Showing papers in "IEEE Transactions on Communications in 1970"
TL;DR: In this paper, the authors proposed a stationary HF ionospheric channel model, where the input (transmitted) signal feeds an ideal delay line and is delivered at several taps with adjustable delays, one for each resolvable ionosphere modal component.
Abstract: Specially designed HF ionospheric propagation measurements were made and analyzed to confirm the validity and bandwidth limitations of a proposed stationary HF ionospheric channel model. In the model, the input (transmitted) signal feeds an ideal delay line and is delivered at several taps with adjustable delays, one for each resolvable ionospheric modal component. Each delayed signal is modulated in amplitude and phase by a baseband tap-gain function, and the delayed and modulated signals are summed (with additive noise) to form the output (received) signal. Statistical specifications for the tap-gain functions involved three hypotheses: 1) that each tap-gain function is a complexGaussian process that produces Rayleigh fading, 2) that the tapgain functions are independent, and 3) that each tap-gain function has a spectrum that in general is the sum of two Gaussian functions of frequency, one for each magnetoionic component. Statistical tests were performed on daytime and nighttime measurements confirming the validity of the three hypotheses, and thereby the validity of the model. For practical applications, the model can be considered valid over a bandwidth equal to about one fourth of the reciprocal of the effective (weighted) time spreads on the ionospheric modal components. The model should be useful both in theoretical analyses of communication system performance and for channel simulator designs.
256 citations
TL;DR: The purpose of this paper is to summarize the most significant of these results, to link various models and approaches, and to indicate areas of study in which additional research seems both desirable and feasible.
Abstract: The problem of designing networks that can survive an enemy attack or natural disaster has received considerable attention in recent years. Work in this area has focused on the formulation of survivability criteria, the development of analysis methods to rank networks with respect to these criteria, and the generation of networks which are optimal with respect to these criteria. Many partial results for a variety of network models are available. The purpose of this paper is to summarize the most significant of these results, to link various models and approaches, and to indicate areas of study in which additional research seems both desirable and feasible.
175 citations
TL;DR: The digital adaptive equalization technique presented is characterized by a relatively simple algorithm and unique design features that permit fast adaptation in the decision-directed mode, and simplicity, small size, low cost, and precision are achieved.
Abstract: The digital adaptive equalization technique presented is characterized by a relatively simple algorithm and unique design features that permit fast adaptation in the decision-directed mode. For example, at 4800 bit/s the initial error rate in this mode may be as high as 25 percent, yet adaptation is assured. The equalization process is applied to the modified duobinary technique, but the principle is sufficiently general to employ any kind of system-correlative or multilevel. The present model employs primarily large scale integration (LSI), medium scale integration (MSI), and few integrated circuits; no discrete circuitry is included. As a result, simplicity, small size, low cost, and precision are achieved.
123 citations
TL;DR: In this paper, the effect of predetection bandwidth restriction on the error performance of binary frequency modulation (FM) systems for various frequency deviations has been investigated with an examination of the probability of bit error.
Abstract: The spectra of digital frequency modulation (FM) signals have been observed to be narrow compared with the spectra of digital baseband signals or the corresponding amplitude modulation (AM) and phase modulation (PM) signals. This suggests a high efficiency of transmission in terms of occupied bandwidth for binary FM. Here the desirability of digital FM systems is investigated further with an examination of the probability of bit error. The FM receiver considered consists of a predetection filter, a limiter-discriminator, and a postdetection filter. New results are presented on the effect of predetection bandwidth restriction on the error performance of binary FM systems for various frequency deviations. Theoretical results have been obtained by a Fourier analysis of the distorted FM signals and by Rice's click analysis of FM noise. The data presented permits an optimum selection of system parameters such as frequency deviation ratio and predetection filter bandwidth. It is found that in narrow-band operation a peak-to-peak frequency deviation of about 0.7 times the bit rate and a bandpass filter bandwidth of about 1.0 times the bit rate yield a minimum probability of error. Experimental measurements have been made and there is, in general, good agreement with the theory. A comparison of the error performance of noncoherent binary FM, coherent AM, and coherent PM in a restricted band is made. Narrow-band noncoherent FM proves superior to AM in any bandwidth. In narrow-band operation, the noncoherent FM gives a somewhat better error performance than the coherent PM. For example, for an error rate of 10-4and a bandpass filter bandwidth of one times the bit rate, the performance of a noncoherent FM system is 0.7 dB better than that of a coherent PM system.
116 citations
TL;DR: It is shown that an extended structure of this type is also optimum for the completely general case of M multiplexed PAM signals transmitted over random-time and frequency-dispersive multichannel systems with I inputs and N outputs, where each output is subjected to arbitrarily correlated additive noise.
Abstract: It is well known that the optimum receiver for pulse-amplitude-modulated (PAM) signals, transmitted over a single channel, consists of a cascade of continuous and transversal filters. It is shown that an extended structure of this type is also optimum for the completely general case of M multiplexed PAM signals transmitted over random-time and frequency-dispersive multichannel systems with I inputs and N outputs, where each output is subjected to arbitrarily correlated additive noise. The random channels may also be arbitrarily correlated. In general, the receiver consists of MN continuous filters and M tapped delay lines with M2sets of taps. An adaptive version is also described. The results include, as special cases, the situations considered by Gonsalves and Tufts [8], Shnidman [5], and Kaye [3]. Also, included as special cases are new results for diversity systems, multi-input systems such as multipair cables, and multiple-terminal systems such as multiple access repeaters in which cochannel interference occurs.
85 citations
TL;DR: The equalizer uses a modified zero-forcing tap-gain-adjustment algorithm, which is no more costly to implement than the zeroforcing algorithm, but which has superior convergence properties in the presence of severe distortion.
Abstract: A small and relatively inexpensive automatic equalizer has been developed for use with a vestigial sideband AM data set presently utilized to provide a high-speed Bell System voiceband data service. The equalizer uses a transversal filter structure comprising a tapped analog delay line with a variable-gain circuit associated with each tap. Its operation is adaptive in the sense that the tap-gain-adjustment information is derived from the received data, so that the tap gains are continuously optimized. A number of adaptive tap-gain-adjustment algorithms and transversal filter realizations were considered. The algorithms range in order of implementation complexity from "zero forcing," involving binary correlation, to "mean square," which requires linear correlation. The transversal filter realizations vary from an early scheme, which accomplished the variable-gain function with up-down counters controlling a switchable resistive ladder network, to an all-digital approach, where the analog delay line is replaced by shift registers and the variable-gain function is achieved with binary multipliers. Although the all-digital approach, using the mean-square algorithm and implemented with large-scale integration, will undoubtedly be the best choice in the near future, the equalizer to be described was selected on the basis of its being the cheapest to implement at the present time. The equalizer uses a modified zero-forcing tap-gain-adjustment algorithm, which is no more costly to implement than the zeroforcing algorithm, but which has superior convergence properties in the presence of severe distortion. The variable-gain function is achieved by means of a field-effect transistor (FET) used as a variable resistance. A single operational amplifier is used both to control the resistance of the FET and to provide feedback necessary to insure precise circuit performance without requiring close tolerances on the FET characteristics.
79 citations
TL;DR: In this article, a queuing model with limited waiting room (buffer), batch (burst) Poisson arrivals, and a synchronous server (transmission channel) with constant service time (constant transmission rate) was studied.
Abstract: A queuing model with limited waiting room (buffer), batch (burst) Poisson arrivals, and a synchronous server (transmission channel) with constant service time (constant transmission rate) is studied. Using average burst length and traffic intensity as parameters, the relationships among buffer size, overflow probabilities, and an approximation to the expected queuing delay of each burst due to buffering are obtained. These relationships are represented in graphs which provided a guide to the buffer design problem. A simple example is given which applies these results to the design of a buffer system in a time-sharing computer system. Although the problem arose in studies of design of statistical multiplexers, the queuing model developed is quite general and may be useful for other industrial applications.
69 citations
TL;DR: In this article, the nonlinear phase versus input power behavior of traveling-wave tubes (TWT) and its impact on intermodulation noise and intelligible crosstalk is treated.
Abstract: A modern communications satellite typically amplifies many frequency-modulated radio-frequency carriers in a single common amplifier. The nonlinear phase versus input power behavior of traveling-wave tubes (TWT) and its impact on intermodulation noise and intelligible crosstalk is treated. A theoretical model is presented, which allows calculation of the multicarrier performance of the TWT, and an experimental verification of the model is given.
64 citations
TL;DR: Exact error probabilities are given for a binary phase-shift-keyed (PSK) signal which is transmitted over a nondistorting channel but additively corrupted by Gaussian noise and cochannel interference.
Abstract: Exact error probabilities are given for a binary phase-shift-keyed (PSK) signal which is transmitted over a nondistorting channel, but additively corrupted by Gaussian noise and cochannel interference. Both coherent detection and quasicoherent (noisy reference) detection are considered. The analysis admits a wide class of interferences, but numerical results are limited to interferences modeled by a sum of randomly phased sinusoids. This includes most extraneous analog frequencymodulation (FM), phase-modulation (PM), or digital PSK signals, or combinations of the same which are cochannel with the desired PSK signal.
63 citations
TL;DR: The traffic-carrying capability of three types of common-user mobile radio systems are derived and compared; they are also compared with a private-user system.
Abstract: The traffic-carrying capability of three types of common-user mobile radio systems are derived and compared; they are also compared with a private-user system. The systems considered make use of multichannel capability for mobile units or a dispersed array of base stations, or both.
53 citations
TL;DR: A fairly exhaustive, tutorial, and partly critical review of a major part of the extensive literature pertaining to theoretical and experimental results on the propagation of laser signals in the atmosphere to obtain a communication channel model of the atmosphere for laser frequencies.
Abstract: The advent of the laser has generated much interest in its use for communication. Consideration is given in this paper to obtaining a communication channel model of the atmosphere for laser frequencies. For this purpose a fairly exhaustive, tutorial, and partly critical review of a major part of the extensive literature pertaining to theoretical and experimental results on the propagation of laser signals in the atmosphere is given. Some of the basic characteristics of the channel for Which detailed results are given are 1) channel time spread (or, equivalently, channel coherence bandwidth) due to multipathing and dispersion, 2) channel gain, 3) channel doppler spread and coherence time, 4) channel spatial coherence and amplitude covariance functions. Consideration is given to both point receivers and area receivers, heterodyne and direct detection receivers, nonplanar (i.e., spherical and beam) wave propagation, and infinite plane-wave propagation. The results are given for clear weather conditions. Brief attention is given to additive noise. In reviewing the extensive literature dealing with the propagation of optical signals through a turbulent medium, an attempt is made to put this varied, diverse, and in some instances controversial material into proper perspective; to note where theoretical results derived for one set of conditions were applied to experimental results obtained for another set of conditions; to correct some invalid conclusions drawn from some theoretical results as far as signal degradation effects and parameter variations; and to indicate where out-of-date vertical profile models were used with updated models being given, and where possible the new theoretical predictions based on these models provided. The paper should, as a result, be of help to those familiar with the literature as Well as those who are not. For those not acquainted with the literature, in addition to providing a fairly in-depth knowledge of the characteristics of propagation through a turbulent medium and an indication of the status of our understanding of this subject, the paper should provide a stepping stone for obtaining a more thorough understanding of the subject and for doing work in the field.
TL;DR: In this paper, a method for analyzing intermodulation spectra in limiting amplifiers using a statistical approach has been presented for the case of a large number of input signals and the accuracy appears to be good down to seven signals.
Abstract: A method is presented for analyzing intermodulation (IM) spectra in limiting amplifiers using a statistical approach Engineering approximations are derived for the case of a large number of input signals The accuracy appears to be good down to seven signals The approach permits analysis of arbitrary locations (in frequency) of the input signals Possible ways of reducing IM interference in limiting satellite repeaters are discussed The analysis is generally restricted to the ideal hard-limiting amplifier however, its applicability to other nonlinear characteristics is illustrated Additional sources of degradation, such as amplitudemodulation-frequency-modulation (AM-FM) conversion and system phase nonlinearity, are not treated
TL;DR: In this article, the mutual coupling effects on an M -branch maximum ratio combiner with correlated signal fading in general are considered and the results in the mobile radio case show that the average power delivered to the receiver becomes less as the antenna spacing of an antenna array decreases (i.e., the mutual couplings increases) with either an optimum load network or an optimum resistive load network connected to the antenna array at the front end of the receiver.
Abstract: The mutual coupling effects on an M -branch maximumratio combiner with correlated signal fading in general are considered. The results in the mobile radio case show that the average power delivered to the receiver becomes less as the antenna spacing of an antenna array decreases (i.e., the mutual coupling increases) with either an optimum load network or an optimum resistive load network connected to the antenna array at the front end of the receiver. However, there is not much effect on the power as long as the antenna spacing is equal to or greater than a certain value dependent on the number of the branches M . For M \leq 4 , the required antenna spacing is equal or greater than 0.2λ. The value for M > 4 increases gradually as M increases. It can be obtained following the same technique. The mutual coupling effect on the cumulative distribution of the signal from an M branch mobile diversity array is also small. Hence for an in-line array with M \leq 4 , the antenna spacing of 0.2λ can still provide most of the usual diversity advantage.
TL;DR: An algorithm for adjusting the tap gains of a tapped delay line (TDL) equalizer which converges for all channels to the setting, which minimizes the mean-square error at the filter output due to both noise and intersymbol interference is described.
Abstract: Results are reported which are of interest in the automatic adaptive equalization of telephone lines or other narrowbandwidth channels where data transmission is limited primarily by intersymbol interference. We describe an algorithm for adjusting the tap gains of a tapped delay line (TDL) equalizer which converges for all channels to the setting, which minimizes the mean-square error (rose) at the filter output due to both noise and intersymbol interference. This algorithm has been implemented using digital microcircuits and metal-oxide semiconductor (MOS) shift registers to construct a 63 tap all-digital TDL filter covering 19.7 ms. The modulation system used is vestigal sideband with bandwidth 1.6 kHz. The data rate is 3200 pulses/s, or 9600 bit/s with 8 data levels. A description of the hardware realization and performance data over various telephone lines is given.
TL;DR: The error performance of binary DPSK transmission for the case of noise and general cochannel interferences is considered to include multiphase signaling and the uncertainty in the approximation is small enough to make the results quite usable for systems design.
Abstract: This paper is a sequel to a previous paper which considered the error performance of binary DPSK transmission for the case of noise and general cochannel interferences. The present work extends that analysis to include multiphase signaling. Whereas the binary analysis produced exact expressions for the probability of error, the extension to M -ary DPSK considered here leads to upper and lower bounds to the symbol error probability. The bounds have a ratio of 2. However, by a heuristic argument it is possible to estimate the proximity of the error probability to one of the bounds, and so the uncertainty in the approximation is small enough to make the results quite usable for systems design. Computations are displayed for quaternary ( M = 4 ) single difference (conventional) DPSK.
TL;DR: In this paper, the authors proposed a detector which has no normalization and is superior to the ideal detector below zero to 10 dB input S/N, where the noise at the output of the detector is additive for all input S /N ratios and in this sense is comparable to synchronous detection of AM.
Abstract: Conventional frequency-modulation (FM) detectors show a marked decrease in output S/N below an input S/N of 5 to 10 dB. This threshold effect is caused by "clicks" or phase slips of multiples of π. This effect is related to a normalization (clipping) which is very beneficial at high S/N . A detector is described which has no normalization and is superior to the ideal detector below zero to 10 dB input S/N . The noise at the output of the detector is additive for all input S/N ratios and in this sense is comparable to synchronous detection of AM. Plots of output noise spectrum and output S/N with and without sine-wave modulation are given. Another detector is also proposed that is asymptotic to the ideal detector at high S/N and the aforementioned detector at low S/N . Comparison is made with conventional detectors. The results indicate that this detector may be especially useful in the detection of frequency-shift keying and phase-shift keying in telemetry systems.
TL;DR: The steady-state phase noise performance of an absolute value type of early-late gate bit synchronizer is developed using the Fokker-Planck method and yields the best performance (in the sense of minimum phase noise) at every value of R s.
Abstract: The steady-state phase noise performance of an absolute value type of early-late gate bit synchronizer is developed using the Fokker-Planck method. The results are compared with the performance of two other commonly used bit synchronizer circuit topologies on the basis of either 1) equal equivalent signal to noise in the loop bandwidth in the linear region, or 2) equal loop bandwidth at each input signal-to-noise ratio R s . These comparisons are made as a function of R s . In both cases, the absolute value type of early-late gate yields the best performance (in the sense of minimum phase noise) at every value of R s .
TL;DR: The dynamic stability of a simple two station network is analyzed mathematically and the extensive analysis of the dynamic behavior of various system configurations using a computer simulation of the time history response to various perturbations is reported.
Abstract: An analysis of the performance of a synchronous digital communications system in which the clock rate of each station is established as the average of the clock rates of all incoming signals to each station is presented. Using both-end control, the system frequency is the average of the free-running frequencies of the phase locked oscillators, weighted by the number of inputs to each station. In single-end control, the frequency is a similar average plus a term, which is a function of changes in transmission path delay. Rules are derived defining how a transmision path delay change will be reflected in changes of total phase difference at each transmission path terminal. These phase changes reflect requirements for elastic storage of data at each transmission path termination. The dynamic stability of a simple two station network is analyzed mathematically. A subsequent technical paper [7] will report on the extensive analysis of the dynamic behavior of various system configurations using a computer simulation of the time history response to various perturbations.
TL;DR: In this paper, an M branch equal-gain predetection space-diversity combiner is analyzed for the case where the signals received from the individual branches of such a combiner are partially correlated.
Abstract: An M branch equal-gain predetection space-diversity combiner is analyzed for the case where the signals received from the individual branches of such a combiner are partially correlated. The level crossing rate (LCR) of the combined signal envelope equals the probability density function times a function of the direction of motion. When a linear antenna array is used on a mobile radio receiver, the direction of motion does not affect the shape of the LCR curve of the combined signal but does affect the absolute values, especially when the antenna spacing d is less than 0.5 λ. There is apparently no effect for d > 1 \lambda
TL;DR: Using the moments of the subsequence weight distribution, a relationship is shown between the subsequences of the pseudorandom binary sequence and subsequences composed of random binary digits.
Abstract: The properties of subsequences of long m sequences are studied using the moments of the subsequence weight distributions. These moments are shown to be an aid for selecting good m sequences for correlation-detection problems. In particular, the first four moments are described in detail for subsequence lengths M \leq 100 digits for six different m sequences. Two simple algorithms are described for determining the third and fourth moments. An algorithm for calculating the fourth moment and a detailed description of this moment for the six m sequences are presented in this paper. Using the moments, a relationship is shown between the subsequences of the pseudorandom binary sequence and subsequences composed of random binary digits. Estimates for the moments of the subsequence weight distribution are obtained by sampling the m sequence. A statistical analysis of these results is presented.
TL;DR: The results of a recent study to determine and evaluate effective methods of applying forward-acting error-control-coding techniques to multichannel high-frequency (HF) data modems are presented.
Abstract: The results of a recent study to determine and evaluate effective methods of applying forward-acting error-control-coding techniques to multichannel high-frequency (HF) data modems are presented. A unique feature of this study is a conscious exploitation of the fine-grain bit-error behavior, both in time and in frequency, that is characteristic of multichannel HF data transmission systems. Attention is focused on the ability of four promising coding techniques to combat the characteristic error behavior. These techniques involve the use of bit-interleaved binary block codes, diffuse convolutional codes, block-interleaved two-stage concatenated codes with binary-coded inner stages and nonbinary-coded outer stages, and a class of two-stage binary codes with random-error-correcting inner stages and burst-error correcting outer stages.
TL;DR: The steady-state behavior of a data-transition tracking loop, used as a bit synchronizer in a phase-coherent receiver, is considered and it is shown that by adjusting the quadrature channel gain along with the integration interval, a significant improvement in phase noise and cycle slip performances can be achieved.
Abstract: The steady-state behavior of a data-transition tracking loop, used as a bit synchronizer in a phase-coherent receiver, is considered. Optimization of mean-square phase noise and mean time to first cycle slip is performed when the average power of the reference cross-correlating signal is constrained. It is shown that by adjusting the quadrature channel gain along with the integration interval, a significant improvement in phase noise and cycle slip performances can be achieved over that system which integrates in the quadrature channel over the full symbol period. All the results are derived for a first-order loop filter merely to indicate the approach to the problem and the relative value of optimizing the system.
IBM1
TL;DR: The digital echo modulation eliminates the need for precise analog filters and minimizes the effects of component instability; in addition, it gives a great flexibility to the design, allowing special features such as efficient half-speed operation.
Abstract: A new technique for the transmission of binary data over band-limited channels is described. The signaling method is a generalization of Nyquist's telegraph theory, it allows pulse modulation without intersymbol interference within a frequency spectrum centered at any multiple of one quarter of the signaling rate. The signals are derived from pulse trains generated by a combination of logic circuits and weighting resistors. The number of pulses constituting a signal element varies typically from 5 to 9 in function of the desired total bandwidth. Signaling at 80 percent of the Nyquist rate has been easily achieved with 40-dB out-of-band attenuation. The digital echo modulation eliminates the need for precise analog filters and minimizes the effects of component instability; in addition, it gives a great flexibility to the design, allowing special features such as efficient half-speed operation.
TL;DR: A decomposition of these graphs into a class of subgraphs called step- l cycles, of which a directed Hamilton cycle is a special case, which is demonstrated in simultaneous k commodity flows.
Abstract: Algorithms are given to synthesize optimally invulnerable directed graphs with an arbitrary number of vertices and a minimum number of branches. We give a decomposition of these graphs into a class of subgraphs called step- l cycles, of which a directed Hamilton cycle is a special case. The use of this property in simultaneous k commodity flows is demonstrated.
TL;DR: The application of a binary transversal filter as transmitting filter and a digital circuit as modulator makes it possible to construct data transmitters consisting only of transistors and resistors.
Abstract: The application of a binary transversal filter as transmitting filter and a digital circuit as modulator makes it possible to construct data transmitters consisting only of transistors and resistors. The modulation distortion, caused by the keying of a low-frequency square-wave carrier, can be compensated by a modification of the transfer function of the transmitting filter, provided the carder frequency is chosen to be a multiple of half the bit rate. This principle makes it possible to realize the whole data transmitter in one single integrated circuit. The binary transversal filter, provided with an analog-to-digital converter, can also be used to construct the data receiver for the greater part with digital circuitry.
TL;DR: In this paper, the noise bandwidth of Butterworth, Bessel, and Chebyshev filters is tabulated, and it is proven that B N is invariant when a low-pass to bandpass transformation is made.
Abstract: The noise bandwidth B N of Butterworth, Bessel, and Chebyshev filters is tabulated. The Butterworth filter is shown to have the minimum B N to 3-dB bandwidth ratio for any secondorder filter. It is proven that B N is invariant when a low-pass to bandpass transformation is made. An error is corrected in recent tables of Chebyshev filters.
TL;DR: Second-order phase-locked loops, operating in the presence of frequency detuning, are used as a means of providing phase synchronization and the results are applicable to the problem of establishing noisy reference losses in a broad class of coherent PCM telemetry systems.
Abstract: In phase-coherent communication systems which use phase-locked loops to provide synchronization of the data detector, the communications engineer is frequently faced with the problem of determining the effects which noisy timing has upon detection efficiency. This paper is concerned with determining these effects when second-order phase-locked loops, operating in the presence of frequency detuning, are used as a means of providing phase synchronization. The results are also applicable to the problem of establishing noisy reference losses in a broad class of coherent PCM telemetry systems, e.g., PCM/PSK/PM. Also presented are results which can be used to determine steady-state statistical dynamics of second-order loops.
TL;DR: In this paper, the error probability of differential phase shift keying and non-coherent frequency-shift keying (FSK) was analyzed as a function of the pulse carrier frequency offset, and it was shown that the frequency offset which causes the DPSK error probability to be larger than the FSK error probability is approximately 0.74.
Abstract: Expressions and curves of error probabilities of differential phase shift keying (DPSK) and noncoherent frequency shift keying (FSK) are obtained as a function of the pulse carrier frequency offset. These indicate that the frequency offset which causes the DPSK error probability to be larger than the FSK error probability is approximately \Delta\omegaT = 0.74. This crossover frequency appears to be a linear function of the signal-to-noise ratio over the range-4 dB to 14 dB. The crossover point is extended by sub-bit processing and the previous curves are modified for this case.
TL;DR: In this paper, the bit error probability performance of both frequency-shift-keyed (FSK) and differential phase-shift keyed (DPSK) binary systems are described when disturbed by an interference consisting of a mixture continuous wave (CW) tone and narrow-band random noise in addition to the usual broad-band thermal noise background.
Abstract: The bit error probability performance of both frequency-shift-keyed (FSK) and differential phase-shift-keyed (DPSK) binary systems are described when disturbed by an interference consisting of a mixture continuous wave (CW) tone and narrow-band random noise in addition to the usual broad-band thermal noise background. Performance of the two systems is shown to be identical with the exception that the equivalent DPSK performance occurs at 3 dB less signal-to-thermal-noise ratio. The combination of CW tone and noise interference severely degrades system performance under a broad range of conditions.
TL;DR: The method is based on Bayes estimation, opitmization theory for discrete systems, and discrete invariant imbedding and results of two examples indicate the efficacy of the method.
Abstract: In digital communications, the theory of optimum data detection requires precise knowledge of the bit transition time at the receiver before bit-by-bit detection can be implemented. This paper presents a method for the optimum design of discrete sequential algorithms for bit synchronization which may readily be implemented with digital hardware or with a digital computer. The method is based on Bayes estimation, opitmization theory for discrete systems, and discrete invariant imbedding. Results of two examples indicate the efficacy of the method.