Configuring Software-Defined WLAN Test Systems

Publish Date: Jul 12, 2013 | 10 Ratings | 4.10 out of 5 |  PDF


Today, it is hard to picture life without wireless local area network (WLAN) devices. In fact, the use of WLAN or WiFi extends to many products such as access points, routers, and even cell phones. As the market for WLAN devices has grown, so has the number of engineers performing WLAN measurements. This white paper offers basic insights into WLAN test measurement options and an overview of the physical layer. It also explains how you can use software-defined RF measurement systems to quickly and accurately perform a full suite of WLAN measurements. After reading this white paper, engineers unfamiliar with WLAN testing should have a basic understanding of the types of measurements involved.

Table of Contents

  1. Introduction to the WLAN Physical Layer
  2. Overview Of RF Virtual Instrumentation
  3. Introduction to the NI WLAN Measurement Suite
  4. Typical WLAN Measurements
  5. Transmit Power
  6. Error Vector Magnitude
  7. Spectrum Mask Measurement
  8. Conclusion
  9. Resources

1. Introduction to the WLAN Physical Layer

WLAN standards are defined and maintained by the IEEE 802.11 working group, which consists of representatives ranging from chip vendors to access point manufacturers. The group has defined a variety of 802.11 standards including versions from 802.11a to 802.11z and beyond. However, for WLAN devices, the most common protocols are IEEE 802.11a, b, g, n, and ac.

In 1999, the working group ratified the 802.11a and 802.11b standards to enable WLANs. The IEEE 802.11a standard provides data rates of up to 54 Mb/s in the 5 GHz unlicensed industrial, scientific, and medical (ISM) frequency band. By contrast, the IEEE 802.11b standard provides data rates of up to 11 Mb/s in the 2.4 GHz ISM band. In 2003, the IEEE 802.11g was introduced to provide data rates of up to 54 Mb/s in the 2.4 GHz ISM band as well. The newest version is IEEE 802.11n, which incorporates additional features such as multiple input, multiple output (MIMO) and concurrent channel use to provide expected data rates of 300 Mb/s in both the 2.4 and 5 GHz frequency bands.

The two basic transmit schemes used by WLAN are direct-sequence spread spectrum (DSSS) and orthogonal frequency division multiplexing (OFDM). In addition, the underlying modulation schemes can range from CCK to quadrature schemes such as BPSK and 64-QAM. Table 1 lists the standards that use specific transmit schemes and modulation types. 

Table 1. Transmit Schemes and Modulation Types for Various 802.11 Revisions

Unlike other OFDM-based standards such as WiMAX (IEEE 802.16d/e) and 3GPP Long Term Evolution (LTE), the OFDM signals in WLAN use the same modulation scheme for every subcarrier. Thus, for IEEE 802.11a/g signals, the modulation scheme directly correlates to the maximum data rate and coding rate for a specific signal. Table 2 illustrates this relationship.

Table 2. Relationship between Data Rate, Coding Rate, and Burst Duration

Table 2 shows that high data rates such as 54 Mb/s require the use of higher-order modulation schemes such as 64-QAM. Moreover, the time duration for a standard burst with 1,024 data bits is significantly larger for lower-order modulation schemes. Given the correlation between longer burst durations and longer measurement times, understanding this relationship is important when optimizing a test system for measurement speed. In general, when performing an error vector magnitude (EVM) measurement on a single burst, you can achieve a faster measurement result when you configure the instrument to acquire only as much data as necessary for the measurement. For example, when measuring a 64-QAM burst, setting an acquisition time length of 200 µs produces faster measurement times than setting it to 10 ms or more.

Back to Top

2. Overview Of RF Virtual Instrumentation

You can choose from a variety of instruments to test WLAN devices by using the National Instruments software-defined approached to WLAN testing. To show the differences between traditional instrumentation and virtual instrumentation, this white paper provides a basic overview of the architecture of a virtual PXI measurement system.

PXI instruments combine a high-performance multicore controller, high-speed PCI/PCI Express data bus, and optimized measurement algorithms to deliver industry-leading measurement speeds. Software used for WLAN measurements is the NI WLAN Measurement Suite, featuring the NI WLAN Analysis and WLAN Generation toolkits. The recommended NI hardware includes the NI PXIe-5663 vector signal analyzer and the NI PXIe-5673 vector signal generator. The NI PXIe-5663 is capable of signal analysis from 10 MHz to 6.6 GHz with up to 50 MHz of instantaneous bandwidth. The NI PXIe-5673 offers signal generation from 85 MHz to 6.6 GHz along with 100 MHz of instantaneous bandwidth. You can pair either instrument with additional generators or analyzers to perform phase-coherent measurements as well. Figure 1 shows a typical WLAN device test system configuration featuring a vector signal generator and vector signal analyzer.


Figure 1. A PXI System Configured for WLAN Measurements

Software-defined instruments are ideally suited for automated test applications. Architecturally, the main difference between PXI modular instruments and traditional instruments is the processing core. While both systems use many similar components, the major distinction is that PXI systems can incorporate high-performance multicore central processing units (CPUs). Figure 2 shows that traditional and PXI instruments feature many common core components, including memory, high dynamic range analog-to-digital converters (ADCs), and a high-performance RF front end.

Figure 2. A user-defined CPU is a critical component of PXI RF instruments.

Multicore CPUs on PXI modular instruments deliver world-class signal processing capability. As a result, many PXI-based measurement systems achieve measurement times that are significantly faster than traditional instruments. Generally, the CPU performance from chipset vendors such as Intel and AMD typically increases over time in accordance with Moore’s law. Thus, you can upgrade only the controller of a PXI system as newer processors are released. As a result, you can significantly improve the measurement speed of a given test system at a fraction of the cost of replacing the entire system.

A second benefit of the software-defined approach to instrumentation is the ability to test multiple wireless standards on the same hardware platform. This benefit is particularly useful to engineers developing multi-standard consumer products or system on a chip (SOC) devices. In the past, engineers testing devices with a GPS receiver, WLAN radio, and FM tuner had to purchase several unique instruments. With software-defined instrumentation, they can purchase a common set of hardware and use a unique software toolkit for each standard they need to test. Figure 3 illustrates this concept.



Figure 3. Architecture of a Software-Defined Instrument

Figure 3 shows that you can use a general-purpose RF front end (either generator or analyzer) with a Windows-based CPU to create a software-defined instrument. With NI software-defined RF instruments, you can test WLAN, GPS, GSM/EDGE/WCDMA, WiMAXTM, BluetoothTM, DVB-T/ATSC/ISDB-T, FM/RDS/IBOC, and many other wireless standards. 

Back to Top

3. Introduction to the NI WLAN Measurement Suite

Given the software-defined characteristics of PXI instrumentation, bundles such as the NI WLAN Measurement Suite and equivalent software are critical components of a measurement system. The WLAN Measurement Suite consists of the NI WLAN Generation Toolkit and NI WLAN Analysis Toolkit. Both toolkits contain an API in LabVIEW, LabWindows/CVI, and ANSI C/C++, and both work in with PXI RF vector signal generators and analyzers. At a high level, think of the WLAN Generation Toolkit as a software package used to create IEEE 802.11a/b/g/n/ac signals. The WLAN Analysis Toolkit simply provides measurement results from signals acquired with a vector signal analyzer. Observe a block diagram of this approach in Figure 4.


 Figure 4. Architecture of a WLAN Test System

Using either the property nodes or the programming API, you can configure settings such as the specific standard, data rate, burst interval, and carrier frequency. Figures 5 and 6 illustrate how to adjust typical settings with either a property node or the programming API.

Figure 5. WLAN Measurement Settings Configured with a LabVIEW Property Node


Figure 6. WLAN Measurement Settings Configured with the LabVIEW programming API


Figure 6a. WLAN Measurement Settings Configured with the LabWindows/CVI programming API

The getting started example programs are designed to provide a tool for configuring automated measurement applications. For more interactive measurements, you can also use a LabVIEW or LabWindows/CVI demo panel similar to the one shown in Figure 7.

Figure 7. LabVIEW Demo Panel for WLAN Measurements

 Figure 7 shows a basic 802.11g spectrum mask in the frequency domain. Note that all measurements described in the subsequent section were performed using this example. 

Back to Top

4. Typical WLAN Measurements

When characterizing any WLAN component or radio, the specific measurements you use are often dependent on the devices you are testing. For example, if you are characterizing a power amplifier (PA), you might rely on the combination of an EVM and IM3 (third-order distortion) measurement to characterize nonlinearity. However, a carrier offset measurement is not important because it is a function of the RF signal generator. As a general-purpose reference, Table 3 lists some of the most common WLAN measurements. While only a subset of these measurements are described in the following sections, you can use the WLAN Analysis Toolkit to perform many measurements, as shown in Table 3.

Table 3. Measurements Performed with the WLAN Analysis Toolkit

Back to Top

5. Transmit Power

One of the most important WLAN measurements is transmit power. You can measure power in several ways, and different power measurements are required for different WLAN standards. When characterizing an 802.11a/g transmitter, WLAN measurement systems report both peak and average power. For 802.11b devices, typical measurement systems report power ramp-up and ramp-down times as well. Note that while a peak power meter is often a useful tool for measuring power, an RF vector signal analyzer is the fastest option to measure average power on a signal burst. Note that an average power meter can only be used to measure power when the transmitter is configured to output a continuously modulated carrier.

When measuring power with an RF vector signal analyzer, the result is computed over a triggered burst. In this case, you can measure power as the average over an entire burst or over a specific portion of the burst. With the WLAN Analysis Toolkit, you can configure a gated power measurement that measures the power as the average power from a user-defined start time and stop time. In addition, you can use the toolkit to return a power-versus-time trace, as shown in Figure 8, for an IEEE 802.11a/g signal.


Figure 8. Training Sequence, Channel Estimation, and Data in a Power-versus-Time Trace

The power-versus-time trace shown in Figure 8 is frequently used as a debugging tool to ensure that each portion of the burst—training sequence through OFDM symbols—is being transmitted appropriately.

Back to Top

6. Error Vector Magnitude

EVM is one of the most important measurements because it can capture error due to a variety of impairments, including quadrature skew, I/Q gain imbalance, phase noise, and nonlinear distortion. For a modulated signal, the EVM measurement compares the measured phase and amplitude of a signal with the expected phase and amplitude. The NI WLAN Analysis Toolkit calculates it by computing the error vector |E| by the magnitude vector |V|, shown in Figure 9.


Figure 9. Graphical Representation of an EVM Measurement

In general, you can specify EVM either as a percentage (%) or in decibels (dB). However, IEEE 802.11a/g measurements report EVM in decibels, and IEEE 802.11b measurements report EVM in percentages. Equation 1 shows how to convert between the two units.

                                                                     EVMdB = 20log(EVM%/100)

Equation 1. Conversion from Decibels to Percentages

For example, an EVM of 1% is equivalent to -40 dB, and an EVM of 5% is equivalent to -26 dB. When measuring EVM for an entire burst, instruments typically report a root mean squared (RMS) EVM result. For OFDM signals, EVM is calculated as an RMS result over every subcarrier and symbol. For DSSS signals, the result RMS is calculated over every chip.

In many instances, you can visually inspect the EVM performance with a constellation plot. With this trace, which shows the phase and magnitude of each symbol, you can identify specific quadrature impairments. A constellation plot of 64-QAM is illustrated in Figure 10.



Figure 10. Graphical Representation of the EVM Measurement

As shown in Figure 10, an EVM of -46 dB is equivalent to 0.5%. The plot shown in Figure 9 was performed using the NI PXIe-5673 RF vector signal generator and the NI PXIe-5663 RF vector signal analyzer configured in a loopback mode. Both instruments were configured to a center frequency of 2.412 GHz and an RF power level of -10 dBm. At these settings, the instruments therefore have a residual EVM of -46 dB. Note from Figure 9 that the WLAN Analysis Toolkit performs all time-domain measurements in parallel. By configuring a composite measurement, you can evaluate EVM, carrier offset, and carrier leakage as well as quadrature impairments such as I/Q gain imbalance, quadrature skew, and others.

Back to Top

7. Spectrum Mask Measurement

The spectrum mask provides a method to characterize transmitter nonlinearity. In general, you use the spectrum graph as a diagnostic tool to determine whether distortion is present in the signal you are analyzing. Because this is a pass/fail test, you also can report the measurement result as a “spectrum mask margin,” where the “margin” is the power delta, in decibels, of the signal at the most prevalent violating frequency bin. Figure 11 illustrates a spectrum mask measurement on an 802.11b signal.

Figure 11. Spectrum Mask of an 802.11b Signal

IEEE 802.11b signals and IEEE 802.11a/g signals actually use a different spectrum mask. Figure 12 shows the mask for an OFDM 802.11a/g signal.



Figure 12. Spectrum Mask of an 802.11a/g Signal

Figure 13. Spectrum Mask of an 802.11ac Signal

Note that you can use the spectrum mask to evaluate a variety of signal characteristics. For example, transmitter nonlinearity causes the signal sidebands to approach the mask limit. In addition, an improperly configured baseband signal also creates unwanted sidebands on an OFDM signal.

Back to Top

8. Conclusion

As you have observed in this document, you can configure many WLAN measurements using software toolkits. In fact, the WLAN Measurement Suite provides generation and analysis functions for IEEE 802.11a/b/g/n/ac measurements. Using application programming environments such as LabVIEW, LabWindows/CVI, or even .NET, you can configure PXI RF vector signal generators and analyzers to quickly and easily test WLAN products. While these software-defined instruments can test WLAN and many other wireless standards, note that one of the primary benefits of this approach is measurement speed. Read Optimizing WLAN Test Systems for Measurement Speed to learn how to configure WLAN test systems for optimal speed, accuracy, and repeatability.

Back to Top

9. Resources

View additional details on WLAN test systems
View pricing and specs for a pre-configured WLAN Test System

"WiMAX,” “Mobile WiMAX,” “Fixed WiMAX,” and “WiMAX Forum” are trademarks of the WiMAX Forum.
“Bluetooth” is a registered trademark of the Bluetooth SIG

The mark LabWindows is used under a license from Microsoft Corporation. Windows is a registered trademark of Microsoft Corporation in the United States and other countries.

Back to Top

Bookmark & Share


Rate this document

Answered Your Question?
Yes No