Analytical comparisons of CSSB and TDMA/DSI satellite transmission and techniques

At the same time that Time-Division Multiple Access (TDMA) with Digital Speech Interpolation (DSI) is gaining wide acceptance, being planned for almost every modern satellite communication network as the basic transmission technique, a variation of the classical single sideband using companders (CSSB) has recently been proposed, accompanied by the claim of higher transponder capacities. The higher capacity of CSSB is supported under a variety of assumptions (lower speech levels, increased transponder linearity, benign interference environment) that hinders the comparison of CSSB with other well-established satellite transmission techniques such as FDM/FM or TDM/D51. The purpose of this paper is to "normalize" CSSB to a common set of assumptions that permits its capacity to be calculated on the same bases that apply to other satellite transmission methods, and in this manner to identify the ranges of ground and space segment parameters that would render CSSB superior to existing alternatives. In particular, CSSB is compared with TDMA/DSI, using analytical approaches that permit the simultaneous consideration of the principal factors involved in the capacity of both systems. These factors are • speech and compander characteristics • ground segment parameters, G/T, and antenna sidelobe radiation • space segment parameters, in particular EIRP, the transponder nonlinearities, and orbital spacing. These factors have been quantified for a typical 36-MHz C-band transponder with EIRP--34 dBW. The results show that CSSB's claims of higher capacity are realizable only under idealized transponder characteristics or with larger ground segment antenna sizes. In addition, CSSB's performance degrades faster than TDMA/DSI under fading conditions, suggesting that margins must be subsequently improved for this technique.