1. Analysis Results of Performance Metrics for Synchronous HARQ and Asynchronous HARQ Types IEEE 802.16 Presentation Submission Template (Rev. 9) Document Number: IEEE C802.16m-08/856 Date Submitted: 2008-07-14 Source: Jaeweon Cho, Junsung Lim, Hokyu Choi, Jaehee Cho, Voice:+82-31-279-5796 Heewon Kang, Rakesh Taori E-mail:jaeweon.cho@samsung.com Samsung Electronics Co., Ltd. 416 Maetan-3, Suwon, 443-770, Korea Thierry Lestable Samsung Electronics Research Institute, UK Venue: IEEE C802.16m-08/008r2, “HARQ Rapporteur Group Chairs' initial recommendations on definitions of the HARQ performance metrics as well as recommendations on common assumptions to be used for evaluation of the HARQ proposals.” Base Contribution: None Purpose: For discussion and adoption in the 802.16m HARQ Rapporteur Group Notice: This document does not represent the agreed views of the IEEE 802.16 Working Group or any of its subgroups. It represents only the views of the participants listed in the “Source(s)” field above. It is offered as a basis for discussion. It is not binding on the contributor(s), who reserve(s) the right to add, amend or withdraw material contained herein. Release: The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this contribution may be made public by IEEE 802.16. Patent Policy: The contributor is familiar with the IEEE-SA Patent Policy and Procedures: and.http://standards.ieee.org/guides/bylaws/sect6-7.html#6http://standards.ieee.org/guides/opman/sect6.html#6.3 Further information is located at and.http://standards.ieee.org/board/pat/pat-material.htmlhttp://standards.ieee.org/board/pat

2. Analysis Results of Performance Metrics for Synchronous HARQ and Asynchronous HARQ Types Jaeweon Cho, Junsung Lim, Hokyu Choi, Jaehee Cho, Heewon Kang, Rakesh Taori Samsung Electronics Co., Ltd. Thierry Lestable Samsung Electronics Research Institute, UK July, 2008

3. About this Contribution Provide Quantitative and Qualitative analysis results on the HARQ performance metrics of different HARQ types Focus on comparison of the two key attributes of HARQ - i.e. Synchronous vs. Asynchronous, and Adaptive vs. Non-adaptive The considered performance metrics (based on C802.16m-HARQ-08/008r2) - System Throughput - Packet Delay - Signaling Overhead - Power Saving - Retransmission Scheduling Flexibility (Support of persistent scheduling) - Robustness 2

4. 3 DL SLS for Throughput and Delay Analysis TDD, D:U = 4:4 Scheduling algorithm: Proportional Fairness Non-adaptive and chase combing HARQ operation Downlink SLS parameters (based on 802.16m EMD) Simulation assumptionsDescriptions Frequency reuse1 Ant/ Receiver structureSIMO (1X2) / MMSE Channel EstimationIdeal CQI report period1 frame (5ms) ACK/NACK error rate0% Target PER10% Number of user / sector16 Max number of retransmission4 Test scenarioConfiguration Site-to-site distance1.5km BS Tx power46dBm Penetration loss10dB Antenna GainBS: 17dBi, MS: 0dBi Pathloss modelLoss(dB)=130.62+37.6log(R) Lognormal shadowing STD 8dB Channel mix ITU Ped B 3km/hr – 60% ITU Veh A 30km/hr – 30% ITU Veh A 120km/hr – 10%

5. Throughput and Delay Async HARQ does not provide any noticeable gain in throughput Nonetheless, packet delay is significantly increased with Async HARQ 4 Throughput (Mbps) Latency; Avg. (ms) Latency; 90%-tile (ms) Avg. num of Tx / Burst Distribution of Tx (%) 1st2nd3rd4th5th Sync HARQ 7.571.343.21.2481.8113.064.100.850.14 Async HARQ 7.70 (+1.7%) 8.89201.2381.2915.193.140.350.03 * Throughput: Pure system throughput without signaling (MAP) overhead * Latency: Time duration from the 1 st transmission to the successful transmission

6. Signaling Overhead In this analysis, assumed the same fields as in 802.16e Signaling overhead with considering the re-transmission 45.51 bits - Async and Adaptive: 1.23 (Avg. num of Tx; from SLS) x 37 bits = 45.51 bits 30 bits - Sync and Non-adaptive: 1 x 30 bits = 30 bits - Overhead increase for Async and Adaptive: 51.7%! - Overhead increase for Async and Adaptive: (45.51-30)/30 = 51.7%! 5 Field802.16e Asynchronous and Adaptive Synchronous and Non-Adaptive Resource indicationDuration (OFDMA slots)10 bits MCSDIUC/UIUC4 bits HARQ process numberACID4 bits- New packet indicatorAI_SN1 bits- Redundancy versionSPID2 bits- CID 16 bits Total37 bits30 bits

7. 6 Power Saving Sync HARQ is very friendly with power saving 1)The timing of retransmission is pre-determined MS may go sleep mode between retransmissions, till success of Tx 2)In a more aggressive way; Default sub-frame concept + Sync HARQ  Implicit Sleep Mode MS needs to monitoring the pre-assigned sub-frame only MS may go sleep mode during all other sub-frames

8. ReTx Scheduling Flexibility Propose Synchronous HARQ with allowing resource relocation - Synchronous and non-adaptive operation as default. - However, if necessary, the resource can be relocated with a partial signaling, for efficient support of persistent scheduling. Resource relocation (i.e. limited adaptation) - At the time of synchronous retransmission, the resource can be relocated on frequency domain, - Or the resource can be rolled over in unit of interlace. (i.e. BS defers the resource allocation for retransmission) - So, a flexibility in resource location and time is provided. 7

9. Robustness Sync HARQ: Comparable Robustness to Async HARQ; Recovery from Initial MAP loss - Various handling schemes of MAP loss can be applied E.g. MAP loss detection by using MAP ACK_CH (explicit way), or MAP loss detection by Null detection of data ACK_CH (implicit way) Recovery from ACK_CH error - NACK to ACK error The occurrence probability of this error is very low E.g. Occurrence Pr. ~= 20~10% (1 st Tx failure Pr.) x 0.1~0.01% (NACK to ACK error) = 0.02~0.001% - ACK to NACK error When this error occurs, BS retransmits packet but MS does not anything Then, this error can be detected by Null detection of data ACK_CH 8

10. 9 Summary The proposed Sync HARQ over Async/Adaptive HARQ - System throughput : Almost Same - Packet delay : Much Lower - Signaling overhead : Much Lower - Power Saving : Very Friendly - ReTx scheduling flexibility : Provided by Allowing the Resource Relocation - Robustness : Comparable by employing MAP ACK_CH and/or Null detection of data ACK_CH Proposal of working assumption for 802.16m HARQ (C802.16m-HARQ-08/021) - In downlink and uplink, HARQ is based on synchronous retransmission. - Resource at the retransmissions can be fixed or adaptive according to control signaling.