Burton R. Saltzberg

Performance of an Efficient Parallel Data Transmission System

A parallel quadrature AM data transmission system may be implemented with a number of overlapping channels, each carrying a signaling rateb, spacedb/2apart in frequency. When a large number of channels are used, the system allows transmission speeds very close to the Nyquist rate, with little sensitivity to delay and amplitude distortion of the transmission medium. The receiver requires precise phasing of the demodulating carriers and sampling times in order to keep crosstalk between channels small. In the presence of delay and amplitude distortion, better results are obtained when half cosine roll-offs are used for shaping each channel than for full cosine roll-off. This transmission scheme appears to be a promising technique for achieving good performance at high information rates over bandlimited dispersive transmission media.

The capacity of PCM voiceband channels

It is theoretically shown that the capacity of a voiceband channel operating through a digital network is at least 56 kbps (for a 3.5 kHz baseband bandwidth). The capacity may actually be higher but how close one can come to 64 kbps, the limit imposed by the digital network itself, is unknown. The transmission method is based on generating nonuniformly pulsed PAM (pulse amplitude modulation) signals which at the network sampling instants go through one of the slicing levels of the A/D converter of the network. To accomplish this, both the slicing levels and the sampling clock times must be known. This is a hard practical problem and one can only conjecture if a reasonable solution will be possible. The signal may be recovered using a Viterbi detector to reduce intersymbol interference effects, but at present, the number of states required for this detector is very large.<<ETX>>

Error-burst detection with tandem CRCs

The authors investigate the efficacy of using two different cyclic redundancy check (CRC) codes in tandem to increase error-burst detecting capability. For a set of pairs of CRCs which are used in standards, it is found that the guaranteed detectable burst length is less than the sum of the individual guaranteed detectable burst lengths, but not much less. Thus strengthened CRC codes can readily be obtained using existing devices. >

Second-order statistics of logarithmic quantization noise in QAM data communication

When a QAM (quadrature amplitude modulation) data signal is transmitted over a channel that includes PCM codecs, it has frequently been observed and shown that the resulting noise around the final received constellation points grows with the distance of the point from the origin. The authors extend the analysis of this effect, and further show that the noise distribution about each point is elliptical, with greater variance in the radial direction. Formulas giving the signal-dependent noise components as a function of the channel characteristics of the analog transmit and receive access legs are derived. Experimental results confirm the analysis. >

QAM transmission through a companding channel/spl minus/signal constellations and detection

The effect of the companding process on QAM signals has been under investigation for the past several years. The compander, included in the PCM telephone network to improve voice performance, has an unusual affect on digital QAM data signals which are transmitted over the same channel. The quantization noise, generated by the companding process which is multiplicative (and asymmetric), degrades the detectability performance of the outermost points of the QAM constellation more than that of the inner points. The combined effect of the companding noise and the inherent white Gaussian noise of the system, leads to a re-examination of signal constellation design. The authors investigate the detectability performance of a number of candidates for signal constellations including, a typical rectangular QAM constellation, the same constellation with the addition of a smear-desmear operation, and two new improved QAM constellation designs with two-dimensional warping. >

An improved Manchester code receiver

It is shown that a simple modification to a basic Manchester code receiver can significantly improve its performance, both in terms of noise tolerance and intersymbol interference. The modification consists of subtracting a delayed sample of the received signal from the current sample. This improved receiver has potential application in extending reliable operation of wire-pair transmission to 10 Mb/s over a distance of 600 ft, as indicated by an improvement of approximately 5.4 dB over present receivers in tolerance to white noise.<<ETX>>

On the solution of a quadratic integral equation arising in signal design

This paper considers the solution of a quadratic integral equation of the first kind arising in the design of bandlimited signals for binary communication using simple memory-less correlation detection. The signals are disturbed by additive white Gaussian noise. It is shown that a bandlimited signal can be designed which eliminates intersymbol interference for signalling at Nyquist rate. This signal is a solution to a quadratic integral equation. The existence of the solution, as well as a construction of the solution of this equation, are presented. The method of solution of the integral equation is applicable to a broad class of problems.

Coding for a channel with quantization in the presence of an estimable interference

The encoding and decoding schemes presented are aimed at enabling the transfer of data through a channel in which two types of interference are added to the transmitted signal and the sum is quantized. One of these interferences is known (or can be estimated), whereas the second is an additive white Gaussian noise (AWGN). Since the input of the quantizer is not accessible, the known interference can not be removed from the received signal. We show that the error rate for an uncoded transmission through this channel is unacceptably large, even for low noise levels and linear quantization. It is also shown that the problem becomes even more severe when a nonlinear quantization is present. Therefore, coding is essential and a huge coding gain is achievable in this application. An upper-bound on the error rate, contributed by the component codes of a multilevel code, has been developed for multistage decoding. Results of computer simulations of a practical case with optimal and suboptimal decoding algorithms, both developed in this paper, are presented.

Frequency Modulation Using Digital Filtering Techniques

A simple unstable second-order digital filter with one variable multiplier can provide a source of frequency-modulated signals. This modulator is readily time-shared and should prove attractive when a large number of transmitters are required. A system consisting of a multiple installation of identical binary frequency-shift keying data transmitters is described and analyzed. A refinement which totally eliminates amplitude and phase discontinuities is also presented.

Timing Considerations in the Interfacing of Synchronous Data Networks

The problems of clocking data across the interface between synchronous data networks are considered. Maximum time interval error is introduced as a specification which includes both phase and frequency errors.