Experimental and Apparatus History

The original PAN prototype was developed by Tom Zimmerman, used on-off-keying (OOK), and ran at 300 baud. A transmitter and receiver were built, and each was a one-off single task device.

A second design and protoboard implementation was done by Chris Turner.[#ctpan##1#] This design also used OOK, however each device was a transceiver and the devices ran at 2400 baud. The design was based on a series of gain stages, a pair of quadrature integrators followed by digital demodulation in a PIC microprocessor.

Further development of a modified version of Turner's design was carried out by the author. Both OOK and differential binary phase shift keying (BPSK) were used for encoding, with limited success. This design was built on PCBs, however the design ultimately failed due to noise rejection problems. Attempts to resolve this problem through use of an analog gain control (AGC) were made, but the AGC was was also abandoned in favor of the final design.

A final hardware design by Matthew Reynolds, employing a phase-locked-loop (PLL) for analog demodulation was implemented using a frequency shift keying (FSK) encoding scheme. This design, as of May, 1997, has shown the most promise, and has proven reliable for speeds up to 9600 baud in certain geometries.