This paper proposes a deterministic path-loss model for the open-sea environment. The model accounts for different effects including effective reflection, divergence, and diffraction due to rough sea and earth curvature. The model results show excellent agreement with experimental results from our recent measurement campaign, which investigated propagation at 2 GHz with a maximum distance of 45 km. Channel parameters like mean-square surface slope and standard deviation of surface height are evaluated, from which it can be concluded that the shadowing and scattering effects on the reflection ray will influence the fading amplitude within the distance of 0.6 First Fresnel Zone clearance.

A long-distance channel measurement campaign with a maximum distance of 45 km was performed in Trondheimsfjorden, Norway. In this paper, we give a detailed description of the channel measurements including route, RX and TX antenna sites and other channel parameters. The results of received signal level (RSL) obtained by using two different signal processing methods are demonstrated, and it is found that the influence of unwanted interference and noise becomes more than 2 dB at distances beyond 20 km, which proves the necessity of employing an interference-reduction method. The inbound trip RSL is found to be larger (~6 dB) than for the outbound trip, which is assumed to be due to the "inbound" TX antenna resting on the furthermost edge of a large horizontal metal plane, which might act as a mirror to create its image. The path loss has been compared with the free-space model, the Okumura-Hata model, the COST 231-Hata model and the ITU-R P.1546-2 model, and it has been found that the ITU-R model for open cold sea corresponding to field-strength value exceeded at 50% of the locations fits the measurement results best. However, some important deep fadings visible in our measurements are not seen in these models. Finally, the channel spatial correlation coefficient between the two RX-antennas is found to be less than 0.7 beyond a distance of 3 km. The ship turning and shadowing effect caused by other ships might, however, reduce the spatial correlation between the two RX channels.

Digital block-based predistortion (PD) is an efficient scheme to linearize power amplifiers (PA). The performance of this method highly depends on the estimation accuracy of the amplifier characteristics. In this paper, the PA characteristics are approximated by piecewise continuous polynomials in the form of splines. The approximation is based on a block of observed data at the input and output of the PA. This approach leads to a linear parametric approximation problem with moderate implementation complexity, and it also gives the best possible fit to a given set of observations. With cubic splines it is possible to get a fairly accurate estimate even with a short block of data, and hence a fast convergence of the PD algorithm. The performance of the method has been evaluated for an example selected from an application in mobile satellite communications with burst transmission. We use a model of a 2 Watts class A PA that is based on measurements, and the calculations are performed for 16-QAM modulation. The mean square approximation error, the power spectrum and the error vector magnitude (EVM) of the received signal have been analyzed by simulations. For 6 dB back-off, the adjacent channel power ratio (ACPR) level is reduced by approximately 15 dB relative to a transmitter without predistortion, after an adaptation based on only 25 transmitted symbols.

A short description of the transmultiplexer structure is given, and a brief review of some significant CCITT recommendations is presented. The paper briefly explains the influence of these recommendations on the design of the main subblocks of a back-to-back configurated transmultiplexer, the main subblocks being the two baseband filters, the twoo2DFT processors and the two channel filters (polyphase networks). The paper emphasizes how the above recommendations influence the design of the channel filters. As a consequence, the crosstalk mechanism is treated in detail and filter requirements based on these mechanisms are presented. The impact of the noise requirements is discussed and computer simulation results are presented. It is concluded that the channel filters which are realized as direct-form IIR filters could be designed using 16 bit wordlengths. Computer simulation results have shown that theo2DFT processor could also be designed using 16 bit wordlength when an optimum scaling strategy is used; however, the noise margin will be unsatisfactorily low in this case, and the number of bits is consequently increased to 20.

A comparison between different line codes suitable for transmission of 140 Mbit/s digital signals over 1.2/4.4 mm coaxial cable has been made using the computer optimization program, "OPUT." One alternative, the two level partial response class 1 (PR1), is chosen and a laboratory system has been built and tested. A block diagram of the system followed by a description is given. Performance of the system is presented and commented on. It is concluded that a PR1 system can be implemented with reasonable noise margin with a repeater spacing of 2.15 km for a 1.2/4.4 mm coaxial cable.

The Electronics Research Laboratory at the Norwegian Institute of Technology has since 1974 studied PCM-transmission in the local subscriber plant. A short discussion of this work is presented together with choice of design parameters and concept for a digital subscriber set. Construction and laboratory testing of the digital subscriber set are described. The paper concludes with a discussion on future development.