March 2018 doc IEEE 802 11 180435 r

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 WUR Preamble Sequence Design and Performance Evaluation Date: 2018 -03 -05 Authors: Name Affiliation Justin Jia Huawei Technologies Ming Gan Huawei Technologies Wei Lin Huawei Technologies Chenchen Liu Huawei Technologies Jun Zhu Huawei Technologies Submission Address Slide 1 Phone Email justin. jia@huawei. com Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Introduction • Submission Slide 2 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Metrics • Submission Slide 3 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Metrics Analysis • Multiple Auto/Cross-Correlation Metrics may be not necessary – – It is impossible to have two sets of local reference sequence Refs at the correlator of the receiver Data rates can be identified by the sign of output of correlator at the receiver side • • Auto Correlation Metric can ensure the timing performance – – The output sign of correlator in the numberator can be used to distinguish the two data rates The absolute value of second largest peak in denominator is used for our metrics to avoid the timing error and false rate detection • • • Equal number of Ones and Zeros It could be further enhanced by the number of peaks A huge second largest peak with the same sign as the largest peak may cause timing error A huge second largest peak with the opposite sign as the largest peak may cause false rate detection In this contribution, we pick up four candidates of 32 -bit-based basic sequence S 1[2], S 2[3], S 3[4], and S 4[5] and evaluate their performance – – Submission S 1 = [1 0 0 0 1 1 0 0 1 1 1 0 0 0 1]; S 2 = [0 1 1 1 0 1 0 1 1 0 0 0 1 1 0]; S 3 = [1 0 0 1 0 1 1 1 0 0 1 1 1 0 0 0]; S 4 = [1 0 1 0 0 1 1 0 0 1]; Slide 4 Recommended Sequence Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Correlation Property of S 1 • Correlations at the receiver for S 1: Better performance for the high rate Submission Slide 5 ACMetric <High. R> -8. 0 ACMetric <Low. R> 4. 0 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Correlation Property of S 2 • Correlations at the receiver for S 2: Submission Slide 6 ACMetric <High. R> -3. 2 ACMetric <Low. R> 4. 0 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Correlation Property of S 3 • Correlations at the receiver for S 3: ACMetric <High. R> -5. 3 ACMetric <Low. R> 8. 0 Better performance for the low rate Submission Slide 7 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Correlation Property of S 4 • Correlations at the receiver for S 4: Submission Slide 8 ACMetric <High. R> -2. 7 ACMetric <Low. R> 2. 3 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Metrics for preamble sequence • Summary on Metrics comparison – S 3 has the best performance for data rate, however, S 1 performs best on high data rate Submission S 1 S 2 S 3 S 4 ACMetric <High. R> -8 -3. 2 -5. 3 -2. 7 ACMetric <Low. R> 4 4 8 2. 3 Slide 9 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 AGC Related • The concerns of AGC adjustment are listed as follows: – The Max OFF time in high/low data rate preambles, respectively – Silence Period Location[5]: The location of first 4 us OFF period in high/low data rate preambles, respectively Submission S 1 S 2 S 3 S 4 8 us 6 us 6 us 4 us 12 us 20 us 22 us 6 us 16 us 14 us Slide 10 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Simulation Settings[6] • Submission Slide 11 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Performance Comparison Submission Slide 12 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Summary • Through the simulation results and metrics analysis, we obtain – Low rate case: S 3 outperforms S 1, S 2 and S 4 by about 0. 28 d. B, 0. 35 d. B and 0. 54 d. B gain @1% PER, respectively, where S 1 and S 2 have similar overall PER performance – High rate case: S 1 outperforms S 2, S 3 and S 4 by 0. 55 d. B, 0. 3 d. B and 0. 59 d. B gain @1% PER, respectively, where S 2 and S 4 have similar overall PER performance – It is recommended to choose S 3 since the performance in low data rate environment is more concerned. Low Rate Submission High Rate SNR@ 1% PER SNR@ 10% PER S 1 6. 07 d. B 0. 64 d. B 8. 87 d. B 4. 07 d. B S 2 6. 00 d. B 0. 66 d. B 9. 42 d. B 4. 15 d. B S 3 5. 72 d. B 0. 61 d. B 9. 17 d. B 4. 20 d. B S 4 6. 26 d. B 0. 73 d. B 9. 46 d. B 4. 16 d. B Slide 13 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Straw Poll • Submission Slide 14 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Reference [1] IEEE 802. 11 -17/0575 r 9, “Specification Framework for TGba”, Jan. 2018 [2] IEEE 802. 11 -17/1636 r 0, “A Simple WUR Preamble Design”, Nov. 2017 [3] IEEE 802. 11 -18/0123 r 0, “Options for Sync Field Bit Sequence”, Jan. 2018 [4] IEEE 802. 11 -18/0100 r 1, “WUR Preamble Sequence Performance Evaluation”, Jan. 2018 [5] IEEE 18/0096 r 3, “WUR Sync Design”, Jan. 2018 [6] IEEE 802. 11 -17/0188 r 9, “Simulation Scenario and Evaluation Methodology”, Jul. 2017 [7] IEEE 802. 11 -17/1617 r 1, “Dual Sync Designs”, Nov. 2017 Submission Slide 15 Justin Jia, et. al. , Huawei

March 2018 doc. : IEEE 802. 11 -18/0435 r 0 Appendix 1 Comparison based on other Metrics[7] ACMetric (Short) Descriptions Ref CCMetric (Short, Long) ACMetric (Long) Short Preamble*2 -1 Correlating X(n)=xcorr(Ref, Short Preamble) CCMetric (Long, Short) Long Preamble*2 -1 Y(n)=xcorr(Ref, Long X(n)=xcorr(Ref, Long Y(n)=xcorr(Ref, Short Preamble) Algorithm (Non-absolute value) Sequence ACMetric (Short) CCMetric (Short, Long) Sum (using Short Preamble) ACMetric (Long) CCMetric (Long, Short) Sum (using Long Preamble) Overall Sum S 1 8 4 12 2 8 10 22 S 2 5. 3 4 9. 3 2 6. 4 8. 4 17. 7 S 3 8 8 16 2 10. 7 12. 7 28. 7 S 4 5. 3 2. 3 7. 6 2 5. 3 7. 3 14. 9 Submission Slide 16 Justin Jia, et. al. , Huawei