Abstract: An analog-to-digital converter includes a filter having an input for receiving, via a summing node, a signal derived from the analog input signal. A first quantizer produces a first digital signal from a signal derived from the output of the filter. A feedback loop feeds a signal derived from the first digital signal to be combined in analog form with the signal derived from the analog input signal at the summing node, so that the filter in use receives an error signal representing the difference between the signal derived from the analog input signal and the analog representation of the first digital signal. An error signal filter filters the error signal to remove noise outside the passband of the analog-to-digital converter. A second quantizer produces a second digital signal representative of the filtered error signal, and a combining circuitry combines the first and second digital signals to produce a digital output signal representative of the analog input signal.

Abstract: FMCW radar return signals are converted to i.f. in a front end section, digitised in an A-D converter, and reduced to in-phase and quadrature base band components at a lower sampling rate in a filter section. Returns are compared with a separate reference waveform signal from a generator for each range cell by a multiplier and an accumulate and dump function. The reference waveforms for subsequent range cells are delayed by one sample period. The provision of separate correlation of de-ramping for each range cell enables non-linear sweep wave forms to be employed, whereas the conventional method using a single Fourier Transform for all range cells simultaneously restricts operation to a linear sweep waveform.

Abstract: An analogue-to-digital converter in accordance with the invention comprises a feed-forward arrangement in which, in a first stage, the analogue input signal is applied via a filter to a subtractor along a first path and is applied via a second path to an analogue-to-digital converter followed by a digital filter and then converted back into analogue form to be applied via a bandpass filter to the subtractor. The output of the subtractor is therefore an error signal representative of the difference between the analogue input signal and the predicted signal in analogue form. The error signal is applied to an analogue-to-digital converter and is combined with the predicted signal in digital form taken from the output of the digital filter via another filter to produce the digital output signal. Control means are provided to provide phase and gain matching between different parts of the circuit.

Abstract: A narrow band signal is converted into digital form using an analogue to digital converter of relatively small capacity. The input signal is compared with a prediction of it based on previous cycles of the incoming waveform. This produces an error signal which is compared with a prediction of the error signal to produce an "error of the error" signal. A relatively small A/D converter converts this into 8 bit digital form and this is applied to a feedback loop consisting of an adder and delay to give a sixteen bit prediction of the error signal which is applied via D/A converter to the subtractor. The sixteen bit prediction also provides an input to a further adder forming part of a feedback loop to give an eighteen bit prediction of the input signal. The latter is applied via a D/A converter to a subtractor and is used as the system output.