Microwave Journal
www.microwavejournal.com/articles/5275-new-techniques-aid-802-11n-and-mimo-testing

New Techniques Aid 802.11n and MIMO Testing

September 14, 2007

For many years, wireless local area network (WLAN) technologies have worked impeccably when they have an unimpeded path from transmission to reception. The real world is neither flat nor empty, however, and physical barriers have long been the nemesis of WLAN performance. Walls, people and other objects routinely impede wireless signals,creating distortions, multipath fading and retransmissions as a consequence—and ultimately hindering signal integrity and throughput.


Those creating and refining wireless standards have long dealt with the effects of refraction by adding to link budgets,using antenna diversity and other costly techniques. New developments, however, have effectively reversed the IEEE course. Contrary to past wireless technologies that used more power to compensate for its effects, new innovations are now taking advantage of multipath conditions to improve throughput.

The forthcoming IEEE 802.11n uses similar modulation formats as 802.11g, but has the added benefit of applying spatial processing to improve network performance. The technique known as Multiple-Input Multiple-Output (MIMO) employs multiple antennas for both transmission and reception to boost the data transmission rate and quality of wireless signals. In doing so, MIMO takes advantage of the various reflections seen by the receiver. These reflections (multipath) cause the transmissions to arrive at the receivers at different times, different amplitudes and different phase characteristics.

There are three MIMO techniques: Spatial Division Mutliplexing (SDM), precoding and diversity coding. A 2x2 MIMO system (two transmitting antennas and two receiving antennas) employing SDM, for example, utilizes multipath by transmitting different data on the two antennas, but using the same encoding. Each receiver hears signals from both transmissions (Tx1 and Tx2), and digital signal processing (DSP) unique to MIMO allows reconstruction of weak signals. The result is better signal integrity and improved throughput.

MIMO test challenges

In the case of conventional communication using a Single-Input, Single-Output (SISO) approach, system performance is easily determined by characterizing transmit power and Error Vector Magnitude (EVM). With MIMO, however, these measurements are not enough. Even if EVM performance is perfect at each transmitter antenna, the performance of the system can still be substandard due to poor propagation channel conditions.

These circumstances lead to several test, characterization and debug questions: How do you test MIMO DSP algorithms and debug errors? How do you characterize the propagation environment of your MIMO system? How do you capture intermittent transmissions in the presence of larger broadcasts? How do you validate your MIMO transmitter’s performance? How do you benchmark competitive 802.11n equipment?

To answer these questions, it is important to evaluate the quality of a received signal for development, installation and maintenance of an RF digital communication apparatus. If there is signal quality degradation, it is necessary to evaluate the space conditions of multiple propagation paths to investigate the cause of the degradation. In communication by MIMO using SDM, a plurality of transmitting antennas sends different signals and a plurality of antennas receives them. A receiver demodulates the signals using information about the propagation paths between the transmitting and receiving antennas, which have different transfer functions. In a 2x2 MIMO system, there are 2x2 of the transfer functions between the transmitting and receiving antennas, which requires computation using all of the transfer functions to recover the data from one transmitting antenna.

There are essentially two sets of MIMO tests to consider. The first are MIMO system measurements, which help system designers understand how antenna placements interact with the transmission environment (propagation channel). These measurements also show how the physical environment impacts the MIMO transmission. The second are more traditional transmitter measurements—such as EVM, Carrier Frequency Error, Orthogonal Frequency Division Multiplexing (OFDM) flatness, output power and spectrum emission mask (SEM)—to validate conformance to the IEEE standard.

These tests extend beyond the standard EVM testing required by SISO and legacy systems. They are necessary, however, to determine optimum placement of the transmitting and receiving antennas, ensure the distinctness of the propagation channels and make certain that a MIMO system operates effectively.

MIMO test techniques

Advanced real-time spectrum analyzers (RTSAs) are now being used to capture and demodulate MIMO signals for test and analysis. Top RTSAs keep pace with high-speed wireless transmissions with 110 MHz real-time bandwidth, 48,000 spectrums per second capture rate and up to 95 dBc spurious-free dynamic range (SFDR). And with unique Digital Phosphor technology, frequency mask triggers and MIMO-specific analysis software, select RTSAs are helping overcome MIMO test challenges.

In the case of a 2x2 MIMO system, two RTSAs are required since each transmitted signal must be captured and demodulated separately. The instruments must also be triggered simultaneously, for which an Arbitrary/Function Generator (AFG) can be employed. Figure 1 illustrates this test setup, with the addition of an Arbitrary Waveform Generator (AWG) in the absence of a live MIMO transmitter. One RTSA acts as Rx1 and is the Master. The second RTSA acts as Rx2 and its reference clock is slaved to Rx1. The AFG is used to synchronously trigger both RTSAs. Using this test setup, designers can effectively create, capture and demodulate a 2x2 MIMO transmission. An RTSA’s MIMO-specific software or integrated triggering capabilities can then compile the two signals on a single instrument and assimilate Rx1 and Rx2 data for comprehensive analysis.



Figure 1. A 2x2 MIMO test setup using real-time spectrum analyzers and an arbitrary/function generator. In certain cases when a MIMO transmitter isn’t available, an arbitrary waveform generator can be employed.

Once the signals are captured, designers can study a number of factors, from both the transmission and reception ends of the broadcast. From a receiver standpoint, O-gram displays on select RTSAs can help designers understand the propagation channel’s performance over time. Users can view the propagation channel’s Transfer Function, Transfogram, Delay Profile and Delayogram. Transfer Function shows how the transmission’s phase and amplitude change over frequency (subcarrier). Transfogram allows the user to see how the propagation channel impacts the amplitude and phase of each subcarrier over the analysis period. Delay Profile illustrates how the packet delay changes over frequency and time. And the Delayogram goes one step further to show how delay spread changes at the receiver over time. Because similarity between propagation paths hinders performance, a wider spread in the Delayogram should lead to better performance in MIMO systems.

From a transmitter standpoint, EVM, constellation display, carrier frequency error and even power in relation to the subcarriers can all be analyzed. EVM measurements are reported for each packet analyzed, providing an index for measuring modulation quality by comparing a demodulated symbol’s actual and ideal magnitude and phase positions in the constellation diagram. After reviewing EVM measurements, the designer will typically check carrier frequency error over time—especially the portion of transients in each packet—subcarrier power and other spectrum-related performance such as occupied bandwidth and spectrum emission.



Figure 2. Real-time spectrum analyzer display showing MIMO transmission measurements.

In addition to these standard measurements, MIMO requires more sophisticated computation and analysis. As the MIMO environment is subject to rapid and dynamic propagation changes, various fading profiles and test conditions must be applied to test system performance and robustness, and the transmission’s ability to adapt to the environment.

With conventional SISO communication, it is relatively easy to evaluate the conditions of the propagation paths by looking at the transmit power and EVM. With MIMO, however, signal demodulation requires complex computation of multiple transfer functions in order to evaluate the conditions of the propagation paths.

There is currently no measurement apparatus that is suitable for this type of evaluation. Fortunately, new algorithms and techniques that facilitate “Transfer Efficiency” measurements are coming to the fore, enabling effective evaluation of multiple propagation paths. Using these techniques, data of one packet is extracted from signals received by the antennas. The symbol synch of the packet is determined and the frequency and phase are compensated. The data is separated every OFDM symbol to extract Long Training Fields (LTFs) from them. The LTF is an OFDM symbol having a fixed pattern and located at the first portion of a packet. Transfer functions of NxN are calculated using the LTFs.

Transfer Efficiency is then calculated with the transfer functions of NxN according to a series of algorithms. Measuring Transfer Efficiency makes it easy to evaluate MIMO propagation paths, which has been difficult with conventional measurement techniques. Computation of the transfer function becomes more complex as the number of antennas increases, but an RTSA can compute it and provide an easy to understand display that eases the investigation of the cause of signal quality degradation.

This analysis—in conjunction with standard measurements such as EVM—leads to improved development, installation and maintenance of RF digital communication devices employing MIMO technologie

Jim Taber is a Product Planner for the Real-Time Spectrum Analyzer group at Tektronix focusing on new measurement solutions for digital RF systems. Prior to his current role, he has held product planning and marketing positions at other test and measurement companies in signal sources and vector network analyzers. Jim holds a BSEE from the University of Missouri – Rolla and masters in the Management of Technology from National Technological University. He can be reached at jim.taber@tek.com.