Transmitter Design and Measurement Challenges

Transcription

1 Transmitter Design and Measurement Challenges Based on the book: LTE and the Evolution to 4G Wireless Chapter 6.4 4G World 2009 presented by: David L. Barner www/agilent.com/find/4gworld 1 Agilent Technologies, Inc. 2009

2 Agenda Introduction / General & Overall Issues Power & Spectrum Characteristics Vector (Frequency & Time) Measurements Analysis of Signals After Digital Demodulation 2 Agilent Technologies, Inc. 2009

3 Agenda Introduction / General & Overall Issues Power & Spectrum Characteristics Vector (Frequency & Time) Measurements Analysis of Signals After Digital Demodulation 3 Agilent Technologies, Inc. 2009

4 General Challenges Variable & wide bandwidths From 1.4 up to 20 MHz A new TX scheme for UL (SC-FDMA) FDD and TDD modes Stressful signal characteristics in terms of spectrum, power, time variations due to traffic type and loading Multiple antenna techniques & MIMO The need for making complex tradeoffs between In-Channel, Out-of-Channel and Out-of-Band Performance 4 Agilent Technologies, Inc. 2009

5 A Systematic & Structured Approach When measuring complex signals, it is tempting to go directly to advanced digital modulation analysis. However, it is usually more productive and efficient to follow a verification sequence that begins with basic spectrum measurements and continues with vector measurements, before switching to modulation analysis. 5 Agilent Technologies, Inc. 2009

6 Agenda Introduction / General & Overall Issues Power & Spectrum Characteristics Vector (Frequency & Time) Measurements Analysis of Signals After Digital Demodulation 6 Agilent Technologies, Inc. 2009

7 Power & Spectrum Characteristics Familiar measurements apply for LTE: Channel power Occupied bandwidth ACLR SEM Center frequency, flatness Initial verification and measurement of these can be made fairly easily via Spectrum Analysis However, other power measurements can only be made via Vector Signal Analysis (demod) 7 Agilent Technologies, Inc. 2009

8 Power & Spectrum Characteristics LTE s wide bandwidth has interesting implications at allocated spectrum band-edges Many LTE carriers will need to be at the edge of a band / allocation This implies that many channels will be subject to Out-of- Channel and Out-of-Band emission regulations simultaneously Out-of-Channel emissions can be made with ACLR/ACP and SEM measurements 8 Agilent Technologies, Inc. 2009

9 Power & Spectrum Characteristics ACP Measurement 9 Agilent Technologies, Inc. 2009

10 Agenda Introduction / General & Overall Issues Power & Spectrum Characteristics Vector (Frequency & Time) Measurements Analysis of Signals After Digital Demodulation 10 Agilent Technologies, Inc. 2009

11 Power & Spectrum Characteristics Average power measurements are not new, even for timevarying signals LTE also requires accurate power measurements down to the resource element level (1 symbol x 1 subcarrier), and on a selective basis New power / demod measurements such as in-band emissions and OFDM Symbol TX Power introduced (next slide provides details from ) 11 Agilent Technologies, Inc. 2009

12 Power & Spectrum Characteristics F.3.3 Resource Element TX power Perform FFT (z (ν)) with the FFT window timing. The result is called Z (t,f). 2 The RE TX power is then defined as: RETP Z (t,f) 15KHz From this the Reference Signal Transmit power (RSTP) is derives as follows: RSTP 1 n RETP RS RE locations withinsubframe It is an average power and accumulates the powers of the reference symbols within a sub frame divided by n, the number of reference symbols within a sub frame. From RETP the OFDM Symbol TX power (OSTP) is derived as follows: OSTP RETP DL RB all NRB Nsc RE locations of 4thsymbol withinsubframe It accumulates all sub carrier powers of the 4th OFDM symbol. The 4 th,out of 14 OFDM symbols within a subframe, (using frame type 1, normal CP length) contains exclusively PDSCH. From the acquired samples, 10 values (per Frame) for each RSTP and OSTP can be derived. 12 Agilent Technologies, Inc. 2009

13 Vector (Frequency & Time) Measurements Power vs time measurements are of interest in LTE due to the varying structure of the signal (RS locations, channel locations, etc) Initially, these can be made without demodulating the signal, in order to verify absolute power levels using a VSA Wide BW LTE requires large Time Record Length (proportional to large number of FFT points) Example. 20MHz 1Frame for 10ms requires 256,000 pt FFT CCDF behavior can also be measured and in particular, using time-gating with a VSA NOT using time-gating significantly effects PAR during DTX (since lower average power) 13 Agilent Technologies, Inc. 2009

14 Vector (Frequency & Time) Measurements Time Gate Time Gating CCDF 14 Agilent Technologies, Inc. 2009

15 Vector (Frequency & Time) Measurements Spectrogram allows us to interpret the overall signal at a glance It allows us to visually recognize major signal characteristics, especially for complex signals such as DL Any serious power or frequency issues, for example of drift or symbol transitions, will be visible here 15 Agilent Technologies, Inc. 2009

16 Vector (Frequency & Time) Measurements RS PDSCH PDSCH Secondary synch signal 5 unallocated subcarriers on either side of synch signals Primary synch signal PDSCH PBCH PDSCH PDSCH Spectrogram of LTE DL 16 Agilent Technologies, Inc. 2009

17 Agenda Introduction / General & Overall Issues Power & Spectrum Characteristics Vector (Frequency & Time) Measurements Analysis of Signals After Digital Demodulation 17 Agilent Technologies, Inc. 2009

18 Analysis of Signals After Digital Demodulation First we ll focus on basic digital demodulation techniques Correct configuration in this step will help us properly verify basic parameters and also give confidence when investigating more intricate details of the signal 18 Agilent Technologies, Inc. 2009

19 Analysis of Signals After Digital Demodulation Measurement example for setup, including: FDD / TDD UL / DL Bandwidth & span Sync type Number of Antennas MIMO decoding 19 Agilent Technologies, Inc. 2009

20 Analysis of Signals After Digital Demodulation Measurement Example LTE DL 20 Agilent Technologies, Inc. 2009

21 Analysis of Signals After Digital Demodulation Coupled Markers 21 Agilent Technologies, Inc. 2009

22 Analysis of Signals After Digital Demodulation Without EQ With EQ Equalizer Impact 22 Agilent Technologies, Inc. 2009

23 Analysis of Signals After Digital Demodulation Measuring EVM at different points in the CP EVM definition requires measurements be made at 2 different points in time during the CP VSA allows the designer to investigate the impact of timedomain distortion on the CP by changing the window length used to make EVM measurements 23 Agilent Technologies, Inc. 2009

24 Demodulation and Cyclic Prefix EVM Analysis Windows (DL & UL) Total Transmitted Symbol Len CP Len Nominal Symbol Len FFT aligned with CP end FFT aligned with CP center Agilent VSA EVM Setting FFT aligned with CP start EVM is measured at two locations in time and the maximum of the two EVM is reported. i.e. EVM1 measured at EVM Window Start EVM2 measured at EVM Window End Reported EVM = max(evm1, EVM2) If EVM vs Time gets noticeably better when selecting EVM window center, it could be due to high ISI coming from a baseband filter that is very tight (optimized for ACLR). 24 Agilent Technologies, Inc. 2009

25 Analysis of Signals After Digital Demodulation Controlling the measurement interval allows detailed analysis on very specific and precisely-chosen portions of the signal This allows measurement results to be isolated to certain areas of trouble or interest Some terms are introduced to describe the timing and analysis portions associated with the VSA (next slide) 25 Agilent Technologies, Inc. 2009

26 Analysis of Signals After Digital Demodulation Terms: Result Length Length (in slots) of acquired IQ used in analysis and aligned to Frame start boundary event Measurement Offset Delay offset (in slots or symbols) relative to Frame start boundary event and beginning of Measurement Interval Measurement Interval Measurement interval (in slots or symbols),starting from Measurement Offset, used for analysis Note: Symbol by Symbol Resolution! 26 Agilent Technologies, Inc. 2009

27 Analysis of Signals After Digital Demodulation Slot 0, Symbol 5 = S-SS + PDSCH 27 Agilent Technologies, Inc. 2009

28 Analysis of Signals After Digital Demodulation IQ errors: IQ image and LO leakage IQ distortion can be easily seen in the following example resulting in mirror images about the center frequency using spectrum and spectrogram displays An MXG Signal Generator can emulate IQ impairments LO Leakage = Adding MXG IQ Offset (via MXG front panel) IQ Image = Adding MXG IQ Gain Imbalance (via MXG front panel) 28 Agilent Technologies, Inc. 2009

29 Analysis of Signals After Digital Demodulation IQ Imbalance Impact 29 Agilent Technologies, Inc. 2009

30 Summary LTE adds new complexities for transmitter design and test. More than ever, it s imperative to have a structured and systematic approach to signal test. Considered measurements made with an eye to troubleshooting, and cause vs effect, will bring benefits to the design and test engineering community. 31 Agilent Technologies, Inc. 2009

31 Receiver Design and Measurement Challenges Based on the book: LTE and the Evolution to 4G Wireless Chapter 6.5 4G World 2009 presented by: David L. Barner 32 Agilent Technologies, Inc. 2009

32 Key Objectives Understand SISO RX test challenges Understand Agilent SISO solutions available to address RX test challenges 33 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

33 Key LTE FDD enb Test Challenges LTE Conformance Tests Require Sophisticated Signals Various modulation bandwidths (1.4 MHz to 20 MHz) Various modulation types (QSPK, 16QAM, 64QAM) Transport channel coding with specific configurations, i.e. Fixed Reference Channels (FRC) Interfering Signals AWGN Emulation of channel propagation conditions New Conformance Tests Require Special Test Configuration Three performance requirements tests require dynamic changes in signal characteristics Closed loop control of RV index based on HARQ feedback Closed loop control of RF frame timing based on TA feedback Interference and Rx diversity tests require MIMO-like test configurations Agilent Technologies, Inc. 2009

34 enb Receiver Conformance Tests Receiver characteristics Reference sensitivity level Dynamic range Adjacent Channel Selectivity (ACS) Blocking characteristics Intermodulation characteristics In-channel selectivity Spurious emissions These test are performed open loop Performance metric = BLER Performance Requirements Performance requirements for PUSCH Multipath fading propagation conditions UL timing adjustment HARQ-ACK multiplexed on PUSCH High speed train conditions Performance requirements for PUCCH ACK missed detection for sing user PUCCH format 1a CQI missed detection for PUCCH format 2 ACK missed detection for multi user PUCCH format 1a Performance Requirements for PRACH These tests require HARQ feedback Performance metric = throughput Solving test needs: Flexibility to easily create varying signals that simulate real-world conditions Signal generation capability that evolves as the standard evolves to ensure most accurate test results Note: UE conformance tests are still being defined 35 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

35 Agilent 3GPP LTE enb Test Solutions enb Conformance Tests Receiver Characteristics Receiver Characteristics Wanted Signal Interfering Signal Dynamic Range (wanted interferer) Agilent Solution 7.2 Reference Sensitivity Level 7.3 Dynamic Range 7.4 In-Channel Selectivity 7.5 Adjacent Channel Selectivity 7.5 Narrowband Blocking 7.6 Blocking (in-band) 7.6 Blocking (out-of-band) 7.6 Blocking (Co-location with other base stations) FRC A1-1, 1-2, 1-3 QPSK Mod FRC A2-1, 2-2, QAM Mod FRC 1-2, 1-3, 1-4, 1-5 QPSK Mod FRC A1-1, 1-2, 1-3 QPSK Mod FRC A1-1, 1-2, 1-3 QPSK Mod FRC A1-1, 1-2, 1-3 QPSK Mod FRC A1-1, 1-2, 1-3 QPSK Mod FRC A1-1, 1-2, 1-3 QPSK Mod None required for this test -- Signal Studio & MXG AWGN 12.4 db Signal Studio & MXG E-UTRA with all BW 21.5 db Signal Studio & MXG E-UTRA Offsets up to 2.5 MHz* E-UTRA Offsets up to 4.66 MHz* CW or E-UTRA Offsets up to 7.5 MHz* CW Offsets up to GHz CW Freq from 728 MHz to 2690 MHz 48.1 db Signal Studio & MXG 51.1 db Signal Studio & MXG 57.1 db Signal Studio + MXG + PXB 85.1 db Signal Studio & MXG + PSG db Signal Studio & MXG + MXG 7.7 Receiver Spurious Emissions NA NA NA MXA Spectrum Analyzer 7.8 Receiver Intermodulation 7.8 Receiver Intermodulation (Narrow Band Intermodulation) FRC A1-1, 1-2, 1-3 QPSK Mod FRC A1-1, 1-2, 1-3 QPSK Mod CW offset up to 7.5 MHz* & E-UTRA offset up to 18.2 MHz* CW offset up to 415 khz* & E-UTRA offset up to 1780 khz* Notes * from channel edge of wanted signal Either ARB or real-time Signal Studio can be used Tests do not require channel emulation Test are performed open loop, i.e. no HARQ or timing adjustment feedback required 48.1 db Signal Studio & MXG + PXB 48.1 db Signal Studio & MXG + PXB Agilent Technologies, Inc. 2009

36 Agilent 3GPP LTE enb Test Solutions enb Conformance Tests Performance Requirements Performance Requirements Wanted Signal Channel Model PUSCH in Multipath Fading Propagation Conditions UL Timing Adjustment HARQ-ACK Multiplexed on PUSCH High Speed Train Conditions ACK Missed Detection for Single User PUCCH Format 1a CQI Missed Detection for PUCCH Format ACK Missed Detection for Multi User PUCCH Format 1a PRACH False Alarm Probability and Missed Detection FRC A3, A4, A5 QPSK, 16QAM, 64QAM FRC A7, A8 QPSK & 16QAM (SRS is optional) FRC A3-1, A4-3 to A4-8 QPSK, 16QAM FRC A3-2 to A3-7 QPSK (PUCCH is optional) PUCCH ACK PUCCH CQI PUCCH ACK PRACH Preamble EPA 5 Hz EVA 5, 70 Hz ETU 70, 300 Hz Moving Propagation Model a. ETU 200 Hz b. AWGN Channel Configuration 1x2 (2x RX diversity) 1x4 (4x RX diversity) 2x2 (2x RX diversity) 2x4 (2x RX diversity) (Stationary & moving UE) Feedback HARQ HARQ & timing adjustment Agilent Solution Real-time Real-time + Waveform Playback ETU 70 Hz 1x2 (2x RX diversity) -- Waveform Playback High Speed Train with: a. Open Space b. Tunnel for multi-antenna EPA 5 Hz EVA 5, 70 Hz ETU 70, 300 Hz ETU 70 Hz ETU 70 Hz ETU 70 Hz AWGN (no fading) 1x2 (2x RX diversity) 1x4 (4x RX diversity) 1x2 (2x RX diversity) 1x4 (4x RX diversity) 1x2 (2x RX diversity) 1x4 (4x RX diversity) 4x2 (2x RX diversity) (Requires 3 interferers) 1x2 (2x RX diversity) 1x4 (4x RX diversity) HARQ Real-time Real-time or Waveform Playback Real-time or Waveform Playback -- Waveform Playback -- Waveform Playback Notes All tests require channel emulation and AWGN All tests require RX diversity if supported by enb Industry is requesting up to 4-way RX diversity for all tests, i.e. 1x4, 2x4, & 4x4 MIMO Agilent Solution indicates type of Signal Studio for 3GPP LTE FDD software. N5106A PXB MIMO receiver tester and N5182A MXG vector signal generator are also required. Agilent Technologies, Inc. 2009

37 Agilent 3GPP LTE Test Solutions Rx RF/BB Front End Verification Signal Studio - Uplink FDD LTE - ARB basic capability MXA Signal Analyzer Generate simple test signals Create CW signals Create multi-tone signals Generate simple LTE signals Ultimate physical layer flexibility Supports March 09 version of LTE standard Selectable BW from 1.4 MHz to 20 MHz Select PUSCH modulation: QSPK, 16QAM, 64QAM Configurable data payloads Allocate resource blocks in frequency & time MXG Vector Signal Generator Analog I/Q, Digital I/Q, DigRF RF Receiver Front End Measure basic RF parameters Analyze amplitude flatness Measure gain at each stage Analyze phase linearity Determine noise figure Measure EVM of components & subsystems Page 38 Agilent Technologies, Inc. 2009

38 Agilent 3GPP LTE Test Solutions Rx Conformance Test Signal Studio Uplink FDD LTE Real-time capability Real-time LTE Signal Generation PXB accepts closed loop feedback HARQ ACK/NACK signals Timing adjustment feedback LTE signal continuously adjusted based on feedback Predefined Fixed Reference Channel definitions Real-time Channel Emulation PXB MIMO Rx Tester Feedback enb Standards based channel models Custom defined channel models 24 paths of fading 120 MHz modulation bandwidth Simplified power calibration Digital I/Q MXG Vector Signal Generator RF Interfering Signals Add CW blocking signals Add modulated signals for blocking &interoperability test Calibrated AWGN for accurate C/N ratios Page 39 Agilent Technologies, Inc. 2009

39 Agilent 3GPP LTE Test Solutions RF/BB Channel Emulation PXB MIMO Rx Tester RX Diversity Tx0 h00 h01 Rx0 Rx1 Advanced Channel Emulation 120 MHz fading bandwidth 24 paths of fading per channel Up to 8 independent fading channels Custom MIMO correlation settings Configurable antenna parameters Standards based channel models Simplified power calibration MXG Vector Signal Generator Signal Studio RF to RF fading with MXA Simplified Signal Routing & Summing Combine independent channels for diversity or MIMO Windows operating system Intuitive GUI Scalable Architecture Connect to ESG, MXG, & DSIM for signal creation Connect to MXA for RF fading applications Field upgradable with calibrated DSP blocks Page 40 Agilent Technologies, Inc. 2009

40 Agilent 3GPP LTE Test Solutions RF/BB Interference and Interoperability Test PXB MIMO Rx Tester MXG Vector Signal Generator Interoperability testing Signal Studio 0 Configuration flexibility Create: LTE, W-CDMA/HSPA, GSM/EDGE, cdma2000, 1xEV-DO, WiMAX, WLAN Up to four internal baseband generators Sum CW carriers with wanted signal Sum modulated carriers with wanted signal Sum custom Matlab waveforms with wanted signal Add calibrated AWGN for accurate C/N ratios Scalable Test Solutions Tailor capability & performance from SISO to MIMO Easily upgrade as your test needs evolve Connect to ESG, MXG, & DSIM for signal creation Connect to MXA for RF fading applications Field upgradable with calibrated DSP blocks High Performance Real-time uplink FDD LTE signal creation Real-time MIMO channel emulation Simplified power calibration Wide bandwidth ready for LTE Advanced (Rel 10) Page 41 Agilent Technologies, Inc. 2009

41 PXB Closed Loop Test Concept HARQ & Timing Adjustment Tests Throughput Testing Equipment Configuration Signal Studio N7624B 3GPP LTE FDD CMOS 3.3 V inputs from enb HARQ Level Triggered Timing Adjustment Serial Data N5106A PXB HARQ ACK/NACK Timing Adjustment Frame Pulse 10MHz enb LAN GPIB 10MHz Digital I/Q Baseband w/ Fading N5182A MXG RF Dynamically Changing RF Frame Timing based on TA RV Index based on ACK/NACK Agilent Technologies, Inc. 2009

42 Timing Adjustment Conformance Test Concept enb Timing Adjustment transmitted back to UE, to align UE with enb frame timing 1 symbol (2048 Ts) Normal Cyclic Prefix Resource Blocks Details Stationary UE and moving UE transmit in same subframe, but with different subcarriers Moving UE simulates changing propagation path lengths & therefore different arrival times at enb enb must command moving UE to advance or delay timing of transmission such that the signal arrives at enb with proper frame timing, i.e. does not overlap into adjacent symbols Timing adjustment test is performed with even subfames occupied Sounding Reference Signal (SRS) is optional for this test This test is performed with real-time HARQ feedback enb Frame Timing Stationary UE Moving UE simulates changing propagation path lengths In this example, the mobile UE is assigned blue Resource Blocks Moving UE signal can arrive at wrong enb frame timing as path length changes UE transmission interferes with next symbol without timing adjustment Agilent Technologies, Inc. 2009

43 Common RF Front End Measurements Amplitude Flatness Issues LTE can correct some amplitude / phase errors with RS Errors will manifest themselves as EVM Important because LTE BW is wider than other cellular standards Need to test individual components, i.e. Amplifiers, Filters, Mixers, etc How to test LTE signals could be used Some configurations of LTE do not utilize all subcarriers Power not constant over LTE BW Alternate approach with multitone signals Space tones over BW of interest Correction techniques enable flatness of ~0.1 db Signal Studio for Multitone Distortion 44 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

44 Common RF Front End Measurements Amplitude Flatness Amplitude Flatness Test Calibration Configuration Signal Studio for Multitone Distortion MXG Signal Generator RF Output DUT Additional Benefit: Entire System is calibrated! LAN 10 MHz Reference MXA Signal Analyzer RF Input 45 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

45 Common RF Front End Measurements Amplitude Flatness Multitone Performance with Corrections 50 Tones Spaced over 100 MHz Before Corrections After Corrections Note: Scale per div is 0.2 db in each graph Corrected flatness is ~ 0.1 db! 46 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

46 Common RF Front End Measurements Phase Linearity Issues LTE can correct some amplitude / phase errors with RS Errors will manifest themselves as EVM Important because LTE BW is wider than other cellular standards Need to test individual components, i.e. Amplifiers, Filters, Mixers, etc How to test Can t measure phase w/ Spectrum Analyzer High degree of integration may make network analyzer impractical MXG A VSA Measurement Ampl. Flatness Phase Linearity VSA can measure Amplitude flatness and also Phase linearity of modulated LTE signal 47 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

47 Common RF Front End Measurements Analog-to-Digital Converter Issues Analyzing data in digital domain How to test Modulated LTE stimulus Use Logic Analyzer or DSIM/ESG with VSA 89601A VSA Software Signal Studio for 3GPP LTE Digital I/Q DSIM N5102A RF-IC DigRF MXG Signal Generator Digital Stimulus / Analysis 48 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

48 Common RF Front End Measurements Other Things to Consider Automatic Gain Control (AGC) Noise Figure Receiver Error Vector Magnitude Receiver Performance under Impaired Conditions 49 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

49 Receiver Performance under Impaired Conditions -Phase Noise Impairments LTE subcarrier spacing is 7.5 khz or 15 khz Close subcarrier spacing makes RX highly susceptible to phase noise problems Results in degraded EVM Signal generator can be used as LO, but typically have much better phase noise than RX N5182A phase noise characteristic an be degraded very precisely Determine performance required for LO in RX Determine performance of RX with impaired signal from TX Pedestal set to -90 dbc 50 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

50 Receiver Performance under Impaired Conditions -Phase Noise Impairments N5182A MXB Phase Noise Plots (w and w/o Phase Noise Impairments) Pedestal Phase Noise Set at -90 dbc/hz MXG non-impaired phase noise characteristic at -116 dbc/hz ~26 dbc/hz 51 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

51 Receiver Performance under Impaired Conditions -Phase Noise Impairments Two Carriers Spaced at 15 khz (w and w/o Phase Noise Impairments) Phase noise characteristic with pedestal set at -90 dbc/hz MXG noise floor ~26 dbc/hz 52 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

52 Receiver Performance under Impaired Conditions -Phase Noise Impairments EVM degradation due to added Phase Noise impairment at -90 dbc/hz EVM has increased from less than 0.5% to more than 6% Note: Constellation appears more Gaussian than expected rotation. Why? 53 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

53 Receiver Performance under Impaired Conditions -AWGN Impairments LTE Signal with AWGN 5 MHz LTE signal with 20 MHz AWGN at 30 db C/N Although specified, is not a real representation of actual interference due to NB allocations used in OFDM systems -Just a simplified model! -Better for CDMA 54 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

54 Receiver Performance under Impaired Conditions -IQ Impairments Many Systems employ IQ Demoduators Typical impairments I/Q DC offsets, IQ Quadrature Phase I/Q Gain Imbalance I/Q Skew (Delay) Issues Creates LO feedthrough Creates unwanted images Results in degraded EVM (Classic V ) LTE Signal w/ IQ impairments Image LO Feedthrough Original signal (offset +30 MHz) Can add with signal generator Compensate for errors in RX Determine impact from TX 55 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

55 Receiver Performance under Impaired Conditions -IQ Impairments EVM decreases from 0.5% to ~2% LTE signal with 5 ns skew between I & Q More noticeable in EVM vs subcarrier Classic V : Timing error gets progressively worse as subcarriers get farther away CF 56 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

56 Receiver Performance under Impaired Conditions -Interference Can receiver correctly reject interfering carriers? Selectivity tests Blocking tests IMD immunity tests Multi-Carrier output from single MXG Note: Don t require multiple Sig Gens, power combiners, or isolators! W-CDMA LTE 57 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

57 Receiver Performance under Impaired Conditions -Fading Propagation Conditions three major components: Delay spread (Amplitude / phase fluctuations) result of multipath profile Doppler spreading result of TX or RX movement TX/Rx Antenna Correlation Matrix Amplitude fluctuations as a function of time Deep Fade ~ 40dB! 58 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

58 Receiver Performance under Impaired Conditions -Fading Effect of Amplitude and Phase changes on a QPSK constellation QPSK modulated carrier faded with two paths of Rayleigh fading 59 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

59 Baseband Measurements Tools for Creating RF-RF Faded Signals Signal Inputs RF 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 RF N5102A MXA Page Page 60 PXB ESG or MXG Agilent Restricted June 17-18, Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009

60 Questions? www/agilent.com/find/4gworld Excuse me, is this the Society for Asking Stupid Questions? Agilent Technologies, Inc. 2009

61 Backup 62 Agilent Technologies, Inc Receiver Design and Measurement Challenges June 17-18, 2009