Practical Spread Spectrum: An Experimental Transmitted-Reference Data Modem

Direct-sequence spread-spectrum systems may be classified into two broad categories: stored-reference and transmitted-reference systems. The circuits I have described so far in these colu m n ~ ~ ~ ~ ~ ~ ~ ~ have all been of the storedreference kind, in that a replica of the pseudonoise sequence used at the transmitter site was also stored at the receiver. The main problem then facing the designer was to build into the receiver a circuit which would (a) extract the clock from the transmitted signal, and (b) synchronize the two identical, but out of phase, sequences. A radically different approach consists in transmitting simultaneously on a frequency F1 the PN sequence XORed with the data, and on a frequency F2 the PN sequence alone. If the receiver uses a second intermediate frequency (2nd IF) equal to (Fl-F2), it is possible to modulo2 add the two signals and recover the original data5. This approach has a major advantage: its simplicity. It also has two main drawbacks for military applications: It is easy to either jam the receiver by transmitting a carrier equal to the IF, or for unauthorized listeners to decode the signal with a very simple receiver. Neither of these problems is of concern to the radio amateur, and it was therefore decided to design a simple data link using the transmitted reference approach. The description of the circuitry used will be brief, as most of its elements have already been described in some of my previous articles. (This paper does not aim to be a "construction article," but to give enough pointers to the interested that they may decide to use this information as a starting point, and try their hands at experimenting with spread spectrum.)