 amendments to support multimedia traffic 2.4 and 5 GHz bands by superseding and inte- with the required QoS.Also,further advances in grating the IEEE 802.11n-2009 and IEEE power saving mechanisms are expected to keep 802.11ac-2013 amendments.In a few years,every WLAN energy consumption as low as possible. single AP will most likely implement two IEEE 802.11ax-2019 instances that independently oper- CONCLUSION ate at 2.4 and 5 GHz,with a IEEE 802.11ah-2016 We have reviewed some technological options instance at 1 GHz for M2M and long-range com- that could be included in the IEEE 802.11ax- munications,and one IEEE 802.11ad-2012 (or 2019 amendment for next-generation WLANs. the future IEEE 802.11ay-202x)at 60 GHz for Individually,all those solutions offer some per- IEEE Wireless Communications.February 2016 45

IEEE Wireless Communications • February 2016 45 with an external hard disk and instantaneous file exchange between a mobile phone and a pro￾jector/smart television. The use of D2D com￾munications will reduce the amount of airtime required for each transmission by allowing higher data rates and avoiding the use of the AP as a relay. Beamforming is an interesting feature for D2D communication because it may allow con￾current transmissions inside the same WLAN by different groups of nodes. IEEE 802.11ax-2019 aims to operate in the 2.4 and 5 GHz bands by superseding and inte￾grating the IEEE 802.11n-2009 and IEEE 802.11ac-2013 amendments. In a few years, every single AP will most likely implement two IEEE 802.11ax-2019 instances that independently oper￾ate at 2.4 and 5 GHz, with a IEEE 802.11ah-2016 instance at 1 GHz for M2M and long-range com￾munications, and one IEEE 802.11ad-2012 (or the future IEEE 802.11ay-202x) at 60 GHz for very fast millimeter-wave communications. In this situation, the STAs of any of those networks should be able to communicate with the STAs in any of the other WLANs. This is the aim of the IEEE 802.11ak-2017 amendment, which focuses on adding bridging capabilities to WLANs. Finally, next-generation WLANs will use traf￾fic differentiation, flow admission control, and groupcast mechanisms from the IEEE 802.11e- 2007, IEEE 802.11ae-2012 and IEEE 802.11aa- 2012 amendments to support multimedia traffic with the required QoS. Also, further advances in power saving mechanisms are expected to keep WLAN energy consumption as low as possible. Conclusion We have reviewed some technological options that could be included in the IEEE 802.11ax- 2019 amendment for next-generation WLANs. Individually, all those solutions offer some per￾Figure 6. Upcoming IEEE 802.11 amendments. 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 t Extended functionalities IEEE 802.11ay PHY/MAC Feature Amendment Objective Fast handoff between APs IEEE 802.11ai-2016 Handoff duration below 100 ms by implementing preemptive channel sensing and user authentication. D2D/WiFi Direct WiFi Alliance feature STAs with D2D capability can decide to create their own WLAN to communicate directly. Multi-technology APs and bridging capabilities IEEE 802.11ak-2017 Interconnection of multiple WLANs. Video traffic differentiation IEEE 8v02.11aa-2012 Stream prioritization and groupcast mechanisms. AP/service selection IEEE 802.11aq-2018 APs provide pre-association information about the services offered to help STAs to choose between multiple APs IEEE 802.11ah IEEE 802.11ax IEEE 802.11aq IEEE 802.11ad IEEE 802.11ai IEEE 802.11ak IEEE 802.11aa IEEE 802.11ac IEEE 802.11n The use of D2D communications will reduce the amount of airtime required for each transmission by allowing higher data rates and avoiding the use of the AP as a relay. Beamforming is an interesting feature for D2D communication because it may allow concurrent transmissions inside the same WLAN by different groups of nodes

In addition to the higher formance gains by improving the spatial reuse [4]F.N.Abinader Jr.,et al.,"Enabling the Coexistence of and spectrum utilization.However,the analysis LTE and Wi-Fi in Unlicensed Bands,"IEEE Commun. achievable throughput Mag,vol.52,no.11,Nov.2014.Pp.54-61 of the actual performance gains when several of [5]S.Rajagopal,"Power Efficiency:The Next Challenge for provided by those solutions are combined and used simul- Multi-Gigabit-Per-Second Wi-Fi,"IEEE Commun.Mag., IEEE802.11ax2019 taneously is still an open challenge requiring vol.52,no.11,Nov.2014pp.40-45. [6]B.Bellalta et al.,"On the Interactions between Multi PHY/MAC features, significant research efforts in the next years.For ple Overlapping WLANs Using Channel Bonding,"/EEE instance,for a given transmission power,multi- Trans.Vehic.Tech.,2015. we believe that the plexing different users in a single transmission [7]I.Jamil,L.Cariou,and J.-F.Helard,"Improving the using channel bonding and spatial multiplexing Capacity of Future IEEE 802.11 High Efficiency WLANs," most disruptive reduces the received power per Hertz and user, Proc.21st IEEE Int'l.Conf.Telecommunications,2014 performance and user which may require the selection of lower-order Pp.303-07. [8]J.I.Choi et al.,"Achieving Single Channel,Full Duplex Wire- experience modulation schemes and coding rates,and the less Communication,"Proc.16th Annual ACM Int'l.Conf. theoretical gain in throughput obtained by the Mobile Computing and Networking.2010.pp.1-12. improvements in combined use of spatial multiplexing and chan- [9]L.Sanabria-Russo et al.,"Future Evolution of CSMA Pro- tocols for the IEEE 802.11 Standard,"Proc.IEEE /CC '13, next-generation WLANs nel bonding may be lost by the larger transmis- 2013,pp.1274-79. will also be related to sion delay. [10]M.Park,"IEEE 802.11ac:Dynamic Bandwidth Channel Finally,in addition to the higher achievable Access,"Proc.2011 IEEE ICC '11,2011,pp.1-5. the development and throughput provided by IEEE 802.11ax-2019 [11]S.Valentin et al.,"Integrating Multiuser Dynamic OFDMA Into IEEE 802.11 WLANs-LLC/Mac Extensions implementation of the PHY/MAC features,we believe that the most and System Performance,"Proc.IEEE /CC '08,2008,pp disruptive performance and user experience 3328-34. other recently approved improvements in next-generation WLANs will [12]R.Liao et al.,"MU-MIMO MAC Protocols for Wireless Local Area Networks:A Survey,"IEEE Commun.Surveys or ongoing also be related to the development and imple- and Tutorials,2014. amendments. mentation of the other recently approved or [13]C.Shepard et al.."Argos:Practical Many-Antenna Base ongoing amendments,and by developing effi- Stations,"Proc.18th Annual ACM Int'l.Conf.Mobile cient and smart mechanisms to improve the Computing and Networking,2012,pp.53-64. coexistence and cooperation among WLANs [14]X.Zhang et al.,"NEMOx:Scalable Network MIMO for Wireless Networks,"Proc.19th Annual ACM Int'l.Conf. ACKNOWLEDGMENTS Mobile Computing&Networking,2013,pp.453-64. [15]D.Camps-Mur,A.Garcia-Saavedra,and P.Serrano This work was partially supported by the Spanish "Device-to-Device Communications with Wi-Fi Direct: and Catalan governments through the projects Overview and Experimentation,"IEEE Wireless Com- TEC2012-32354 and SGR-2014-1173,respec- mun.,vol.20,no.3,June 2013. tively. BIOGRAPHY REFERENCES BoRIs BELLALTA is an associate professor in the Deparment of Information and Communication Technologies at University [1]IEEE 802.11 WLANs,Working Group for WLAN Stan- Pompeu Fabra(UPF),where he leads the Wireless Network- dards 2015 ing research group.He obtained his degree in telecom- [2]IEEE 802.11 Task Group AX,"Status of Project IEEE munications engineering from Universitat Politecnica de 802.11ax High Efficiency WLAN(HEW)";accessed July Catalunya in 2002 and his Ph.D.in information and com- 2015,http://www.ieee802.org/11/Reports/tgax update. munication technologies from UPF in 2007.His research htm,2015. interests are in the area of wireless networks,with empha- [3]B.Bellalta et al."Channel Bonding in Short-Range WLANs," sis on the design and performance evaluation of PHY/MAC Proc.20th IEEE Euro.Wireless Conf.,2014,pp.1-7. protocols and network architectures. IEEE Wireless Communications.February 2016

46 IEEE Wireless Communications • February 2016 formance gains by improving the spatial reuse and spectrum utilization. However, the analysis of the actual performance gains when several of those solutions are combined and used simul￾taneously is still an open challenge requiring significant research efforts in the next years. For instance, for a given transmission power, multi￾plexing different users in a single transmission using channel bonding and spatial multiplexing reduces the received power per Hertz and user, which may require the selection of lower-order modulation schemes and coding rates, and the theoretical gain in throughput obtained by the combined use of spatial multiplexing and chan￾nel bonding may be lost by the larger transmis￾sion delay. Finally, in addition to the higher achievable throughput provided by IEEE 802.11ax-2019 PHY/MAC features, we believe that the most disruptive performance and user experience improvements in next-generation WLANs will also be related to the development and imple￾mentation of the other recently approved or ongoing amendments, and by developing effi￾cient and smart mechanisms to improve the coexistence and cooperation among WLANs. Acknowledgments This work was partially supported by the Spanish and Catalan governments through the projects TEC2012-32354 and SGR-2014-1173, respec￾tively. References [1] IEEE 802.11 WLANs, Working Group for WLAN Stan￾dards, 2015. [2] IEEE 802.11 Task Group AX, “Status of Project IEEE 802.11ax High Efficiency WLAN (HEW)”; accessed July 2015, http://www.ieee802.org/11/Reports/tgax update. htm, 2015. [3] B. Bellalta et al., “Channel Bonding in Short-Range WLANs,” Proc. 20th IEEE Euro. Wireless Conf., 2014, pp. 1–7. [4] F.N. Abinader Jr., et al., “Enabling the Coexistence of LTE and Wi-Fi in Unlicensed Bands,” IEEE Commun. Mag., vol. 52, no. 11, Nov. 2014, pp. 54–61. [5] S. Rajagopal, “Power Efficiency: The Next Challenge for Multi-Gigabit-Per-Second Wi- Fi,” IEEE Commun. Mag., vol. 52, no. 11, Nov. 2014, pp. 40–45. [6] B. Bellalta et al., “On the Interactions between Multi￾ple Overlapping WLANs Using Channel Bonding,” IEEE Trans. Vehic. Tech., 2015. [7] I. Jamil, L. Cariou, and J.-F. Helard, “Improving the Capacity of Future IEEE 802.11 High Efficiency WLANs,” Proc. 21st IEEE Int’l. Conf. Telecommunications, 2014, pp. 303–07. [8] J. I. Choi et al., “Achieving Single Channel, Full Duplex Wire￾less Communication,” Proc. 16th Annual ACM Int’l. Conf. Mobile Computing and Networking, 2010, pp. 1–12. [9] L. Sanabria-Russo et al., “Future Evolution of CSMA Pro￾tocols for the IEEE 802.11 Standard,” Proc. IEEE ICC ’13, 2013, pp. 1274–79. [10] M. Park, “IEEE 802.11ac: Dynamic Bandwidth Channel Access,” Proc. 2011 IEEE ICC ’11, 2011, pp. 1–5. [11] S. Valentin et al., “Integrating Multiuser Dynamic OFDMA Into IEEE 802.11 WLANs-LLC/Mac Extensions and System Performance,” Proc. IEEE ICC ’08, 2008, pp. 3328–34. [12] R. Liao et al., “MU-MIMO MAC Protocols for Wireless Local Area Networks: A Survey,” IEEE Commun. Surveys and Tutorials, 2014. [13] C. Shepard et al., “Argos: Practical Many-Antenna Base Stations,” Proc. 18th Annual ACM Int’l. Conf. Mobile Computing and Networking, 2012, pp. 53–64. [14] X. Zhang et al., “NEMOx: Scalable Network MIMO for Wireless Networks,” Proc. 19th Annual ACM Int’l. Conf. Mobile Computing & Networking, 2013, pp. 453–64. [15] D. Camps-Mur, A. Garcia-Saavedra, and P. Serrano, “Device-to-Device Communications with Wi-Fi Direct: Overview and Experimentation,” IEEE Wireless Com￾mun., vol. 20, no. 3, June 2013. Biography Boris Bellalta is an associate professor in the Deparment of Information and Communication Technologies at University Pompeu Fabra (UPF), where he leads the Wireless Network￾ing research group. He obtained his degree in telecom￾munications engineering from Universitat Politecnica de Catalunya in 2002 and his Ph.D. in information and com￾munication technologies from UPF in 2007. His research interests are in the area of wireless networks, with empha￾sis on the design and performance evaluation of PHY/MAC protocols and network architectures. In addition to the higher achievable throughput provided by IEEE 802.11ax-2019 PHY/MAC features, we believe that the most disruptive performance and user experience improvements in next-generation WLANs will also be related to the development and implementation of the other recently approved or ongoing amendments