Analog-to-Digital Converter Loading Analysis and Overloading Considerations for Communications Systems

Abstract

In a satellite using an analog-to-digital converter (ADC) in the communication path, a common method of analyzing the uplink performance between ground terminals and a satellite receiver is to examine the ADC loading curve. Such a loading analysis considers how the load factor (LF) influences the aggregate input signal to noise ratio (SNR). This conventional treatment ignores what is happening to a small signal of interest (SOI). What is missing is the relationship between the signal to noise ratio of the SOI (SNRSOI) and the LF. This paper provides novel and unique in-depth analyses and simulations to determine this SNRSOI versus LF relationship and how this small SOI link precipitously fails to close when the ADC is even slightly overloaded. It is clearly shown that the SNRSOI drops off even more rapidly than the SNR as the ADC becomes overloaded. Significantly, this paper shows how large signals into the ADC create noise that the SOI must contend with; spectral plots are shown that can be used to visualize that the SOI must be adequately above the noise created by the larger signals that lie in the SOI band. Furthermore BER curves for the SOI are generated to demonstrate precisely its link closure performance. Notably, it is clearly shown that a variable attenuator can be added in front of the ADC to prevent this catastrophic drop-off in the SNRSOI. Also, it is shown for the first time that the BER can be easily predicted for the case where the only noise in the link is quantization noise, since when filtered this noise is Gaussian-like. This is shown both theoretically and by simulation.