Autonomous loop switching: Interpreting and modifying the internal state of feedback tracking loops

Abstract

Receiver tracking loops are implemented in software in modern space-borne radios. Software implementation allows loop designs to be modified in flight. Not only filter coefficients and gain, but also loop order and type can be modified. This flexibility enables new cognitive and autonomous capabilities. Loop designs can be optimized for each mission phase, and separate loops can be used for acquisition and tracking. Furthermore, the loops can be automatically adapted based on changing signal dynamics or SNR. However, if the loop has tracked away from its quiescent state, the loop will lose lock when the switch occurs unless the internal state of the loop is translated appropriately. This paper considers the internal state of feedback tracking loops. In particular, a physical interpretation of loop state is derived that enables translating the loop state from one design to another. Limitations to autonomous switching, including high-order signal states and noise, are described, and several examples are simulated. Practical applications for both nearearth and deep-space missions are discussed.

Digital signal processing architecture design for gate array based software defined radios

“…This capability enables the software layer to detect an operating condition (e.g. carrier lock or SNR) and react by changing control loop parameters, or by switching the microcode instruction set for a particular event execution cycle to completely swap out the whole control loop [13]. This feature allows a receiver signal acquisition loop could be designed separately from a tracking loop.…”

“…This allows the user to minimize power or maximize capability as needed (i.e. make the power/capability trade in software) 13 . Some of the originally architected receiver path circuitry did not fit into the RTAX2000 device.…”

“…Core nicknames,Table 2, are used from this point forward 13. In the particular Frontier Radio instances, to be discussed later, most of the circuitry for the receive path required a similar clock rate, so all of the cores used a common low rate clock.…”