High-data-rate communications as defined in the WiMAX IEEE 802.16-2004 standard may pave the way for true broadband, multimedia services over wireless networks. Based on orthogonal-frequency-division-multiplex (OFDM) techniques, the WiMAX physical-layer (PHY) and media-access-control (MAC) protocols are outlined in the IEEE 802.16-2004 standard. These protocols have inspired the development of a baseband test transceiver detailed last month in Part 2 of this article series. In this final installment of the article series, some of the different types of measurements possible with the test transceiver will be outlined.

The WiMAX Forum (www.wimaxforum.org) is dedicated to WiMAX interoperability and, as such, is interested in promoting the IEEE 802.16-2004 standard. The group has selected OFDM-256 as an interface format for frequencies below 11 GHz, and that format is the basis for this article series' measurement endeavors. The OFDM-256 air interface provides adaptive functions as well as many optional features. It allows for adaptation of modulation/coding format and adaptation of the cyclic prefix. But in spite of the many optional features, the proposed measurement receiver and test algorithms will work over a wide range of frequencies, bandwidths, and option sets.

The proposed WiMAX measurement receiver (see Fig. 9 in Part 2) integrated the key function blocks needed for baseband WiMAX testing, including a packet detection unit, coarse and fine symbol timing estimation, an inphase/quadrature (I/Q) impairment estimator, frequency-offset correction, channel equalization and estimation, and automatic modulation detection. Employing Fast Fourier Transform (FFT) techniques, the test transceiver employs algorithms specifically designed for making measurements, rather than for operating within a communications network. Part 2 of this article series offered a more detailed look at the different function blocks within the test transceiver and the mathematical basis for some of the estimators.

The IEEE 802.16-2004 standard has provided some transmitter and receiver test requirements. These requirements need to be tested for and conformance demonstrated. Therefore, in the proposed test design, the authors have incorporated ways to measure the requirements enforced in the standard. The following measurement capabilities have been incorporated into the receiver in addition to the ones that will be discussed in more detail shortly:

• Relative constellation error (RCE) in percent or decibels (dB).
• RCE versus symbol number.
• RCE versus subcarrier number.
• Spectral flatness.
• Crest factor (a measure of the peak-to-average power ratio).
• Peak, average, and minimum error vector magnitude (EVM).
• Error vector spectrum/time, including rootmean-square (RMS) error vector.

This list contains standard measurements that are widely discussed in the literature. Other suppliers of test equipment, such as Agilent Technologies (www.agilent.com), Anritsu Co. (www.us.anritsu.com), and Rohde & Schwarz (www.rohde-schwarz.com) are also providing such measurement capabilities in their WiMAX receiver solutions, often in the form of programmable receivers using software-defined-radio (SDR) architectures. Since these measurements are well documented in the literature and in the standard document for 802.16-2004, they will not be covered here. Instead, the following additional measurements will be briefly detailed.

Frequency error is a measurement of the difference of the carrier frequencies generated by the local oscillators at the transmitter and receiver. Rather than the absolute frequency error, the normalized frequency error is a more appropriate value. The frequency error is often measured from the time-domain signal. However, it is also possible to measure the frequency error using frequency-domain samples. In our implementation, we have used time-domain samples for the measurement as described in the previous section.

The sample clock error measurement determines the sampling clock difference at the transmitter and receiver. This measurement in OFDM system is often performed during the pilot tracking period. Since sampling clock introduces a phase rotation that depends on the carrier and OFDM symbol index, variation of the rotation can be used to estimate the sample clock error. Sample clock error is popularly estimated using frequency-domain samples after channel equalization.

Received-signal-strength-indication (RSSI) estimation provides a simple indication of how strong the signal is at the receiver front end. If the received signal strength is stronger than the threshold value, then the link is considered to be good. Compared to other measurements like CINR and BER, RSSI estimation is simple and computationally less complex, as it does not require the processing and demodulation of the received samples. However, the received signal includes noise, interference, and other channel impairments. Therefore, receiving a good signal strength does not tell much about the channel and the signal quality. Instead, it gives an indication whether a strong signal is present or not in the channel of interest.

If the measurement is performed on a wireless channel with a portable measurement device, the received signal power fluctuates rapidly due to fading. In order to obtain reliable estimates, the signal needs to be averaged over a time window to compensate for short-term fluctuations. The averaging window size depends on the system, application, variation of the channel, etc. For example, if multiple receiver antennas are involved at the receiver, the window can be shorter compared to a single antenna receiver. For measurements with a cable connected between the DUT and receiver, this is not an issue. Therefore, even short window of measurements can provide reliable RSSI values.

Carrier-to-interference-ratio (CIR), carrier-to-interference-plus-noise-ratio (CINR), signal-to-interference-ratio (SIR), signal-to-noise-ratio (SNR), and signal-to-interference-plus-noise-ratio (SINR) are the most common ways of measuring the channel quality during (or just after) the demodulation of the received signal. CINR (or SNR, or SINR) provides information on how strong the desired signal is compared to the interferer (or noise, or interference plus noise). Most wireless-communication systems are interference-limited, therefore, CIR and CINR are more commonly used. Compared to RSSI, these measurements provide more accurate and reliable estimates at the expense of computational complexity and with additional delay.

CINR estimation can be employed by estimating signal power and interference power separately and then by taking the ratio of these two. The channel parameter estimates can be used to calculate the signal power. A version of EVM measures the error between what is received and what was expected and can be used for noise-plus-interference-power measurement.