How to Use Fixed Gain Mode on LabSat 4 (RP) to Evaluate the Performance of Different Antennas

What is Fixed Gain Mode?

By default, LabSat 4 utilises an AGC (Automatic Gain Control) system when recording. This system automatically adjusts the internal gain in LabSat 4 to provide optimal recording quality for the input signal. While it results in a more flexible and user-friendly test system, it also affects the normalising of the signal power. 

The power level during a replay will be approximately the same regardless of the input power during recording. Since each RF channel AGC operates independently, differences in the power levels between the channels are also lost due to this normalisation effect. This is not significant for most use cases. If you require a specific output level, you can use LabSat 4's internal attenuation control or external attenuators to set the output power. If you require different powers between channels, you can set the internal attenuation on a per-channel basis.

The Manual Gain Mode in LabSat 4 adds the ability to disable the AGC and fix the input gain to a user-specified level. This will give the system the ability to accurately reproduce differences in signal levels in a recording or to maintain the correct relative power levels between two different recordings. This can be helpful for certain testing applications, such as in a comparison between different antennas.

Advantages and Disadvantages of Fixed Gain Mode

Test Setup

The following illustration is an example system that consists of an antenna that is to be tested, an RF splitter, a u-blox F9P GNSS receiver, and a LabSat 4.

The u-blox receiver, in common with virtually all GNSS receivers, contains its own AGC system for differing input signal levels. It was configured to output both the final GNSS signal SNR values and also the internal AGC level. This AGC level is reported as a percentage of maximum.

Each test consists of logging the u-blox output while simultaneously recording the signal with LabSat 4. The LabSat 4 recording can then be replayed to the u-blox while logging the output during the replay

Analyse the logs to compare the AGC levels and GNSS SNR values that are reported by the F9P when receiving the live sky signal and when receiving the LabSat 4 replay.

Run the test twice with each antenna you want to test, once with LabSat 4 in the default AGC mode and once with LabSat 4 in fixed gain mode.

In our example, we tested three different antenna configurations; a multi-band passive antenna, a multi-band passive antenna plus an in-line amplifier, and a multi-band active antenna.

When we replayed the LabSat 4 recordings, we set the internal attenuation so that when we replayed the fixed gain passive antenna recording the output power was similar to that seen from the live signal. We then used this attenuation setting for all the replays.

Results

AGC Percentage (L1/L5)

	LabSat AGC Enabled		LabSat Fixed Gain
Antenna	Live signal	Replay	Live signal	Replay
Passive	64.3/64.3	42.9/60	64.3/64.3	60.0/60.0
Passive + amplifier	51.4/51.4	26.9/60	51.4/51.4	47.1/45.2
Active	34.3/21.4	25.7/51.4	34.3/21.4	34.3/21.4

As is to be expected, the live sky values for the two data sets are very similar. For simplicity, the charts below only include these values once.

While the fixed gain replay doesn’t result in a perfect reproduction of the live sky characteristics, it does provide a far more accurate reproduction of the differences between the antennas.

The reason for these differences could be due to a difference between the LabSat 4 bandwidth setting and the F9Ps internal bandwidths used in setting AGC levels.

Analysis of Bit Depth and Dynamic Range

These tests were performed with a bandwidth of 48 MHz and a quantization of 8 bits. We used a gain level of 55 dB for the fixed gain recordings. Note that while this was found to be an effective combination of settings for this specific test and equipment, the optimal setting will depend on the final system and configuration.

When you are setting the value for a fixed gain system, one aim is to maximise the available dynamic range while avoiding the risk of slipping or distorting the recorded signal. By analysing the data in the recorded .ls4 file, you can determine how well the available range is being utilised.
The aim is to keep the recorded signal as strong as possible without risking exceeding the maximum value that can be recorded. Statistically, this works out as aiming for the average signal to be around 20 dB below the maximum.

Analysing a short period of the recordings yields the following results:

As you can see, the AGC with active antenna recording was very close to the ideal. Both of the other recordings with the AGC enabled are a little weaker than ideal. This is an indication that the input signals were at the lower end of the range that LabSat 4 is designed for but that the system managed to record them with reasonable fidelity.

The fixed gain with active antenna recording shows a very similar sample distribution to the active antenna with AGC settings. This is by design. The AGC aims to set the highest gain that can reasonably be used without risking distortion. When you are setting a manual gain, the ideal value is the one that provides the same results as the AGC when it receives the maximum expected input signal.

For the other input sources with fixed gain, you can see that the recorded signal levels are significantly lower, reflecting the weaker nature of the signals. In the case of the passive antenna recording, the top 4 bits of the quantization are never used. This demonstrates why fixed gain recordings are only practical when used with larger bit depths. If this recording had been made with 4 bits rather than 8, the fixed gain passive antenna recording would have become the equivalent of a 1-bit recording. If so, the system would not have been able to accurately reproduce any further decrease in signal level and the passive antenna and passive antenna with amplifier recordings would have produced very similar output powers.