Bringing the Magic of Asymptotic Analysis to Wireless Networks

Transcription

1 Massive MIMO Bringing the Magic of Asymptotic Analysis to Wireless Networks Dr. Emil Björnson Department of Electrical Engineering (ISY) Linköping University, Linköping, Sweden International Workshop on Computer-Aided Modeling Analysis and Design of Communication Links and Networks (CAMAD) December 1, 2014

2 Biography 1983: Born in Malmö, Sweden 2007: Master of Science in Engineering Mathematics, Lund, Sweden 2011: PhD in Telecommunications, KTH, Stockholm, Sweden : International Postdoc Grant Worked at Supélec, Paris, France 2014: Assistant Professor Division of Communication Systems, Electrical Engineering, Linköping University, Sweden Optimal Resource Allocation in Coordinated Multi-Cell Systems Book by Emil Björnson, Eduard Jorswieck FnT in Communications and Information Theory 2

3 Outline Introduction Expectation on future cellular networks What is Massive MIMO? Basic Motivation: Asymptotic Properties Theoretical results meet measurements Bringing the Magic of Asymptotic Analysis to Wireless Networks Anticipated implementation and performance gains Recent Results: More Magic from Asymptotic Analysis Hardware design, coordination, hardware utilization all get easier/better 3

4 Introduction WHAT CAN THE PAST TELL US ABOUT THE FUTURE? 4

5 Incredible Success of Wireless Communications Last 45 years: 1 Million Increase in Wireless Traffic Martin Cooper s law The number of simultaneous voice/data connections has doubled every 2.5 years (+32% per year) since the beginning of wireless Source: Personal Communications in 2025, Martin Cooper Martin Cooper Inventor of handheld cellular phones 5 Source: Wikipedia

6 Predictions for the Future Wireless Connectivity A natural part of our lives Voice calls Video streaming Gaming Social networks Next killer app? Rapid Network Traffic Growth 61% annual data traffic growth Faster than in the past! Exponential increase Extrapolation: 20x until x until x until GB/person/month 210 MB/month/person 6

7 Evolving Networks for Higher Traffic Increase Network Throughput [bit/s] Consider a given area Simple Formula for Network Throughput: Throughput bit/s in area = Available spectrum in Hz Cell density Cell/Area Spectral efficiency bit/s/hz/cell Ways to achieve 1000x improvement: More spectrum Higher cell density Higher spectral efficiency Nokia (2011) 10x 10x 10x SK Telecom (2012) 3x 56x 6x 7 New regulations, cognitive radio, higher frequencies Smaller cells, heterogeneous deployments Massive MIMO (Topic of this talk)?x

8 Introduction to MASSIVE MIMO 8

9 Higher Spectral Efficiency Spectral Efficiency of Point-to-Point Transmission Governed by Shannon s capacity limit: log Received Signal Power Interference Power + Noise Power [bit/s/hz/user] Cannot do much: 4 bit/s/hz 8 bit/s/hz costs 17 times more power! Many Parallel Transmissions: Spatially focused to each desired user 9

10 Multi-User MIMO (Multiple-input Multiple-output) Multi-Cell Multi-User MIMO Base stations (BSs) with N antennas Parallel uplink/downlink for K users Channel coherence block: T symbols Theory: Hardware is Limiting Spectral efficiency roughly prop. to min N, K, T 2 2x improvement = 2x antennas and users (since T [100,10000]) Practice: Interference is Limiting Multi-user MIMO in LTE-A: Up to 8 antennas Small gains since: Hard to learn users channels Hard to coordinate BSs End of the MIMO road? No reason to add more antennas/users? 10

11 Taking Multi-User MIMO to a New Level Network Architecture: Massive MIMO Use large arrays at BSs; e.g., N 200 antennas, K 40 users Key: Excessive number of antennas, N K Very narrow beamforming Little interference leakage Spectral efficiency prop. to number of users! min N, K, T 2 K IEEE Marconi Prize Paper Award Thomas Marzetta, Noncooperative Cellular Wireless with Unlimited Numbers of Base Station Antennas, IEEE Trans. Wireless Communications, Analysis based on asymptotics: N Concept applicable at any N

12 What is the Key Difference from Today? Number of Antennas? No, we already have many antennas! 3G/UMTS: 3 sectors x 20 element-arrays = 60 antennas 4G/LTE-A: 4-MIMO x 60 = 240 antennas Massive MIMO Characteristics Active antennas: Many antenna ports Coherent beamforming to tens of users Typical vertical array: 10 antennas x 2 polarizations Only 1-2 antenna ports antenna elements, LuMaMi testbed, Lund University 3 sectors, 4 vertical arrays per sector Image source: gigaom.com

13 Massive MIMO Deployment When to Deploy Massive MIMO? The future will tell, but it can 1. Improve wide-area coverage 2. Handle super-dense scenarios Co-located Deployment 1D, 2D, or 3D arrays One or multiple sectors Distributed Deployment Remote radio heads Cloud RAN 13

14 Basic Motivation ASYMPTOTIC PROPERTIES 14

15 Asymptotic Channel Orthogonality Example: Uplink with Isotropic/Rayleigh Fading Two users, i.i.d. channels: h 1, h 2 ~ CC(0, I N ) Signals: s 1, s 2 with power P Noise: n ~ CC(0, I N ) Received: y = h 1 s 1 + h 2 s 2 + n h 1 h 2 Linear Processing for User 1: y 1 = w 1 H y = w 1 H h 1 s 1 + w 1 H h 2 s 2 + w 1 H n Matched filter: w 1 = 1 N h 1 Signal remains: w 1 H h 1 = 1 N h 1 2 N E h 11 2 = 1 Interference vanishes: w H 1 h 2 = 1 h N 1 H h N 2 E[h H 11 h 21 ] = 0 Noise vanishes: w H 1 n = 1 h N 1 H n N E[h H 11 n 1 ] = 0 Asymptotically noise/interference-free communication: y 1 N s1 15

16 Is this Result Limited to Isotropic Fading? Assumptions in i.i.d. Rayleigh Fading No dominant directivity Very many scattering objectives Less true as N Example: Line-of-Sight Propagation Uniform linear array Random user angles N observations: Stronger signal Suppressed noise What is h H 1 h 2 =? h 1 H h 2 2 h 1 2 h N h 1 H h Main difference: How quickly interference is suppressed

17 Does This Hold for Practical Channels? Initial Measurements: Show similar results Source: X. Gao, O. Edfors, F. Rusek, and F. Tufvesson, Linear Pre-Coding Performance in Measured Very-Large MIMO Channels, VTC h 1 H h 2 2 h 1 2 h N h 1 H h 2 2 Spectral Efficiency Only 10-20% lower than with isotropic fading h 1 H h 2 h 1 h 2 1 N h 1 H h 2 17 Source: J. Hoydis, C. Hoek, T. Wild, and S. ten Brink, Channel Measurements for Large Antenna Arrays, ISWCS 2012

18 Favorable Propagation Favorable Propagation Definition: h H 1 h 2 = 0 for any users No interference leakage Everyone gets the whole cake Asymptotic Favorable Propagation Definition: 1 h N 1 H h 2 0 as N Achieved in Rayleigh fading and line-of-sight two extremes! Same behavior expected and seen in practice There are no experimentally verified massive MIMO channel models 18 Computer-Aided Design Requires Models With Line-of-sight and spherical wavefronts Scattering less rich : Other random distributions Spatial correlation: Directivity and user-correlation Affect if 1 h N 1 H h 2 0 and convergence in mean and variance

19 BRINGING THE MAGIC OF ASYMPTOTIC ANALYSIS TO WIRELESS NETWORKS 19

20 Channel Acquisition Asymptotic Favorable Propagation N : Interference vanishes using linear matched filtering Finite N: Reject remaining interference with zero-forcing filtering Requires channel state information (CSI) at BSs Pilot-based CSI Estimation Frequency-division duplex (FDD): Two-way estimation and feedback Time-division duplex (TDD): Only one-way estimation (exploits reciprocity) Natural choice: TDD Much less overhead! 20 Estimation overhead O(N) Estimation overhead O(K) Example parameters: N = 100, K = 20, T = 200

21 This image cannot currently be displayed. Massive MIMO Transmission Protocol Basic TDD Frame Structure Dimensioned to have fixed channels Coherence time: T c s Coherence bandwidth: B c Hz Depends on mobility and environment Block length: T = T c B c symbols Typically: T [100,10000] Processing Tasks per Frame 21 Channel estimation Linear processing of data signals Fourier transforms (in OFDM) Computing processing filter Complexity: Increase linearly with N and K Straightforward to parallelize Scales as NK with matched filter NK 2 with zero-forcing filter

22 Limited Pilot Resources Limiting Factor: Coherence Block T Not more than T orthogonal pilots CSI errors: Act as noise and are suppressed Multi-cell: Must reuse pilots across cells Pilot Contamination BS cannot tell difference between users Channel estimates are correlated This interference doesn t vanish as N Will Pilot Contamination Kill Massive MIMO? No, but treat it with much respect! Make: Target SINR Pathloss2 from transmitter Pathloss 2 from interferers 22

23 Limited Pilot Resources (2) Solution: Smart Pilot Allocation Cannot be based on current CSI Simple: Pilot reuse factor β Advanced: Exploit spatial statistics Pilot reuse: β = 1 Different statistical angles of arrival Pilot reuse: β = 3 23 Design Tradeoff High β: Less contamination Higher user performance Low β: More users/cell, up to T, can give higher area performance 2β Pilot reuse: β = 4

24 Anticipated Area Spectral Efficiency Simulation LTE-like system parameters Coherence block: T = 500 SNR 5 db, i.i.d. Rayleigh fading 24 Observations Baseline: 3 bit/s/hz/cell (IMT-Advanced) Massive MIMO, N = 100: x20 gain Massive MIMO, N = 200: x30 gain Per scheduled user: 2 bit/s/hz

25 Recent Results MORE MAGIC FROM ASYMPTOTIC ANALYSIS 25

26 Capitalizing on Excessive Degrees of Freedom Benefits from Coherent Transmit/Receive Filtering Desired signals amplified by a factor N Undesired signals are not amplified Noise and hardware distortions are not amplified Ways to Sacrifice Degrees of Freedom Cancel interference at undesired receivers Cancel interference in spatial areas Simplifying hardware design 26

27 Distortions from Hardware Impairments Real Hardware is Non-Ideal Hardware impairments: Phase noise, I/Q-imbalance, non-linearities, etc. Impact reduced by calibration/compensation (not fully removed!) Simple Hardware Model: OFDM signal Bussgang s Theorem: Power loss (can be tracked) Additive distortion noise Key Property: Distortion noise is suppressed just as other noise Tolerate larger impairments Cheap and energy-efficient hardware 27

28 Energy Efficient Hardware Utilization Well-designed Systems are Energy Efficient [bit/joule] EE = Average Sum Rate bit/s Transmit Power + Circuit Power Joule/s What if we optimize over N, K, and rate per user? Use state-of-the-art models with recent parameters Key Property: Massive MIMO gives high energy efficiency Efficient Hardware Utilization Large array gain Many simultaneous users 28 Transmit Power Total: Slightly lower than today Per antenna: Much smaller

29 Cancel Interference to Other Systems Simple Coordination Protocol Estimate received interference subspace Transmit orthogonal to the M dominating interferers Fully Distributed Use only local information No feedback or exchange Scalable! Not Massive MIMO Massive MIMO Key Property: Distributed Coordination N large: Subspace dimension M N Small signal loss, much less interference Helps coexistence with legacy systems 29

30 Hardware-Friendly Signal Shaping Channel Downlink Transmission Noise Received signal: y = h H w + n Effective Signal s How to pick w so that s = h H w? Matched filter: w h s h Pro: Minimize transmit power w 2 Con: x10 power variations over antennas/time, inefficient amplifiers Key Property: Constant Envelope Precoding Elements in w: w 1 2 = = w N 2 for all s Pro: No power variations, much simpler amplifier design Con: More radiated power, same total power when N is large 30

31 31 SUMMARY

32 Summary Massive MIMO: A technique to increase spectral efficiency Massive multi-user MIMO: >20x gain over IMT-Advanced are foreseen High spectral efficiency per cell, not per user Many potential deployment strategies and propagation environments Many Base Station Antennas Near favorable propagation Resilience to hardware impairments High energy efficiency Distributed coordination with other systems Hardware-friendly signal shaping Magic brought by asymptotic analysis 32 Important: Channel coherence block Limits multiplexing gain and per-user performance Pilot contamination mitigated by smart pilot allocation

33 Bringing the Magic to Reality FP7 MAMMOET project (Massive MIMO for Efficient Transmission) Bridge gap between theoretical and conceptual massive MIMO Develop: Flexible, effective and efficient solutions 33

34 QUESTIONS? Dr. Emil Björnson Visit me online:

35 Main References T. Marzetta, Noncooperative Cellular Wireless with Unlimited Numbers of Base Station Antennas, IEEE Trans. Wireless Communications, H. Huh, G. Caire, H. C. Papadopoulos, and S. A. Ramprashad, Achieving Massive MIMO Spectral Efficiency with a Not-so-Large Number of Antennas, IEEE Trans. Wireless Communications, S.K. Mohammed, Erik G. Larsson, Per-Antenna Constant Envelope Precoding for Large Multi- User MIMO Systems, IEEE Transactions on Communications, J. Hoydis, K. Hosseini, S. ten Brink, and M. Debbah, Making Smart Use of Excess Antennas: Massive MIMO, Small Cells, and TDD, Bell Labs Technical Journal, F. Rusek, D. Persson, B. K. Lau, E. G. Larsson, T. L. Marzetta, O. Edfors, and F. Tufvesson, Scaling up MIMO: Opportunities and Challenges with Very Large Arrays, IEEE Signal Proces. Mag., vol. 30, no. 1, pp , E. Björnson, E. G. Larsson, M. Debbah, Optimizing Multi-Cell Massive MIMO for Spectral Efficiency: How Many Users Should Be Scheduled?, Proc. IEEE GlobalSIP, E.G. Larsson, F. Tufvesson, O. Edfors, and T. L. Marzetta, Massive MIMO for Next Generation Wireless Systems, IEEE Commun. Mag., vol. 52, no. 2, pp , E. Björnson, J. Hoydis, M. Kountouris, M. Debbah, Massive MIMO Systems with Non-Ideal Hardware: Energy Efficiency, Estimation, and Capacity Limits, To appear in IEEE Trans. Information Theory. 9. E. Björnson, L. Sanguinetti, J. Hoydis, M. Debbah, Optimal Design of Energy-Efficient Multi- User MIMO Systems: Is Massive MIMO the Answer?, Submitted to IEEE Trans. Wireless Communications.