The search for extraterrestrial intelligence has sought radio beacons devoid of information content. It seems likely, however, that a civilization transmitting a radio signal intended for our detection will also be motivated to embed information within the signal, especially in view of the large speed-of-light latencies. Successful exchange of information by radio with intelligent civilizations in distant solar systems requires an understanding of the end-to-end communication system design, including resources available to transmitter and receiver and properties of radio propagation in the interstellar medium. Although interstellar space is nearly an ideal vacuum, it contains sufficient low- density plasma to profoundly affect radio transmission over interstellar distances. The primary impairments are attenuation, thermal noise, plasma dispersion, scattering, and interference in the vicinity of the receiver. The most difficult technical challenge is initial discovery of a signal, and the primary obstacles are the infeasibility of coordination between transmitter and receiver and related “needle in a haystack” issues. Impairments are actually helpful as an implicit form of coordination through constraining design choices as well as constraining the size of the “haystack”. In this talk, Dr. Messerschmitt will address end-to- end communication system design emphasizing noise, dispersion, and interference, deferring scattering to future work. He will show that an effective means of countering interference without compromising noise immunity is spread spectrum signaling, and proceed to characterize the effect of plasma dispersion upon these broadband signals. The conclusion is that while design considerations provide guidance as to carrier frequencies and bandwidth and time duration of signals, there is also a demonstrated tradeoff between transmit power and the computational burden placed on the receiver.