Submission doc.: IEEE 802.11-15/914r1 July 2015 Guido R. Hiertz et al., EricssonSlide 1 Enlarged minimal contention window size Date: 2015-07-14 Authors:

Slides:



Advertisements
Similar presentations
Doc.: IEEE /080r1 Submission January 2001 Jie Liang, Texas InstrumentsSlide 1 Jie Liang Texas Instruments Incorporated TI Blvd. Dallas,
Advertisements

Doc.: IEEE /0665r1 Submission May 2012 Anh Tuan Hoang et al (I2R) Slide 1 Prioritized PS-Polls Date: Authors:
Discussion on OFDMA in HEW
Submission doc.: IEEE 11-14/1426r1 November 2014 Gustav Wikström et al., EricssonSlide 1 DSC and legacy coexistence Date: Authors:
Submission Kai Kang, SHRCWC May 2013 A Mechanism to Provide QoS in IEEE e MAC Date: Authors: Slide 1.
Submission doc.: IEEE /1225r1 Considerations on CCA for OBSS Opearation in ax Date: Slide 1Huawei Authors:
802.11g & e Presenter : Milk. Outline g  Overview of g  g & b co-exist QoS Limitations of e  Overview of.
Submission doc.: IEEE /1454r1 November 2014 Jarkko Kneckt (Nokia)Slide ax Power Save Discussion Date: Authors:
Doc.: IEEE /0861r0 SubmissionSayantan Choudhury Impact of CCA adaptation on spatial reuse in dense residential scenario Date: Authors:
Doc.: IEEE /1207r1 Submission Imad Jamil (Orange)Slide 1 OBSS reuse mechanism which preserves fairness Date: Authors: September 2014.
Doc.: IEEE /1443r0 SubmissionEsa Tuomaala Adapting CCA and Receiver Sensitivity Date: Authors: Slide 1 November 2014.
Discussion on OFDMA in IEEE ax
802.11ax scenario 1 CCA Date: Authors: March 2015
Submission doc.: IEEE /1454r0 November 2014 Jarkko Kneckt (Nokia)Slide ax Power Save Discussion Date: Authors:
Doc.: IEEE /1126r0 Submission September 2012 Krishna Sayana, SamsungSlide 1 Wi-Fi for Hotspot Deployments and Cellular Offload Date:
Submission doc.: IEEE /870r2 July 2015 Guido R. Hiertz et al., EricssonSlide ax in 2.4 GHz Date: Authors:
A Virtual Collision Mechanism for IEEE DCF
Submission doc.: IEEE /1015r1 September 2015 Guido R. Hiertz et al., EricssonSlide 1 Proxy ARP in ax Date: Authors:
Submission doc.: IEEE /1013r0 September 2015 Guido R. Hiertz et al., EricssonSlide ae & ax Date: Authors:
Submission doc.: IEEE /0801r1 Akira Kishida, NTT Issues of Low-Rate Transmission Date: Authors:
Doc.: IEEE /0648r0 Submission May 2014 Chinghwa Yu et. al., MediaTekSlide 1 Performance Observation of a Dense Campus Network Date:
Doc.: IEEE /065r0 Submission January 2001 Brockmann, Hoeben, Wentink (Intersil) g MAC Analysis Menzo Wentink Ron Brockmann.
Submission doc.: IEEE /1108r0 Technical Feasibility for LRLP September 2015 Chittabrata Ghosh, IntelSlide 1 Date: Authors:
Doc.: ax Submission July 2014 Slide 1 Proposed Calibration For MAC simulator Date: Authors:
Doc.: IEEE /1207r0 Submission Imad Jamil (Orange)Slide 1 OBSS reuse mechanism which preserves fairness Date: Authors: September 2014.
Doc.: IEEE /0637r0 Submission May 2014 James Wang et. al., MediaTekSlide 1 Spatial Reuse and Coexistence with Legacy Devices Date:
Doc.: IEEE /0523r0 Submission April 2014 Imad Jamil (Orange)Slide 1 MAC simulation results for Dynamic sensitivity control (DSC - CCA adaptation)
Submission doc.: IEEE /1014r0 September 2015 Guido R. Hiertz et al., EricssonSlide 1 Multiple BSSID element Date: Authors:
Submission doc.: IEEE /0372r2 Slide 1 System Level Simulations on Increased Spatial Reuse Date: Authors: Jinjing Jiang(Marvell) March.
Submission doc.: IEEE /1289r2 Michelle Gong, IntelSlide 1 RTS/CTS Operation for Wider Bandwidth Date: Authors: Nov
Doc.: IEEE /0814r0 Submission July 2015 Simulation Results for Box5 Calibration Ke Yao, et, al. (ZTE) Slide 1 Date: Authors: NameAffiliationAddress .
Doc.: IEEE /1172r2 Submission September 2014 Eisuke Sakai, Sony CorporationSlide 1 Multicast Performance in OBSS Date: 2014/9/15 Authors:
Doc.: IEEE / ax Submission M. Shahwaiz Afaqui DSC calibration results with NS-3 Authors: Nov
Doc.: IEEE /1280r1 November 2015 SubmissionStéphane Baron et. al., Canon Traffic priority for random Multi User Uplink OFDMA Date: Slide.
Discussion on ax functional requirements
Doc.: IEEE / ax Submission M. Shahwaiz Afaqui DSC calibration results with NS-3 Authors: Nov
Doc.: IEEE /0619r3 Submission May 2012 Haiguang Wang et. al, I2R, SingaporeSlide 1 Overlapping IEEE ah Networks of Different Types Date:
Doc.: IEEE /0294r1 Submission Dynamic Sensitivity Control Channel Selection and Legacy Sharing Date: Authors: Graham Smith, DSP GroupSlide.
Doc.: IEEE /0097r0 SubmissionJarkko Kneckt (Nokia)Slide 1 Bandwidth Specific TXOP Limits Date: Authors: January 2011.
Resolutions to Static RTS CTS Comments
Submission doc.: IEEE /0662r0 May, 2016 Jing Ma, NICTSlide 1 Further consideration on channel access rule to facilitate MU transmission opportunity.
Submission doc.: IEEE /1359r0 November 2015 Yu Wang, Ericsson et al.Slide 1 System Performance Evaluation of ae Date: Authors:
Doc.: IEEE /1357r0 Submission November 2008 Guido R. Hiertz et al., PhilipsSlide 1 Simple improvement for EDCA usage in s Date:
Secondary Channel CCA of HE STA
EA C451 (Internetworking Technologies)
802.11ax in 2.4 GHz Date: Authors: July 2015
QoS Handling of Trigger Frame
IEEE : Wireless LANs ALOHA, Slotted ALOHA
Proposed response to 3GPP ED request
Simulation Results for Box5
Below 6GHz 11vht PAR scope and purpose discussion
Updated Simulation Results for Box5
Simulation Results for Box5
Raising the PAR Date: Authors: January 2014 January 2014
Enhanced IEEE by Integrating Multiuser Dynamic OFDMA
Increased Network Throughput with Channel Width Related CCA and Rules
Simulation Results for Box5
VTS SG PAR Scope Topics Date: Authors: January 2008
Box5 Calibration Results
Considerations on CCA for OBSS Opearation in ax
Box 5 Calibration Result
Airtime Analysis of EDCA
OBSS Requirements Date: Authors: July 2008 July 2008
Coex Simulation and Analysis
System Level Simulator Evaluation with/without Capture Effect
[SDMA operation within ]
OBSS Requirements Date: Authors: July 2008 July 2008
DSC Calibration Result
Channel usage in NGV: follow-up
Month Year doc.: IEEE /1081r0 May, 2016
Presentation transcript:

Submission doc.: IEEE /914r1 July 2015 Guido R. Hiertz et al., EricssonSlide 1 Enlarged minimal contention window size Date: Authors:

Submission doc.: IEEE /914r1 July 2015 Guido R. Hiertz et al., EricssonSlide 2 Abstract OFDM PHYs foresee a minimal Contention Window size of 15. With many stations contending for access to the wireless medium (WM), this initial Contention Window might be too small. A simple approach is presented that increases the number of contention slots without impacting any performance relevant aspects.

Submission doc.: IEEE /914r1 Minimal Contention Window size, 2.4 GHz band DSSS & FHSS CWmin = b-1999 HR/DSSS CWmin = g-2003 Long slot time mode CWmin = 31 Short slot time mode CWmin = DSSS set baseline in 2.4 GHz Most g deployments in “compatibility” mode Use of CWmin = 31 prevailing n-2009 uses CWmin = 15 in 2.4 GHz July 2015 Guido R. Hiertz et al., EricssonSlide 3

Submission doc.: IEEE /914r1 Minimal Contention Window size, 5 GHz band a-1999 First use of OFDM CWmin = n-2009 MIMO upgrade CWmin = ac n upgrade CWmin = 15 All standards for the 5 GHz band use CWmin = 15 July 2015 Guido R. Hiertz et al., EricssonSlide 4

Submission doc.: IEEE /914r e – Quality of Service e introduces support for QoS Distributed Coordination Function (DCF) replaced by Enhanced Distributed Channel Access (EDCA) EDCA introduces four priorities EDCA priorities use different minimal and maximal contention window sizes Priorities adapt scaled CWmin value All n & ac STAs implement EDCA All priority values derived from CWmin Apart from AIFSN July 2015 Guido R. Hiertz et al., EricssonSlide 5

Submission doc.: IEEE /914r1 CWmin design A small, initial Contention Window causes high collision probability Airtime waste, growing contention windows, high jitter, retransmissions, … Compromise: Single STA (peak) performance vs. optimal performance with many STAs No One-fits-all value July 2015 Guido R. Hiertz et al., EricssonSlide 6

Submission doc.: IEEE /914r1 Dense deployments One BSS may serve many STAs With EDCA, AP can enlarge CWmin per priority Reduces peak performance Reduces collision probability & may increase performance Current default values target single BSS performance EDCA parameter set adjustment seldom used Some implementations do not adopt parameters signaled by AP No influence on EDCA parameter set selection in neighboring BSSs July 2015 Guido R. Hiertz et al., EricssonSlide 7

Submission doc.: IEEE /914r1 How to reduce collision probability? Enlarge CWmin Helps to reduce collision probability How to select optimal value? Dynamic, adaptive process needed How to signal/agree needed CWmin between different BSSs? Enlarging default CWmin with ax Introduces coexistence problem PHYs prior to ax use CWmin = 15 Would decrease ax access probability compared to prior PHYs July 2015 Guido R. Hiertz et al., EricssonSlide 8

Submission doc.: IEEE /914r1 Enlarge CWmin AND reduce aSlotTime Consider 5 GHz aSlotTime = 9 µs CWmin = 15 9 µs × 16/2 = 72 µs Select ax aSlot & CWmin to achieve mean wait duration as previous PHYs Const = aSlotTime × (CWmin + 1)/ ax default set aSlotTime = 4.5 µs CWmin = 31 Twice the number of contention slots compared to n & ac Mean wait duration: 4.5 µs × 32/2 = 72 µs July 2015 Guido R. Hiertz et al., EricssonSlide 9

Submission doc.: IEEE /914r1 aSlotTime definition aSlotTime = aRxTxTurnaroundTime + aMACProcessingDelay + aAirPropagationTime + aCCATime Implementation dependent with ac aCCATime aMACProcessingDelay aRxTxTurnaroundTime Static (range dependent) aAirPropagationTime July 2015 Guido R. Hiertz et al., EricssonSlide 10

Submission doc.: IEEE /914r ax aSlotTime = 4.5 µs Assume 300 m range aAirPropagationTime = 1 µs Thus today, aCCATime + aMACProcessingDelay + aRxTxTurnaroundTime ≤ 8 µs To meet aSlotTime goal reduce previous sum from 8 µs to 3.5 µs Save 5 µs to sustain 300 m range assumption aCCATime < 4 µs with a-1999 aMACProcessingDelay < 2 µs with a-1999 aRxTxTurnaroundTime < 2 µs with a-1999 With , all values are noted as implementation dependent Only values for interoperability specified anymore July 2015 Guido R. Hiertz et al., EricssonSlide 11

Submission doc.: IEEE /914r1 Feasibility study aMACProcessingDelay Max time for MAC to issue PHY-TXSTART.request Today’s hardware is more powerful than a hardware in 1999 aRxTxTurnaroundTime Max time to change from Rx to Tx Hardware evolution provides for substantially faster turnaround than in 1999 aCCATime Max time that CCA has available to detect start of a valid IEEE transmission Major part in aSlotTime Today’s hardware needs less time than previous implementations aSlotTime = 4.5 µs seems possible July 2015 Guido R. Hiertz et al., EricssonSlide 12

Submission doc.: IEEE /914r1 Further changes, observations AIFS = SIFS + n × aSlotTime With ax double n to compensate for halved aSlotTime Keep Duration (AIFS ac ) = Duration (AIFS ax ) With enlarged CWmin & reduced aSlotTime ax STAs achieve advantage over non ax STAs ax STAs may access earlier Acceptable “penalty” on legacy STAs to be discussed July 2015 Guido R. Hiertz et al., EricssonSlide 13

Submission doc.: IEEE /914r1 Conclusion aSlotTime = 4.5 µs seems possible with ax Increases Contention Window, reduces collision probability Suitable with dense deployments = ax scope Halved aSlotTime gives earlier access to ax STAs Simulations to follow to show impact Can be balanced with legacy devices Simple and non- intrusive approach July 2015 Guido R. Hiertz et al., EricssonSlide 14