Implementation

Implementing our design was fast, because the circuits turned out to be very simple. The stages of the hardware implementation were as follows:

1.

We got our transmitter and receiver working on a single breadboard, sharing clocks, power supplies, and transmission by wire. Our input data stream was the state of a pushbutton, and our output was a red LED. At this point we kept the clock rate fairly low.

2.

We implemented the circuit on separate breadboards and capacitively coupled the transmission line to the receiver, thus allowing the use of separate power supplies. Both transmitter and receiver still shared clock signal. Clock rates above the f3db point of our coupling worked fine. We tested the circuits up to 500kHz.

3.

We used separate clocks for both transmitter and receiver (two function generators plugged into different power strips and separated by several feet to prevent accidental entrainment) in order to see how well the system performed with incompletely synchronized clocks. At this point we were using a single long wire as our transmission line, which undoubtedly introduced some RF noise. Even with some clock drift, the system continued to work very well, and could be reliably synchronized by hand at 500kHz.

Stages we didn't get to:

1.

Implementing a phase-locked loop or other auto-synchronizing mechanism for the LFSR clocks, although we did come up with a proposed schematic for doing this.

2.

Sending more interesting data, such as digitized audio.

3.

Microcontroler or microprocessor based strong encryption front end.