In a scheme to increase the overall data return from a phase-modulation, suppressed-carrier telemetry transmitter, the telemetry playback symbol rate is adjusted essentially continuously. More precisely, the playback symbol rate is adjusted frequently (as often as once per symbol period) in small increments. The adjustment of the rate is made in accordance with the principle that the supportable data rate at any given instant is a function of the instantaneous total signal-power-to-noise spectral-density ratio (PT/N0) at the receiver. The scheme was devised for transmission of telemetric data from deep-space missions, but could also be applied to satellites in orbit around the Earth.
PT/N0 is a known function of the receiving-station gain-to-noise-temperature ratio (G/T), which, in turn, is a known function of the position of the transmitter relative to that of the receiver. Thus, once the trajectory of the transmitter relative to the receiver position has been predicted, then G/T or PT/N0 as a function of time can be estimated (see figure); this estimate can be computed aboard the spacecraft or uploaded to the spacecraft prior to a portion of a trajectory (a tracking pass) and the corresponding interval of time during which the telemetry signal is expected to be received. Then during the tracking pass, the playback symbol rate at the transmitter is adjusted in accordance with the predicted G/T.
Traditionally, the telemetry symbol rate (equivalently, the telemetry data rate) for a deep-space mission is either fixed or is changed only a few times during a pass. A change in the data rate often entails a large, instantaneous jump, which can sometimes cause the ground receiving system to lose symbol lock. Upon losing lock, the receiver begins to lose some or all of the incoming telemetry data and must go through a signal reacquisition process in order to regain lock. The time lost in reacquisition could be large enough to nullify any advantage gained by changing the data rate. Although there are ways to minimize the probability of losing receiver lock during data-rate changes, they are operationally complicated and often require very precise predictions of the times when the rate changes will take place.
In the present scheme, a precise prediction of the instantaneous data rate is not needed. [However, coarse data-rate predictions are needed for initial acquisition and for reacquisition after mode changes (not to be confused with rate changes).] In this scheme, the symbol clock in the transmitter has a continuous phase; in other words, even when the clock frequency changes abruptly, the clock phase remains continuous. Because of the continuity of phase and the smallness of the clock-frequency increments, the symbol-tracking loop in the receiver is subjected to only small transient phase errors that do not cause it to lose lock. Experiments and computational simulations for some typical cases have shown that a receiver can, indeed, track the small frequent rate changes and that the telemetry returns achievable by use of this scheme exceed, by 1 to 2 dB, those achievable by use of the best-single-rate scheme.
This work was done by Miles K. Sue, Jeff B. Berner, Selahattin Kayalar, and Henry Hotz of Caltech; Peter Kinman of Case Western Reserve University; and Harry Tan of Q-Plus for NASA's Jet Propulsion Laboratory. For further information, access the Technical Support Package (TSP) free on-line at www.nasatech.com/tsp under the Electronics & Computers category.