Engineering topics
Publications and source records attributed to Salter, W. E..
Report presents very thorough and rigorous analysis of bandwidths and filters required to achieve a specific error. Included is an investigation of contribution to error of 3- and 6-pole Chebyschev, Butterworth, and Bessel filters. Also included is methodology for designing DSB/FM telemetry system to make optimum use of available bandwidth while reducing error toward its theoretical minimum.
A tape record/playback system was tested by first deriving an analog test signal and a band-limited digital reference signal from a pseudo-noise sequence generator driven by a clock signal. It recorded the signals on respective tracks of the system during operation in a record mode. During the playback mode of operation of the system, a delayed analog reference signal without time base variations was reconstructed from the played back reference signal. It was compared with the played back test signal in order to obtain an error signal that was a measure of the performance of the system.
A test set for communications systems is described which includes a pseudo noise sequence generator providing a test signal that is fed to a pair of signal channels. The first channel includes a spectrum shaping filter and a conditioning amplifier. The second channel includes a variable delay circuit, a spectrum shaping filter matched to the first filter, and an amplifier. The output of the first channel was applied to the system under test. The output of the system and the output of the second channel are compared to determine the degree of distortion suffered by the test signal due to the communications system.
Telemetry bands are being reassigned to UHF at 1500 and 2200 MHz. Conversion primarily requires changes in equipment used in RF link, while many of same subcarrier oscillators, mixer amplifiers, and frequency discriminators can be used.
Preemphasis schedules are given for 11 constant-bandwidth FM subcarriers modulating an S-band transmitter at various receiver SNRs (9 dB, 15 dB and 25 dB). The criterion for establishing these preemphasis curves is the achievement, at various receiver IF SNR, of equal receiver output SNR for all channels. The empirically derived results are compared with a simplified, analytically derived schedule and the primary differences are explained. The S-band preemphasis schedule is also found to differ from the lower frequency VHF case.
A pseudonoise (PN) test set was built to provide a relatively easy means of accurately determining the end-to-end rms error introduced by a communication system when subjected to wideband data. It utilizes a filtered pseudorandom sequence generator as a wideband data source, providing a convenient means for digitally delaying the input reference signal for comparison with the distorted test communication system output. In addition to providing a means to measure the end-to-end rms error and the average delay of a communication system, the PN test set also provides a means to determine the tested system's impulse response and correlation function. The theory of PN testing is discussed in detail along with the most difficult aspects of implementation, the building of matched filter pairs. Both analytical and empirical results are reported which support the contentions that this is an accurate and practical way to acquire figures of merit for complete communication systems.
End-to-end root-mean-square (rms) tests performed on a constant bandwidth FM/FM system with various settings of system parameters are reported. The testing technique employed is that of sampling, digitizing, delaying, and comparing the analog input against the sampled and digitized corresponding output. Total system error is determined by fully loading all channels with band-limited noise and conducting end-to-end rms error tests on one channel. Tests are also conducted with and without a transmission link and plots of rms errors versus receiver signal-to-noise (S/N) values are obtained. The combined effects of intermodulation, adjacent channel crosstalk, and residual system noise are determined as well as the single channel distortion of the system.
Pre-emphasis schedules are given for 11 constant-bandwidth FM subcarriers modulating an S band transmitter at three receiver signal to noise ratios (i.e., 9, 15, and 25 dB). The criterion for establishing these pre-emphasis curves is the achievement, at various receiver intermediate frequency signal to noise ratios, of equal receiver output signal to noise ratios for all channels. It is realized that these curves may not be the optimum pre-emphasis curves based on overall efficiency or maximum utilization of the allotted spectrum, but they are near-optimum for data with channels which require equal output signal to noise ratios, such as spectral densities. The empirically derived results are compared with a simplified, analytically derived schedule and the primary differences are explained. The S band pre-emphasis schedule differs from the lower frequency VHF case. Since most proportional bandwidth and constant bandwidth systems use ground based recorders and some use flight recorders (as the Saturn systems did on VHF proportional bandwidth telemetry), the effects of these recorders are discussed and a modified pre-emphasis schedule is presented showing the results of this study phase.
Waveform distortion resulting from adjacent channel crosstalk and from amplitude and phase nonlinearity in channel filters in DSB/SC telemetry link