MIMO MIA or the different faces of MIMO

MIMO MIA! …or the different faces of MIMO! Concepts of 3 GPP LTE 9 Oct 2007 Page 1 1 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Agenda • Just a little MIMO theory • MIMO in the LTE air interface • LTE MIMO conformance testing • Testing MIMO in the real world Concepts of 3 GPP LTE 9 Oct 2007 Page 2 2 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Agilent LTE Book www. agilent. com/find/ltebook www. amazon. com In print April 16 th The first LTE book dedicated to design and measurement 30 Authors 460 pages Concepts of 3 GPP LTE 9 Oct 2007 Page 3 3 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Book overview Chapter 1 LTE Introduction Chapter 2 Air Interface Concepts Chapter 3 Physical Layer Chapter 4 Upper Layer Signaling Chapter 5 System Architecture Evolution Chapter 6 Design and Verification Challenges Chapter 7 Conformance Test Chapter 8 Looking Towards 4 G: LTE-Advanced Concepts of 3 GPP LTE 9 Oct 2007 Page 4 4 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Agenda • Just a little MIMO theory • MIMO in the LTE air interface • LTE MIMO conformance testing • Testing MIMO in the real world Concepts of 3 GPP LTE 9 Oct 2007 Page 5 5 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Basic channel access modes The Radio Channel Transmit Antennas Receive Antennas SISO Transmit Antennas The Radio Channel Receive Antennas SIMO Single Input Single Output Single Input Multiple Output (Receive diversity) MISO Multiple Input Single Output (Transmit diversity) Concepts of 3 GPP LTE 9 Oct 2007 Page 6 6 Page MIMO Multiple Input Multiple Output (Multiple data streams) Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

MIMO principles • Transmitting multiple data streams in the same space and time used to be called interference! • So how does MIMO work? 1. MIMO capacity gains come from taking advantage of spatial diversity in the radio channel 2. Depending on channel conditions and noise levels, the rank (number of simultaneous streams) can be varied 3. The performance can be optimized using precoding • These three MIMO principles can seem complex to understand particularly abstract mathematical descriptions • But we intuitively already know these MIMO principles in the way they apply to our perception of audio Concepts of 3 GPP LTE 9 Oct 2007 Page 7 7 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Understanding MIMO spatial diversity through Audio - Single Stream (Mono) M M SISO SIMO M M MISO M M SIMO + MISO ≠ MIMO Note, the combination of SIMO and MISO further improves robustness but does not provide any MIMO capacity gain since there is only one stream of data Concepts of 3 GPP LTE 9 Oct 2007 Page 8 8 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Understanding MIMO spatial diversity through Audio - Dual Stream (Stereo) R L Interference! L R Interference! R L MIMO! For MIMO to work: • • • Must have at least as many receivers as transmitted streams Must have spatial separation at both transmit and receive antennas More transmitters enables beamforming in addition to MIMO Concepts of 3 GPP LTE 9 Oct 2007 Page 9 9 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Understanding MIMO precoding through Audio • MIMO Precoding is a pre-emphasis technique used to improve the separation of the streams at the receiver due to unhelpful coupling in the channel • In audio systems precoding is similar to stereo “balance” • If the receiver is not positioned directly between the speakers the received streams will be at different levels • Adjusting the balance at the transmitter can mitigate the problem • Balancing requires feedback from the receiver to the transmitter Concepts of 3 GPP LTE 9 Oct 2007 Page 1010 Page L R Not enough R Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Understanding MIMO precoding through Audio • The receiver could just amplify the right channel but in the presence of noise the corrected signal would degrade: • Precoding the transmission as L, 5 R optimizes signal recovery L R L 5 R Solution! L + NL, 0. 2 R + NR L + NL, R + 5*NR Concepts of 3 GPP LTE 9 Oct 2007 Page 1111 Page L + N L, R + N R Problem! Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Understanding MIMO Rank adaptation through Audio • In good radio conditions an FM stereo receiver will attempt to decode both the left and right signals (streams) • When the noise gets too high the receiver switches to mono and the quality improves although stereo is lost • This is the audio equivalent of rank adaptation where the number of streams is reduced under poor conditions • Transmit matrix encoded FM stereo as L + R, L – R • Receive (L + R) + N 1, (L – R) + N 2 • Since N 1 and N 2 are largely correlated, adding the two streams (maximum ratio combining) cancels most of the noise Concepts of 3 GPP LTE 9 Oct 2007 Page 1212 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

The role of channel correlation and noise in system performance • In a ideal 2 x 2 system the potential capacity gain is 2 x • The actual gain depends on how easily the receiver can descramble the simultaneous transmissions – this depends on the amount of unwanted correlation and noise • In audio systems channel correlation and noise also affects perceived stereo performance – Spaced living room speakers - lots of correlation degrades stereo, susceptible to external noise – Open headphones – zero correlation, good stereo but still susceptible to noise – Closed headphones – zero correlation, minimal noise Concepts of 3 GPP LTE 9 Oct 2007 Page 1313 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

So what makes a good channel for MIMO? • A perfect MIMO channel is like the closed headphones: channels 2 and 3 don’t exist T 0 ch 1 ch 2 T 1 3 ch ch 4 Channel H R 0 R 1 1 0 0 1 • By simple observation it follows that R 0 = T 0 and R 1 = T 1 • This is the case that creates double the capacity • But suppose we create a simple static channel like this: • How do we know if it will provide capacity gain? Concepts of 3 GPP LTE 9 Oct 2007 Page 1414 Page Channel H 0. 8 0. 2 0. 3 -0. 9 Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

The MIMO challenge: Recovering the signal • If all four channels are the same the original signal cannot be recovered since R 0 = R 1 T 0 ch 1 ch 2 T 1 3 ch ch 4 Channel H R 0 1 1 R 1 1 1 R 0 = T 0 + T 1 and R 1 = T 0 + T 1 • But put in a phase inversion e. g. on ch 3 we get: R 0 = T 0 + T 1 and R 1 = T 1 – T 0 thus T 0 = (R 0 - R 1)/2 and T 1 = (R 0 + R 1)/2 Channel H 1 1 -1 1 • The original signal is completely recovered even though the apparently unwanted ch 2 and ch 3 exist Concepts of 3 GPP LTE 9 Oct 2007 Page 1515 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

The MIMO challenge: Recovering the signal • So is the earlier example good or bad for MIMO? R 0 = 0. 8 T 0 + 0. 3 T 1 R 1 = 0. 2 T 0 - 0. 9 T 1 Giving: Channel H 0. 8 0. 2 0. 3 -0. 9 T 0 = 1. 15 R 0 + 0. 39 R 1 T 1 = 0. 26 R 0 - 1. 03 R 1 • We can recover the original signal • In fact any H matrix other than the unity matrix can be resolved PROVIDED there is no external or internal noise! • So what kinds of channels are robust to noise? Concepts of 3 GPP LTE 9 Oct 2007 Page 1616 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

The MIMO challenge: Recovering the signal • The receiver can untangle the two signals because it knows the coupling coefficients, based on the reference signals, but reference estimation is susceptible to noise • But pilot estimation is susceptible to noise • If the estimate is wrong the recovered signal is impaired • Consider these equations for T 0 from different channels: T 0 = 1. 15 (R 0 + N 0) + 0. 39 (R 1 + N 1) T 0 = 27. 3 (R 0 + N 0) + 16. 5 (R 1 + N 1) • Errors in T 0 recovery happen due to estimation errors in the coefficients or large coefficients amplifying noise N 0 and N 1 • It is possible to analyze the matrix H to predict the impact of noise on signal recovery Concepts of 3 GPP LTE 9 Oct 2007 Page 1717 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Condition Number Measures the short term MIMO channel performance R 0 = 0. 8 T 0 + 0. 3 T 1 R 1 = -0. 9 T 1 + 0. 2 T 0 К = Condition number 0. 957 / 0. 815 = 1. 17 The condition number is the ratio of the singular values of HHT Channel HT 0. 8 0. 2 0. 8 0. 3 -0. 9 0. 2 -0. 9 Eigenvalues Singular values 0. 73 -0. 11 0. 914 0. 957 -0. 11 0. 85 0. 666 0. 815 Channel HTH The d. B value of К approximates the increase in SNR required to recover the signal Concepts of 3 GPP LTE 9 Oct 2007 Page 1818 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

MIMO needs better SNR than SISO High К increases SNR requirements further The extra SNR required to achieve the same recovered signal quality as SISO rises as the condition number rises Concepts of 3 GPP LTE 9 Oct 2007 Page 1919 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Ped. A Channel Condition Number vs. Freq. Condition number and channel response across 10 MHz, 10 ms 0 d. B Concepts of 3 GPP LTE 9 Oct 2007 Page 2020 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Impact of condition number frequency dependency • The previous example of how the condition number varies across the channel during one 10 ms frame and how the pattern varies a few frames later depending on speed • This variability is both a challenge and an opportunity • OFDMA systems can transmit at different frequencies within the channel to target that part of the channel offering the best MIMO gains • CDMA systems cannot do this and have to accept the average performance across the channel Concepts of 3 GPP LTE 9 Oct 2007 Page 2121 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Antenna influence on performance • The dynamic condition number example did not isolate effects from different components, including the antenna • In real life, the instantaneous channel matrix H is made up from the interaction of three components: • The static 3 D antenna pattern of the transmitter • The dynamic multipath and Doppler characteristics of the radio channel • The static 3 D antenna pattern of the receiver • The overall antenna contribution is the product of the transmit and receive antennas known as the channel correlation matrix Concepts of 3 GPP LTE 9 Oct 2007 Page 2222 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Antenna correlation • The correlation between antennas is a primarily a function of distance and polarization • For non polarized antennas the correlation decreases with larger separation in the y axis - usually expressed in terms of wavelength λ Concepts of 3 GPP LTE 9 Oct 2007 Page 2323 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Examples of low and high antenna correlation Spaced non polarized: High correlation Concepts of 3 GPP LTE 9 Oct 2007 Page 2424 Page Compound spaced and cross polarized: Low correlation Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Antenna correlation by type AS = Azimuth Spread Concepts of 3 GPP LTE 9 Oct 2007 Page 2525 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Generating the overall channel correlation matrix Transmit antenna correlation Receive antenna correlation The α and β terms are complex and will vary by frequency The correlation matrix Rs is the Kronecker product RBS RMS Concepts of 3 GPP LTE 9 Oct 2007 Page 2626 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Example channel correlation matrices Cross polarized, UE (0, 90) BS (-45, 45), -8 d. B XPR ratio Channels balanced Diagonal = 1 Cross polarized, UE (-10, 80) BS (-30, 60), -8 d. B XPR ratio Channels unbalanced Not ideal Concepts of 3 GPP LTE 9 Oct 2007 Page 2727 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Computing the instantaneous channel The complex instantaneous channel coefficients are obtained by applying each path of the desired fading profile to each channel of the correlation matrix Ch 1 T 0 ch 1 ch 2 T 1 3 ch ch 4 R 0 R 1 Ch 2 Ch 3 Ch 4 Ch 1 Ch 2 Ch 3 Ch 4 The received signals and condition number are dynamic in both the time and frequency domains according to the chosen fading profile Concepts of 3 GPP LTE 9 Oct 2007 Page 2828 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Real life performance Variation due to instantaneous correlation Variation in the frequency domain not shown Most macrocell activity takes place in this region Variation due to fading and variable interference Concepts of 3 GPP LTE 9 Oct 2007 Page 2929 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Precoding example for condition number 20 d. B Second stream is noise limited No precoding Channel quality is unbalanced Precoded with 1, 1, -1, 1 for equal EVM Concepts of 3 GPP LTE 9 Oct 2007 Page 3030 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Agenda • Just a little MIMO theory • MIMO in the LTE air interface • LTE MIMO conformance testing • Testing MIMO in the real world Concepts of 3 GPP LTE 9 Oct 2007 Page 3131 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

LTE downlink transmission modes 3 GPP TS 36. 213 subclause 7. 1 LTE has seven different downlink transmission modes: 1. 2. 3. 4. 5. 6. 7. Single-antenna port; port 0 Transmit diversity Open-loop spatial multiplexing Closed-loop spatial multiplexing Multi-user MIMO Closed-loop Rank=1 precoding Single-antenna port; port 5 SISO MIMO – no precoding MIMO -separate UE MISO - beamsteering MISO – beamsteering Each mode is suited to different channel and noise conditions Concepts of 3 GPP LTE 9 Oct 2007 Page 3232 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

The LTE MIMO toolset • The LTE standard recognizes the complexity of the MIMO channel and has developed a very flexible air interface • OFDMA allows for frequency-selective scheduling with 180 k. Hz and 1 ms granularity (one resource block) • Comprehensive channel state information • Channel Quality Indicator (CQI) – • Precoding Matrix Indicator (PMI) – codebook based • Rank Indication (RI) • Subband reporting for CQI & PMI, RI is wideband only • Highly configurable reporting mechanisms to account for different scenarios • The UE can select what subbands to report on Concepts of 3 GPP LTE 9 Oct 2007 Page 3333 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

CQI definition 3 GPP TS 36. 213 Table 7. 2. 3 -1 For each CQI reporting period the UE is required to return the highest CQI index that would have resulted in an error probability of less than 10% for a single transport block transmitted using the reported modulation and code rate. Subband CQI reporting can be configured down to the resource block level Concepts of 3 GPP LTE 9 Oct 2007 Page 3434 Page CQI index 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 modulation QPSK QPSK 16 QAM 64 QAM 64 QAM code rate x 1024 out of range 78 120 193 308 449 602 378 490 616 466 567 666 772 873 948 efficiency 0. 1523 0. 2344 0. 3770 0. 6016 0. 8770 1. 1758 1. 4766 1. 9141 2. 4063 2. 7305 3. 3223 3. 9023 4. 5234 5. 1152 5. 5547 Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

PMI definition 3 GPP TS 36. 211 Table 6. 3. 4. 2. 3 -1 For single stream transmission the precoding produces beamsteering For the 4 layer case there are 16 entries Subband PMI reporting can be configured down to the resource block level Concepts of 3 GPP LTE 9 Oct 2007 Page 3535 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Agenda • Just a little MIMO theory • MIMO in the LTE air interface • LTE MIMO conformance testing • Testing MIMO in the real world Concepts of 3 GPP LTE 9 Oct 2007 Page 3636 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

LTE MIMO conformance testing • The performance requirements for LTE are based on a number of simplifications to real world operation • This often involves a modular approach of doing open loop testing of parts of the functionality rather than a more end-to -end approach • This is a bit like measuring engine performance and other components rather than going for a test drive or a real track • The modular approach is useful and separates the test equipment from the DUT but does not tell the whole story • The consequence is that conformance test results cannot be easily mapped to real life conditions to predict typical performance Concepts of 3 GPP LTE 9 Oct 2007 Page 3737 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

MIMO conformance testing vs. real world Attribute Conformance testing Real world operation Correlation matrix High, medium and zero not linked to reference antenna design Real correlation based on actual antenna pattern Fading channel Extended PA, VA, TU Channels with dynamic taps Adaptive Modulation & coding Off – UE becomes fading channel discriminator On – coding aims for constant symbol to noise at UE receiver CQI, PMI, RI Separate open loop tests Included as part of throughput Cell-edge Interference signal Static wideband Gaussian Narrowband frequencyselective based on loading Live antenna testing Developing open loop Over The Air proposals Closed loop Real loading due to body/hand effects Scheduling None, Single UE Multiple UE, real scheduler with frequency selectivity base on subband CQI/PMI Transmission mode Fixed Variable based on prevailing conditions Concepts of 3 GPP LTE 9 Oct 2007 Page 3838 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Agenda • Just a little MIMO theory • MIMO in the LTE air interface • LTE MIMO conformance testing • Testing MIMO in the real world Concepts of 3 GPP LTE 9 Oct 2007 Page 3939 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Testing MIMO in the real world • Most of the simplifications in conformance testing can be overcome with alternative test methods to get closer to real world performance • We will now look at a few of the possibilities Concepts of 3 GPP LTE 9 Oct 2007 Page 4040 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

NEW N 5106 A PXB MIMO Receiver Tester • The flexibility of the PXB can be used to verify MIMO receiver performance throughout the design cycle, at baseband or RF Signal Inputs Signal Creation Tools Signal Outputs Analog I/Q Direct from PXB Connect to any DUT or RF vector signal generator with analog I/Q inputs Digital I/Q N 5102 A RF RF MXA PXB Concepts of 3 GPP LTE 9 Oct 2007 Page 4141 Page ESG or MXG Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

PXB creates real correlation based on reference antenna designs Rx antenna pattern, omni, 3 sector or 6 sector Rx antenna #1 location and polarization Rx antenna #2 location and polarization Concepts of 3 GPP LTE 9 Oct 2007 Page 4242 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Flexible Antenna Configuration and Correlation Path 1 Path 2 In this example, there are 6 paths, each with complex cross coupling coefficients Concepts of 3 GPP LTE 9 Oct 2007 Page 4343 Page Path 6 Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

PXB customizable fading simulation including dynamic channel taps Concepts of 3 GPP LTE 9 Oct 2007 Page 4444 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Flexible MIMO test using System. Vue • TD-LTE or LTE-FDD MIMO Baseband data is generated by System. Vue and sent to PXB • Flexible Fading applied by PXB • Two phase locked ESGs/MXGs driven by PXB generate Receiver test signals for the DUT • Two MXAs capture received signals from DUT output and send to System. Vue • System. Vue demodulates and decodes MIMO signals to measure receiver performance 2 x. N 9020 A Signal Analyzer 2 x. E 4438 C Signal Gen System. Vue N 5106 A PXB DUT Concepts of 3 GPP LTE 9 Oct 2007 Page 4545 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Modeling MIMO crosstalk in System. Vue Specify LO Phase Noise d. Bc/Hz @ Freq. Offset Specify 1 d. B Comp. Pt. Concepts of 3 GPP LTE 9 Oct 2007 Page 4646 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Measuring impact of crosstalk and phase noise on demodulated MIMO streams • These measurement were made using the MIMO features of the Agilent 89601 A Vector Signal Analyzer which fully integrates with the System. Vue design software -29 d. B Tx 0 / Rx 1 QPSK Concepts of 3 GPP LTE 9 Oct 2007 Page 4747 Page -29 d. B Tx 0 / Rx 1 64 QAM Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Upcoming TOL webcast • For further information on System. Vue: LTE MIMO System-Level Design and Test 5/27/2009 Greg Jue Concepts of 3 GPP LTE 9 Oct 2007 Page 4848 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Testing closed loop AMC with CQI, PMI and RI with integral fading • Open loop testing with no AMC avoids having to define the reference behaviour of the test equipment • However, it is still necessary to investigate closed loop • The E 6620 A wireless communications test set is designed to go beyond basic conformance to test closed loop MIMO up to 4 x 2 • Central to this is the inclusion of a baseband fading emulator E 6620 A Wireless Communications Test Set • This solution is the basis for development of scheduling algorithms and transmission mode selection criteria Concepts of 3 GPP LTE 9 Oct 2007 Page 4949 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Conclusion: Making MIMO work and testing it is tough! • The standards are very flexible and complex • The conformance tests are simple and largely open loop with corner case SNR and artificial correlation • Real life is way more complex • • • Real antennas Real channels Real schedulers with multiple UE per cell Dynamic configuration for CSI reporting Real TX/RX distortion impacting channel feedback Non Gaussian frequency-selective cell-edge interference But Agilent is here to help you clear the way for MIMO Concepts of 3 GPP LTE 9 Oct 2007 Page 5050 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

LTE Lifecycle LTE VSA SW Logic Analyzers Spectrum and signal Analyzers, Scopes, LA and ADS Signal Studio Battery Drain & Scopes Characterization Signal Generators ADS and System. Vue PXB MIMO Rx Tester RDX for Dig. RF v 4 RF Module Development RF Proto Design Simulation RF Chip/Module RF and BB Design Integration L 1/PHY BTS and Mobile BB Chipset Development L 1/PHY FPGA and ASIC N 9912 A RF Analyzer Spectrum Analyzers Dig. RF v 4 System Design Validation System Level RF Testing BTS or Mobile Pre. Conformance Manufacturing Protocol Development L 2/L 3 Agilent/Anite SAT Protocol Development Toolset E 6620 A Wireless Communications Platform Concepts of 3 GPP LTE 9 Oct 2007 Page 5151 Page Network Deployment Systems for RF and Protocol Conformance Drive Test Distributed Network Analyzers Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009

Finding MIMO: Don’t stop now! Learn more at www. agilent. com/find/MIMO and www. agilent. com/find/lte LTE Brochure (5989 -7817 EN) The 3 GPP MIMO song: ftp. 3 gpp. org/tsg_ran/WG 1_RL 1/TSGR 1_56/Docs/R 1 -091041. zip PXB Brochure (5989 -8970 EN) MIMO Poster (5989 -9618 EN) Webcasts on • LTE Concepts • LTE Uplink • LTE Design and Simulation • LTE signaling Concepts of 3 GPP LTE 9 Oct 2007 Page 5252 Page Taking LTE MIMO from Standards to Starbucks Moray Rumney 30 th April 2009