Presentation is loading. Please wait.

Presentation is loading. Please wait.

MIMO Support for Interference Mitigation

Similar presentations


Presentation on theme: "MIMO Support for Interference Mitigation"— Presentation transcript:

1 MIMO Support for Interference Mitigation
IEEE Presentation Submission Template (Rev. 9) Document Number: IEEE S80216m-08_587 Date Submitted: Source: Kiran Kuchi, J Klutto Milleth CEWiT India Venue: Denver, USA Purpose: To facilitate the choice of uplink MIMO and to support interference mitigation techniques during the IEEE meeting in Denver, USA Notice: This document does not represent the agreed views of the IEEE 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 Patent Policy: The contributor is familiar with the IEEE-SA Patent Policy and Procedures: < and < Further information is located at < and < >.

2 MIMO Support for Interference Mitigation
Kiran Kuchi, Klutto Milleth, Vinod Ramaswamy, K. Giridhar, Bhaskar Ramamurthi CEWiT, India 5/2/2019

3 Outline India Specific Requirements
MIMO Support for Interference Mitigation Proposed MIMO Schemes OL 2D-POD for 2 and 4 Tx antennas OL STBC+2D-POD for 4 Tx antennas OL SM+2D-POD for 4 Tx antennas Pilot Support Simulation results 5/2/2019

4 India Specific Requirements
Limited Spectrum Frequency reuse-1 is the most likely deployment mode Typical urban cell size has m radius Interference limited in both uplink and downlink High Spectrum Efficiency Low Cost and Ease of Deployment Support for Heavy Voice Traffic Efficiency comparable to GSM 5/2/2019

5 Interference Suppression
Poor cell edge SINR in reuse-1 networks Severe co-channel interference (CCI) Multiple CCI up to 3-dominant interferers Conventional interference suppression techniques 2-antenna MMSE receiver can suppress a single dominant CCI Jointly detect signal dominant co-channel interference Reuse existing 16e MIMO channel estimation and MLD joint detection algorithms Most chipset vendors have MIMO MLD type receiver implementation Add an additional noise whitening filter for additional suppression Proposed algorithms can fully reject at least two dominant CCI 5/2/2019

6 MIMO Support for Interference Mitigation
Conventional MIMO techniques both OL and CL schemes are optimized for diversity gain in the presence of AWGN Diversity benefit should not be the sole consideration for cell edge users who operate in very low SINR conditions Better MIMO support needed for efficient interference suppression Choose schemes which maximize both diversity advantage and interference suppression gain Phase Offset Diversity (POD) and/or single stream CL-MIMO fulfils the above stated these requirements 5/2/2019

7 Proposed MIMO Schemes 5/2/2019

8 2D-Phase Offset Diversity (2D-POD)
In IEEE m data is allocated in groups of resource blocks/slots. For example, a slot is defined as 18 subcarriers and 6-OFDM symbols The following matrix-A with 2D-POD is proposed. Data in each slot is multiplied with a distinct phase rotation and transmitted through second antenna Time dependent phase offset Frequency dependent phase offset Slot wise phase rotations are optimized to maximize the diversity advantage The phase rotations maximally de-correlates channel 2D-POD outperforms both CDD and STBC/SFBC The matrix A can be generalized to arbitrary number of transmit antennas is the slot index in Time –Frequency Plane 5/2/2019

9 2D-Phase Offset Diversity
Advantages Slot by slot channel estimation in virtual (single) antenna mode Half pilot density compared to STBC/SFBC or improved channel estimation with equal pilot density as STBC/SFBC Suitable for irregular sub-frames Uplink resources will be allocated more in time and narrow in frequency for power efficiency reasons. 2D-POD is well suited in this situation. Compatible with advanced interference cancellation (IC) algorithms Receiver can fully reject two dominant CCI Significant gain in cell edge performance STBC/SFBCs are not compatible with conventional IC algorithms Diversity should not be the sole consideration for cell-edge users 2D-POD provides a good trade-off between diversity and IC gain In both uplink and downlink 2D-POD for cell edge users in an FFR zone STBC/SFBC based schemes for high SINR users in another zone On uplink 2D-POD is useful even in high SINR case (les pilots) 5/2/2019

10 2D-POD Vs CDD 2D-POD uses different phase values in each RB
The phase values can be optimized depending on the resource allocation Receiver does not have to know about the phase rotations used by the BS CDD has two modes Large Vs small delay CDD uses phase rotations in frequency dimension only. This scheme can be viewed as a 1D-POD scheme Higher signaling complexity since the BS needs to signal the delay value to the MS The receiver optimizes the channel estimation algorithm based on the delay value Higher receiver complexity 2D-POD has dB gain over CDD/1D-POD 2D-POD outperforms CDD in both complexity and performance 5/2/2019

11 4-antenna SFBC+2D POD For 4-antenna single stream OL case
SFBC+2D-POD matrix A Antenna permutation in time/frequency compared to existing 16e matrix A Same or better performance Half the pilot overhead Low complexity channel estimation and symbol decoding 5/2/2019

12 4-Antenna SM+2D POD For 4-antenna 2-stream OL case SM+2D-POD matrix B
Antenna permutation over time/frequency Six possible matrices. A subset can be used in different slots Compared to existing 16e matrix B Better performance Half the pilot overhead Low complexity channel estimation and low complexity decoding SM MLD receiver can be reused 5/2/2019

13 MIMO Support for Interference Mitigation
Interference suppression feasibility (two dominant interferers) with OL/CL MIMO Interference suppression not feasible when CL and STBC OL users cause mutual interference Users can switch between 2D –POD and single stream CL-MIMO with full IC support MIMO Mode used by the Desired Signal MIMO Mode used by CCI Whitened MLD Receiver Support MMSE Receiver Support CL/POD Yes STBC No Yes, but inferior performance 5/2/2019

14 Pilot Support In a localized allocation, 2D-POD and CL MIMO use dedicated pilots Users can switch between 2D-POD and single stream CL-MIMO without degrading IC receivers In distributed allocation, several localized resource blocks (RBs) will be distributed in frequency domain 2D-POD uses common pilots We prefer quasi-orthogonal pilots (i.e., orthogonal pilots with a PN cover) which facilitate low-complexity multi-user channel estimation Pilots collide with pilots 5/2/2019

15 Simulation Results 5/2/2019

16 Simulation Assumptions
QPSK, CTC 1/2, FEC BLOCK SIZE = 384 bits or 280 bits Localized sub-channelization or distribution of several localized blocks (Block distribution) PED A and PED B Channels, 3kmph 2D -MMSE Channel estimator (within a resource block)‏ 5.56 or 11.12% total pilot density Performance is reported in the presence of AWGN 5/2/2019

17 Performance in the presence of white noise
5/2/2019

18 (2 x 2 )‏ 2D-POD Vs 1D-POD with distributed allocation
Notation 2F-2T denotes a total of 4 RBs. Pairs of RBs are contiguous in time but these time pairs are distributed in frequency PED-A channel 2D-POD outperforms 1D-POD by more than 2.0 dB at high SNR 5/2/2019

19 (2 x 2 )‏ 2D-POD Vs 1D-POD in localized allocation
Notation 2F-2T denotes a total of 4 RBs. The four RBs are arranged in a rectangular grid PED-A channel 2D-POD outperforms 1D-POD by 1.5 dB dB at high SNR 5/2/2019

20 Performance comparison of STBC and POD (2 x 2 )‏
Receiver Ant correlation 0.5 POD is 0.3 dB worse than STBC with half the pilot overhead 2D-POD and STBC have similar performance when both schemes have same pilot overhead 5/2/2019

21 STBC+2D POD Vs STBC+1D-POD in distributed case
Two Pairs of RBs are distributed in frequency Receiver Antenna correlation is 0.9 2D-POD outperforms 1D-POD by dB at high SNR 5/2/2019

22 Performance of STBC+2D POD with ideal channel and estimated channel
Uncorrelated antennas at receiver Channel estimation loss is less than 1.0 dB 5/2/2019

23 Results with 2-Tx antennas in the presence of CCI
5/2/2019

24 Link performance of 2D-POD in the presence of CCI
Assumptions QPSK, CTC 1/2, FEC BLOCK SIZE = 384 bits Localized sub-channelization PED A Channel, 3kmph 2D -MMSE Channel estimator (within a resource block)‏ 11.12% total pilot density A combination of orthogonal and non-orthogonal pilots used for improved channel estimation 2 interferes with SIR profile [0db 3dB] The rest of the interfere power is modeled as white noise Performance is plotted as a function of signal to rest of the interfere power, which we denote as SNR Receivers – MMSE, Whitened MLD 5/2/2019

25 2D-POD in Interference-limited scenario (2 x 2)‏
Transmitter uses 2D-POD but receiver works in virtual single antenna mode Whitened MLD is able to fully suppress two dominant CCI Conventional MMSE shows an error floor since it can suppress a single interferer only Full interference rejection gain in the presence of highly correlated antennas STBCs can not support MLD Therefore, POD should be the baseline transmit diversity scheme for cell edge users 5/2/2019

26 STBC Vs 2D-POD in Interference-limited scenario (2 x 2)‏
2D-POD outperforms STBC with MMSE receiver STBC does not support MLD 5/2/2019

27 Conclusion 2D-POD outperforms 1D-POD/CDD
Both in localized and distributed allocations 2-TX OL 2D-POD and STBC have comparable performance in the presence of white noise 2D-POD outperforms STBC in the presence of CCI 2D-POD supports advanced IC receivers MMSE pre-whitened MLD STBCs do not support MLD We propose basic 2D-POD as the baseline OL transmit diversity for 2 and 4-Tx antennas for cell edge users with interference mitigation support STBCs for high SINR users For 4-TX single stream OL, we propose STBC+2D-POD which outperforms STBC+CDD For 4-TX OL SM we propose SM+2D-POD, which outperforms SM+CDD 5/2/2019


Download ppt "MIMO Support for Interference Mitigation"

Similar presentations


Ads by Google