Understanding adjacent channel leakage ratio (ACLR)

Leakage and adjacent channel leakage ratio (ACLR)

It is not uncommon for some signal power to “leak” beyond the edges of its assigned channel or bandwidth. This leakage is most often caused by intermodulation distortion, which often occurs when active devices such as amplifiers are compressed due to high input power. There are also other causes, such as excessive phase noise or imperfections in the transmitter or system. Leakage is always undesirable because it can create interference for users of adjacent spectrum or channels.

Adjacent channel leakage ratio (ACLR) is the standard way of quantifying this leakage. ACLR is a very important measurement for devices or systems that use wideband, digitally modulated signals. ACLR is also referred to as adjacent channel power ratio (ACPR) or adjacent channel power (ACP).

To measure ACLR, the spectrum is divided into a transmit channel and one or more adjacent channels. The power in each of these channels is then measured. ACLR is defined as a ratio of two powers. The first the power in the transmit channel, and the second is the power in each adjacent channel. In the image below, the shaded box indicates the equivalent channel power after the power is integrated over the channel limits. This difference in power is usually given in dBc, i.e., the number of decibels down from the transmit or “carrier” channel.

Transmit, adjacent and alternate channels

All ACLR measurements require measuring the power in the transmit channel and in at least one adjacent channel. In most cases, adjacent channels are paired and symmetric around the transmit channel. Adjacent channels can be designated as “upper” and “lower” for clarity. Many ACLR measurement standards also specify an additional set of channels above and below the directly adjacent channels, and these are often called “alternate” channels to distinguish them from the channels that are truly “adjacent” to the transmit channel.

Note that channels normally don’t directly touch, and there are small “gaps” between them.

Channel bandwidth and spacing

Channel bandwidth and spacing are defined relative to the center of each channel. In many cases the bandwidth of the transmit channel and the bandwidth of the adjacent and alternate channels are identical. The small gap between the channels occurs when channel spacing is wider than the channel bandwidths.

Note, however, that when configuring ACLR measurements, the distance or spacing of channels is usually specified relative to the center of the transmit channel. That being said, most ACLR measurements are made by selecting a pre-defined standard rather than by manually configuring these parameters.

ACLR measurements

ACLR is measured using a spectrum or signal analyzer. When choosing a spectrum analyzer to measure ACLR, high dynamic range is the most important parameter, since ACLR measurements require accurate measurement capabilities at both very high and very low power levels.

As mentioned above, ACLR measurements involve summing up or integrating the total power in each channel, so ACLR measurements are normally performed using a specialized automatic measurement function. Results are displayed both graphically and numerically.

When it comes to configuring ACLR measurements, some wireless communications standards specify measurement parameters. With some analyzers, you can configure the settings by loading a setup file that specifies the parameters. They may also be able to automatically configure optimal settings based on a user-defined channel configuration.

If manual configuration is required, there are some standard guidelines regarding measurement parameters:

• Span should be wide enough to cover all channels, with an additional recommended margin of 10%.
• Resolution bandwidth should be set to 1-4% of the channel bandwidth. Resolution bandwidth involves a tradeoff between speed and accuracy: a narrow bandwidth means higher accuracy but slower sweep times.
• Video bandwidth should be set to at least three times the resolution bandwidth.
• A root mean square (RMS) detector should be used, since ACLR is a power measurement.
• Increasing the sweep time is a good way to improve measurement reproducibility, although it will of course also increase the overall measurement time.
• Noise cancellation should be enabled, if the analyzer supports it.

Example of ACLR measurement results

Here is an example of how ACLR measurements are typically displayed:

• A graphical view shows the measured power over the span, with the transmit, adjacent and alternate channel boundaries displayed.
• Shaded bars show the measured power for each channel, and it is these powers that are displayed numerically.
• The bandwidth of each channel is given along with the offset for adjacent and alternate channels.
• The power in the transmit channel is always given in absolute units of dBm, since this power will be used when computing the ratio of channel power to “leaked” power.
• For adjacent and alternate channels, ACLR is calculated and displayed in units of dBc, or dB down from the transmit “carrier.”
• If any channel limits are violated, this is indicated using a different color or sign for easy identification.