1 / 13

MAC Efficiency Analysis for HEW SG

MAC Efficiency Analysis for HEW SG . Authors:. Date: 2013-05-13. Dense deployments of 802.11 STAs were discussed in WNG SC as usage models for HEW SG (see reference list) Wireless Office Cellular offload (home, corporate, hotspot) Residential / apartment building Shopping mall

kaethe
Download Presentation

MAC Efficiency Analysis for HEW SG

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. MAC Efficiency Analysis for HEW SG Authors: Date: 2013-05-13 Minyoung Park (Intel Corp.)

  2. Dense deployments of 802.11 STAs were discussed in WNG SC as usage models for HEW SG (see reference list) • Wireless Office • Cellular offload (home, corporate, hotspot) • Residential / apartment building • Shopping mall • Airport waiting halls/lounges • Stadium • Lecture Hall • Airliner / railroad car • 802.11 MAC is based on contention based medium access. • This presentation analyzes 802.11 MAC efficiency for such dense deployment scenarios. Introduction Minyoung Park (Intel Corp.)

  3. MAC Overhead and Efficiency TXOP up to 64 subframes Max A-MPDU size ~ 1MB • MAC efficiency increases as • TXOP  fixed overhead (e.g. DIFS and SIFS) • Number of BAs  aggregation overhead • CW (or # of stations) + collisions  contention overhead CW (variable) Actual data pr RTS SIFS pr A-MPDU SIFS pr A-MPDU SIFS DIFS SIFS SIFS pr CTS pr BA pr BA overhead overhead overhead Total MAC payload (bits) MAC throughput = Time consumed for transmitting total MAC payload (sec) MAC throughput MAC efficiency = PHY rate Minyoung Park (Intel Corp.)

  4. MAC Throughput with Contention Collision overhead TXOP CWmax doubled CTS Timeout* CW=4 CW=2 CW=8 STA-A SIFS RTS RTS RTS SIFS SIFS A-MPDU SIFS DIFS DIFS CTS DIFS CTS SIFS BA CWmax doubled freeze CW=2 CW=10 CW=2 STA-B RTS SIFS A-MPDU SIFS RTS SIFS RTS DIFS CTS SIFS BA DIFS CTS DIFS freeze • Aggregate throughput: STA-A + STA-B overhead overhead Collision overhead Actual data Actual data overhead CW=8 CW=2 RTS SIFS A-MPDU SIFS RTS SIFS RTS SIFS SIFS A-MPDU SIFS DIFS CTS SIFS BA DIFS CTS DIFS CTS SIFS BA Idle CW time slots When there is no collision between the stations, the station with the smallest CW will access the channel and thus CW overhead gets smaller as the number of stations increases Overhead due to collision is shared byall the stations in the network *) CTS Timeout = SIFS + aPHY-RX-START-Delay + aSlotTime Minyoung Park (Intel Corp.)

  5. Number of contending STAs: 1~100, 200 The collision probability increases as the number of STAs increases Contention Analysis 1 Minyoung Park (Intel Corp.)

  6. Idle slots during contention window • The overhead due to idle time slots during contention windows decreases as the number of STAs increases Contention Analysis 2 Minyoung Park (Intel Corp.)

  7. Simulation Setup for Single BSS case • 1 AP, 1 ~ 50 STAs • Payload 1500 Bytes • A-MPDU: max 64 frames • No limitation in size • TXOP = 1, 2, 3 mS • STA configuration • 2x2 MIMO, 40 MHz • Transmit power = 17dBm • PHY rate • Maximum PHY rate = 270 Mbps • Link adaptation enabled • MCS feedback during RTS/CTS exchange • Channel model • 5GHz, TGn channel B • Simulation time: 10 seconds Bi-directional traffic AP STA 25 meters Minyoung Park (Intel Corp.)

  8. TXOP Limit = 3mS As the number of STAs increases, MAC efficiency drops due to increased contention overhead MAC Efficiency Minyoung Park (Intel Corp.)

  9. Medium access delay increases as the number of STAs increases and the TXOP increases Medium Access Delay • Medium access delay is measured as shown in the following figure STA2’s TXOP STA1’s TXOP STA1’s TXOP A-MPDU Medium access delay RTS, CTS, BA Minyoung Park (Intel Corp.)

  10. MAC efficiency measured for different TXOP Limits: • TXOP = 1, 2, and 3 mS • MAC efficiency drops as TXOP gets shorter due to increased fixed MAC overhead per TXOP MAC Efficiency – Different TXOPs Minyoung Park (Intel Corp.)

  11. Overlapping BSSs Case • Setup • Number of APs = 7 • Number of STAs per BSS = 30 STAs • Total number of STAs = 210 STAs • TXOP = 3 mS • Simulation time = 7 seconds • 2.4 GHz • Other parameters are same as the single BSS case (shown in slide 7) • MAC efficiency measured at BSS1 • Average medium access delay = 648 mS • Increased interference and contention from overlapping networks lowers MAC efficiency 30 m 45 m BSS1 Each BSS has 30 STAs Minyoung Park (Intel Corp.)

  12. Dense deployment scenarios have been discussed for HEW SG • 802.11 MAC efficiency is analyzed for dense deployment scenarios • MAC efficiency drops as STA density increases • SG needs to define metrics to measure QoE • Minimum required average (or CDF) throughput per STA • Maximum medium access delay • Average area throughput • … • Investigate existing features in other task groups (11aa/ac/ad/ah/…) to see how those features affect PHY/MAC efficiency in dense deployment scenarios • HCCA (scheduling), 11aa OBSS management, 11ac MU-MIMO, etc. Summary and Next Steps Minyoung Park (Intel Corp.)

  13. https://mentor.ieee.org/802.11/dcn/13/11-13-0331-05-0wng-high-efficiency-wlan.ppthttps://mentor.ieee.org/802.11/dcn/13/11-13-0331-05-0wng-high-efficiency-wlan.ppt https://mentor.ieee.org/802.11/dcn/13/11-13-0287-03-0wng-beyond-802-11ac-a-very-high-capacity-wlan.pptx https://mentor.ieee.org/802.11/dcn/13/11-13-0343-00-0wng-operator-oriented-wi-fi.pptx https://mentor.ieee.org/802.11/dcn/13/11-13-0309-00-0wng-next-gen-wlan.pptx https://mentor.ieee.org/802.11/dcn/13/11-13-0313-00-0wng-usage-models-for-next-generation-wi-fi.pptx https://mentor.ieee.org/802.11/dcn/13/11-13-0314-00-0wng-on-future-enhancements-to-802-11-technology.pptx https://mentor.ieee.org/802.11/dcn/13/11-13-0113-00-0wng-applications-and-requirements-for-next-generation-wlan.pptx https://mentor.ieee.org/802.11/dcn/13/11-13-0514-00-0hew-usage-scenarios-and-applications.pptx References Minyoung Park (Intel Corp.)

More Related