AX25, IL2P & FX.25

IL2P

Improved Layer 2 Protocol is extremely robust in comparison to FX25, but that is at the expense of dropping backwards compatibility with standard AX25 packets. So IL2P is often run on an additional port on SSB radios at a different center frequency. When setting IL2P mode, make sure it is set to IL2P Only, and that and FX25 options are disabled for the IL2P modem.

Notes from the QTSM developer:

As the latest QtSM supports IL2P I've added a tag of the mode the packet was received in.  Values are AX25, FX25 (packet corrected by fec, not just sent using FX25) or IL2P.
MEM is added if the packet has been corrected by Memory ARQ (where sound samples from retries are combined to form a good packet).

# Indicates this decoder was corrected by MEM ARQ. # Has always been there, but adding the MEM tag is new.

FX.25

Ensure FX.25 is enabled for both receive and transmit.

FX.25 adds forward error correction by wrapping the standard AFSK AX25 packet in such a way that is still 100% compatible with older hardware and software decoders that do not have FX.25 capability.

For FX.25 capable decoders, FX.25 improves throughput in noisy conditions as well as some moderate improvement during signal fade (QSB)– reducing the need for retransmission requests and increasing channel throughput in unidirectional environments”, as per the paper published here (mirrored attachment supplied below).

The Setup used on 10/20 & 40m

Dual FX.25 and single IL2P
See our QRG pages for detailed info.

Additional decoder pairs

Enable additional “decoder pairs”- recommendation 1 or 2 per port for SSB operation. FM does not benefit from additional pairs, so should be set to 0 for FM.

The modern day AFSK decoder such as UZ7HO Soundmodem, QtSoundModem and Direwolf is capable of running multiple parallel decoders that are tuned a few tens of Hertz above and below the main decoder center frequency. These extra pairs aid in situations where a remote station may appear to be slightly off frequency, which is quite common due to the nature of some older and/or aging radios.

Adding additional pairs of decoders does come at the expense of additional workload on the system.

Audio Levels & Settings

Input:

Ensure that your sound modem is not being overdriven by the radio- the waterfall in UZ7HO and QtSoundModem can be helpful in determining the input audio level. The majority of the waterfall should be green with signals/packet traces of transmissions appearing in yellow. If you are having trouble decoding packets that are clearly above the noise level then overdriving the microphone input of the computer’s sound card is likely the problem.

Output:

Keep an eye on the radio’s ALC meter, while transmitting a packet the meter should not really move, turn the operating system’s audio up until the ALC starts to move then back the slider down a little so that is is barely moving.

Ensure there are no audio DSP filters, digital noise reduction or microphone compressors engaged when using packet as it can cause severe splatter to adjacent users as well as unevenly distribute the power between the AFSK tones used for packet operation.

Also ensure that the operating system is not interfering with your outbound audio, such as “3D sound” or similar type DSP filtering. Some Windows versions have “enhanced sound” enabled by default, so will need to be disabled manually.

DCD:

Digital Carrier Detect is usually in the form of a slider in the graphical versions of Soundmodem, this can be set by default to approximately 70% and then adjusted as necessary. The aim of the DCD is to prevent your station from transmitting over the top of an incoming packet.

TX Delay:

The additional spacer tones added to the header of packets to compensate for radio transmit and receive switchover delay. It is commonly set between 300-500 milliseconds.

TX Tail:

The additional spacer tones that form the footer of the packet- commonly set by default at 50 milliseconds.