Production Test and Spectral Monitoring

Transcription

1 1

2 Production Test and Spectral Monitoring Stephen Plumb

3 Key RF Building Blocks Symbol Name Types Function Amplifier (2 port) Power Amplifier Low Noise Amplifier Amplify signal before transmission (high power). Amplify weak received signal (add little noise). LO Mixer (3 port) Upconverter Downconverter Translate lower frequency to higher frequency. Translate higher frequency to lower frequency. Oscillator (1 port) Voltage-controlled oscillator Quartz crystal Generate frequency based upon voltage input. Generate a single stable frequency. Filter (2 port) Passive Active Digital Filter out undesired signal content. 3

4 Generic Digital Phone 4

5 B. Common Wireless Standards for Cellular Applications GSM/EDGE UMTS (WCDMA, HSPA, etc) LTE 3GPP IS-95 CDMA2K 1xED/DO 3GPP2 WiMAX Others 5

6 C. Testing Methods Semiconductor Transceivers/PCB Integrated mobile station or base station 6

7 Testing Methods for Cellular Semiconductors Usually require multiple testers, all controlled from a PC RF usually works in loopback IC communication could be digital or a simple digital protocol. RF Tester Power Supply Other 7

8 Testing Methods for Transceivers/PCB DUT communication enables/disable some test scenarios Test can be very extensive depending on the different standards the transceiver supports Customers work constantly with suppliers to get the specific test modes they need in order to further reduce test cost RF Tester Power Supply Other 8

9 Testing Methods for Mobile Handsets Signaling Mode (call processing) Non-signaling Mode (single ended) CW Test Tester Tester RF out RF in RF out RF in DUT Control 9

10 GSM/EDGE Overview Modulation GSM: GMSK EDGE: Offset 3π/8 8-PSK 10

11 GSM/EDGE Generation Payload Control Configures the data to be transmitted in each burst Can select which training sequence to use Three types of bit patterns: o Pseudorandom o Pattern o User Defined 11

12 GSM/EDGE Generation I/Q Perturbations I/Q Perturbations can be added to the generated signal to simulate generator distortion Has ability to artificially add noise into the signal Useful for testing receiver sensitivity to distortions in the received signal 12

13 GSM/EDGE Measurements GSM Signal Analysis Transmit Power Power Versus Time Output RF Spectrum Phase/Frequency Error EDGE Signal Analysis Transmit Power Power Versus Time Output RF Spectrum EVM 14

14 GSM Phase and Frequency Error (PFER) Measures modulation quality in GSM signals Reports mean and maximum statistics when analyzing multiple bursts Reports symbol with peak phase error to assist in generation of useful custom test packets. 15

15 EDGE Error Vector Magnitude (EVM) Defined as the ratio between the magnitude of the error vector and the ideal symbol vector Very similar in function to the PFER measurement in GSM Can compensate for amplitude droop in the measurement 16

16 Adjacent Channel Power Set Offset Channels Set Averaging [Images from API] 17

17 F. Applications for Cellular Test Transceiver Power amplifier RF switching PCB board test Calibration Final test for phone 18

18 PCB Board Test DUT control requirements o Custom design regulates test capabilities Typical tests o Calibration Power Test o Modulation accuracy measurements o Spectral quality measurements o Call processing is sometimes performed if test time and cost are not drivers 21

19 Final Test for Phone DUT control requirements o Very similar to PCB but still custom to the design Typical tests o Power Test ensures the antenna was connected successfully o Other tests can perform depending on the manufacturing process (Signaling test) 22

20 Summary Cellular devices include all cellular components such as power amplifiers, switches, transceivers, and so on All cellular devices use similar type of measurements divided in spectral type, time domain and modulation accuracy NI toolkits help perform custom tests using TestStand 23

21 What Is Spectrum Monitoring? RF Measurements & Signal Processing That Shows How Radio Spectrum Is Being Used. Spectrum Monitoring always Signal Intelligence 24

22 Who Does Spectrum Monitoring? Government: Ensure spectrum users obey licensing to minimize interference among users. Commercial Spectrum Owners: Monitor operation of systems and troubleshoot interference problems. Defense Intelligence: Characterize the radio frequency environment to support their missions. 25

23 Spectrum Monitoring Challenges: Limited spectrum White space spectral efficiency Frequency purchase & allocation Unwanted signals Antenna research Femtocell development & deployment Signal parameters (frequency, power, bandwidth) High throughput (smart phones, WIFI hotspots) Issues: 1. Interference 2. Crowded spectrum 3. Frequency allocation 26

24 Issue 1: Interference Interference occurs when unwanted signals are received at signal strengths that desensitize the receiver. Drivers: Increasing numbers of transmitters create environment where interference is more prevalent. Compliant signals in the licensed and unlicensed spectrum Potential components of an interfering signal. 27

25 Issue 2: Crowded Spectrum Co-Channel Interference occurs when a distant base station signal is too strong. Drivers: Creates hand-off problems Increasing number of cell sites and providers Sources: Adjacent cell sites & transmitters ISM & public bands Adaptive radios Shared spectral space Differing international frequency allocations 28

26 Issue 3: Frequency Allocation Sources: Government Regulation of Spectrum Commercial Network health Interference detection and identification Defense Surveillance Threat targeting Jamming / Deception Surveillance for Terrorist Activity 29

27 Spectrum Monitoring System Requirements Spectrum Scan Compliance to ITU High Dynamic Range Preselection Preamplifier Roofing Filters Modulation Recognition/Signal Identification Signal Processing DDC Channelization Data storage Synchronization to timing standards (GPS) 30

28 Equipment Choices Spectral Monitoring Instruments: Power Meter Signal Strength Meter Spectrum Analyzer Vector Signal Analyzer (VSA) Signal Monitoring Receiver All are useful for spectral monitoring to some extent 31

29 Spectral Monitoring Demo - USRP 32

30 Analyzer Architectures Zero IF Direct I/Q downconversion Pros o Low-cost components, fast tuning Cons o Images Single Stage Pros o Lower cost, fast tuning Cons o Images Multi Stage Superhet Pros o Image rejection for spectral measurements Cons o Expensive components o Slower sweeps RF In RF In RF In Amplifier Amplifier Atten Atten Mixer Amplifier Mixer N C O IF Filters ADC ADC IF Out ADC DDC ADC 33

31 Distortion products generated in over-the-air measurements 34

32 Spectrum Monitoring Receivers Add.. Preselection Filters Post-filter Preamplifiers Roofing filters (selectable BW analog IF filters) 36

33 Spectrum Monitoring Receiver RF Section Preselection Filters: Reduces front-end harmonic distortion Preamplifier: Improves sensitivity. But, not at the expense of distortion! 37

34 Spectrum Monitoring Receivers IF Section Roofing Filters: Reduce distortion of final IF amplifiers and digitizer spurs 38

35 Equipment and Measurement Considerations Receiver Fast tuning speed Preselector to eliminate interfering signals Exceptional Second Harmonic Intercept Superior Third Order Intermodulation High Dynamic Range via roofing filters Great sensitivity Signal Processing Graphical display Extensible to multiple channels Transportable, size and weight Advanced Features GPS / Time-stamping Real time data processing Storage and Playback Specialized application software 39

36 NI PXIe-5667 Spectrum Monitoring Receiver 20 Hz to 3.6 / 7 GHz Frequency Range Dynamic Range >110 db and NF<12dB Up to 50 MHz instantaneous bandwidth Second Order Intercept (SOI) of 80 dbm and TOI >17dBm RF Spectrum with scan rates of up to 30 GHz/sec Real-time signal analysis and advanced signal processing in FPGA hardware NI PXIe-5693 Pre-selection Filter Module 20 MHz to 7 GHz Preselection DC to 30 MHz Bypass path 16 sub-octave pre-selection filters Integrated preamplifier 4 notch filters for FM & TV broadcast frequencies NI PXIe-5694 IF Conditioning Module 7 IF Roofing Filters: 30 khz to 50 MHz IF Outputs: MHz, 87.5 MHz, 21.4 MHz 40

37 NI PXIe GHz Spectrum Monitoring Receiver PXIe-5622 Digitizer PXIe-5694 IF Conditioning PXIe-5603 Downconverter PXIe-5653 LO PXIe-5693 Preselector 41

38 NI PXIe GHz Spectrum Monitoring Receiver PXIe-5694 IF Conditioning PXIe-5693 Preselector PXIe-5622 Digitizer PXIe-5605 Down Converter PXIe-5653 LO 42

39 PXIe-5667 System Specifications Fully Compliant to ITU Receiver Specifications Characteristic PXIe-5667 ITU Handbook Frequency Range 20 Hz GHz 20 Hz GHz 20 MHz 3 GHz Preselection Sub-octave Filters Sub-octave Filters Intermediate Frequency (Roofing) Filters Noise Figure 7 Filters: 30 khz 50 MHz 12 db < 3 GHz 14 db < 7 GHz db Second Harmonic Intercept >+80 dbm +40 dbm IP3 >+17 dbm +10 dbm Phase 10 khz > -129 dbc/hz -100 dbc/hz Scan Rate 30 GHz/sec -- Dynamic Range >110 db -- 43

40 NI FlexRIO System Architecture PXI/PXIe NI FlexRIO Adapter Module Interchangeable I/O Analog or digital NI FlexRIO Adapter Module Development Kit (MDK) NI FlexRIO FPGA Module Virtex-5 FPGA 132 digital I/O lines Up to 512 MB of DRAM PXI Platform Synchronization Clocking/triggers Power/cooling Data streaming 44

41 Signal Processing with LabVIEW DDCs / DUCs / Filtering / Channelization / Decimation In-band power, adjacent-channel power, burst timing, BER, EVM, MER, ρ, and more Advanced RF, spectral measurement, and modulation software Custom modulation formats and channel coding and impairments Wavelet and filter-bank design for short-duration signal characterization, noise reduction, and detrending Time-frequency analysis -- analytical, graphical tools for signals with evolving frequency content Time-series analysis -- statistical analysis for description, explanation, prediction, and control Access Xilinx CORE Generator IP Libraries in LabVIEW FPGA 45

42 NI FlexRIO Peer-to-Peer Architecture >800 MB/s one-way >700 MB/s both ways ~10 us latency Up to 16 streams per FPGA 46

43 Multiple Transmitters & Receivers (x8) 47

44 What is a Spectrogram? A time-varying spectral representation that shows how the spectral density of a signal varies with time. Also known as spectral waterfall. 48

45 How to use Spectral Mask Triggering? Describe a spectral mask based on the signal of interest using a piecewise linear function Configure for a real-time spectrum acquisition ideally using an FPGA coprocessor Arm the trigger and start the acquisition. A violation of the mask will fire the trigger 49

46 Spectral Masking Triggering Frequency Mask Mask Violation Trigger Indicator 50

47 Conclusion New Standards and Requirements demand a more flexible system configuration. Modular system design tools allow users to adapt. Consider the best receiver architecture for your requirements. Consider future requirements. 51

48 52