Weak signal modes

Introduction

In 1997, CEPT/ERC Recommendation 62-01 E gave radio amateurs access to the 2200 m band. At this wavelength amateur sized antennas are inefficient, resulting in very weak signals. In addition noise levels are very high. As a result the signal to noise ratio (SNR) of ham signals on this band is often very low. This limited the distances that could be covered in CW, the only mode used in the early days.
In most cases it was not possible to increase the ERP (that is limited to 1 W anyway), so the most obvious way to improve the SNR is to reduce the receiver bandwidth. This results in less unwanted signals and noise while leaving the level of the wanted signal unchanged. But the reception of any signal requires a minimal receiver bandwidth depending on the type of modulation. SSB has a typical bandwidth of 2.4 kHz, the bandwidth of a CW signal depends on the speed but is in any (practical) case less than 100 Hz. Using a RX filter with a narrower RX bandwidth than that of the transmitted signal will distort the received signal.
So if we want to improve SNR by reducing the RX bandwidth we will have to reduce the bandwidth of the transmitted signal as well. In CW this is done by reducing the speed. This lead to the use of narrow bandwidth modes such as very slow CW (QRSS) and its variant DFCW (dual frequency CW).
At a later stage new weak signal modes were developed, using FEC (Forward Error Correction) to improve the SNR. These are full digital modes at a rather high bit rate and thus bandwidth, compared to QRSS. This lead to a bunch of new modes, some of them meanwhile well known such as WSPR and JT9, others less known or a bit forgotten such as WOLF.
In this article I will try to give an overview of these weak signal modes and their developments, more or less in a chronological order.