FEC-A uses a convolutional error correction scheme based on data bits being read into a shift register, the length of which may be changed. Values of 72 and 128 bits are common.
|
Parameter |
Value |
|
Frequency range |
HF |
|
Operation modes |
Broadcast/Simplex FEC |
|
Modulation |
FSK |
|
Symbol rate |
96, 144 and 192 Bd, variable 30-650 Bd |
|
Receiver settings |
DATA, CW, LSB or USB |
|
Input format(s) |
AF, IF |
|
Additional Info |
ARQ1A with convolutional coding |
The FEC-A mode is started by selecting a standard baudrate from the Baudrate item in the Demodulator menu. Frequency shift and baud rate are determined using FSK Analysis. For automatic tuning Auto may be selected.
The synchronization or idle state is easily recognized by its sound. It is an alternating mark-space keying sequence (mark-space ratio approx. 40% - 60%).
Shift register length is set using the S-Reg item in the Options menu. Incorrect selection of the S-Reg parameter causes incorrect error correction,and the data output rapidly becomes corrupted. If error correction is disabled (ECC off), the length of the shift register will not affect decoding (S-Reg not visible). This feature allows any FEC-A signal to be decoded.
FEC-A will detect and correct transmission errors up to a certain limit; in the case of extreme interference, error correction may worsen the situation, so reception without error correction may improve performance.
FEC-A uses the ARQ-1A alphabet. Every second bit of the bit stream is used for the convolutional error correction and thus each codeword consists of 14 bits.