TESTING METHODS AND ERROR BUDGET ANALYSIS OF A SOFTWARE DEFINED RADIO By Richard Overdorf

Size: px

Start display at page:

Download "TESTING METHODS AND ERROR BUDGET ANALYSIS OF A SOFTWARE DEFINED RADIO By Richard Overdorf"

Lily Fox
5 years ago
Views:

1 TESTING METHODS AND ERROR BUDGET ANALYSIS OF A SOFTWARE DEFINED RADIO By Richard Overdorf

2 SDR Considerations Data rates Voice Image Data Streaming Video Environment Distance Terrain High traffic/low traffic Priority

3 SDR s Flexibility Interoperability One radio with ability to communicate with everyone. Upgradable

4 Frequency Configurable Cognitive Radio Front ends that are configurable require one of the following Very wide bandwidth Flexible front-end NCO

5 SDR Testing Analog vs. Digital Words Different Signal Formats Internal FPGA Signals Cross Format Analysis New Development Processes SD R DSP Digital (SSI) BB (I-Q) IF/RF Logic Analyzer Oscilloscope Signal Analyzer

6 Golden Box Testing Quantitative data is not available Makes statistical process control difficult Inability to separate impairments Can easily mask other problems with radio Can lead to interoperability issues Golden Radio DUT

7 What is Residual BER? Bit Error Rate (BER) Residual Error Floor Received Signal Level (dbm) 0 Threshold Overload

Loop-back tests remove impairments, leading to false results.

8 BER Testing Methods Loopback Testing Easy Measurement Measures Overall Performance Issues in Loopback BER Some impairments are additive across the wireless system. Loop-back tests remove impairments, leading to false results. System budgets are essential for interoperability. Bit Error Ratio Tester (BERT) Agilent ADS Signal Sources - digital/analog/rf

9 EVM { Q Measured Signal φ Magnitude Error (IQ error mag) Error Vector Reference Signal Phase Error (IQ error phase) Error Vector Magnitude I

10 Waveforms SCM Development Flexibility Troubleshooting CDMA Security Ability to carry multiple users Transmit more efficiently OFDM Robust in the presence of SCM interferes Adapt and handle challenging channels

Digital Impairments FPGA resources vs signal quality quality Filters Filter ripple, rolloff, filter

word length cost speed Agilent System Vue Filter Design tool Overflows Truncation reduces DR Rounding

quality Overflows: distort signals and will reduce BER/EVM Truncation: reduces DR, reduced S/N

11 Digital Impairments FPGA resources vs signal quality quality Filters Filter ripple, rolloff, filter rejection Better filter requires more filter taps More taps means more multiplies Multiplies increase word length cost speed Agilent System Vue Filter Design tool Overflows Truncation reduces DR Rounding requires additional operations Ripple: typically removed by receiver equalizer but can effect signal quality Overflows: distort signals and will reduce BER/EVM Truncation: reduces DR, reduced S/N effects EVM/BER Insufficient rejection or rolloff: cross channel interference reducing BER/EVM in co-channels

Digital Impairments FPGA resources vs signal quality step Accumulator index 0 0.00 1 0.31 2 0.63 3 0.94 4 LUT 1.26 5 1.57 6 1.88 7 2.20 8 2.51 9 2.

12 Digital Impairments FPGA resources vs signal quality step Accumulator index LUT LO NCO Numerically controlled oscillators Frequency resolutions Improve with larger LUT size Amplitude Quantization spurs Improve with greater word length or dithering Phase truncation (quantization) Improve with enhanced dithering techniques or greater LUT resolution

Noise Effects Noise Figure SCM CDMA OFDM Phase Noise Phase Noise in OFDM Results in each subcarrier interfering with other subcarriers Modulates each sub-carrier to the point that

13 Noise Effects Noise Figure SCM CDMA OFDM Phase Noise Phase Noise in OFDM Results in each subcarrier interfering with other subcarriers Modulates each sub-carrier to the point that they no longer look like simple sinusoids within the FFT interval Causes the nulls of the sin(x)/x spectrum to fill in, creating interference between every subcarrier and its neighbors

14 Other Impairments Primary Secondary Channel Loss Range Natural Barriers Interference Spurious Power Supply AM/AM Distortion PA ADC Quantization AM/PM Distortion PA PA Leveling Stability Delay Distortion/ISI BB Filters IF Filters

15 System Budget Concept Transmitter Receiver Modem IF PA LNC IF Modem Modulator Demodulator FPGA/GPP FPGA/GPP Allows the engineer to separate the modulator, transmitter, receiver, and demodulator issues Helps ensure interoperability between different receivers or transmitters Essential for assuring customer premises equipment units will not dribble years after the base station is installed Gives the ability to upgrade modem and ensure that the current RF will support it Budgeting helps control costs It mathematically relates key analog metrics to digital bit errors used to evaluate the system

16 RF Design vs. DSP Many times you can cut costs by implementing tracking and equalizing Cost come in development and DSP power PA LNC

17 Testing Impairments in SDR ADS model to input multiple waveforms Variable impairments

18 Impairments Using ADS

19 Using EVM to trouble shoot the radio 11.2 MHz Digital IF 70 MHz analog IF 2.3 GHz RF 5.4 EVM 11.2 MHz IF 5.7 EVM Digital IQ 9.2% EVM

20 Phase Noise effects on CDMA and WiMAX Example of effects of BER in a SDR BER WiMAX W-CDMA khz offset WiMAX W-CDMA *Waveforms are NOT equal in data throughput

BER vs 2 nd LO PN vs Amp gain for WiMAX 0.30 0.28 0.26 0.24 0.22 0.20 0.18 0.16 0.14-55 dbc 0.

21 BER vs 2 nd LO PN vs Amp gain for WiMAX dbc BER.DF.Amp_Gain BER -65 dbc -60 dbc -55 dbc -65 dbc

22 How to Test Impairments in SDR 89601A Digital Impairments IQ impairments IF/RF impairments SD R DSP Digital (SSI) BB (I-Q) IF/RF Logic Analyzer Oscilloscope Signal Analyzer

23 Phase Noise Measurements Techniques Spectrum analyzer technique Phase detector techniques discriminator method PLL/reference source method Direct Spectrum Phase Detector db k 1 M 100 M db k 1 M 100 M db c/h z db db c/ Hz db -180 db Frequency Offset from Carrier -180 db Frequency Offset from Carrier

power specifications for mixers, filters, amplifiers ACP Test for leakage into

24 Other Measurements CCDF Peak to average power measurement Fully characterize the power statistics of a digitally modulated signal Vital in setting signal power specifications for mixers, filters, amplifiers ACP Test for leakage into other channels Inter-modulation products Phase Noise SEM Test for unwanted emissions

25 Conclusions and Review The dynamic nature of SDR requires in depth testing Multi-dimensional system budgeting is highly recommended to drive down costs Beneficial to have tools that work from start of design to deployment

TESTING METHODS AND ERROR BUDGET ANALYSIS OF A SOFTWARE DEFINED RADIO

TESTING METHODS AND ERROR BUDGET ANALYSIS OF A SOFTWARE DEFINED RADIO Richard Overdorf (Agilent Technologies, Inc., Santa Rosa, CA, USA; richard_a_overdorf@agilent.com) ABSTRACT Ideally a Software Defined