Methods are presented for the transmission of wideband digital signals above and below the frequency division multiplexing (FDM) baseband spectrum of analog microwave systems. In particular, it is shown that a Data Above Voice (DAV) system employing a 1.544 Mbit/s data channel above FDM channels and a Data Under Voice (DUV) system employing 1.544 Mbit/s channel below 720 FDM channels, are practical hybrid systems. Theoretical models of hybrid transmission are disucssed and the DAV/DUV methods are compared. A conceptual hybrid system design procedure is introduced and field measurement results on an 8 hop 6 GHz system are presented, comparing favorably with theoretical predictions.

This paper presents a class of digital transformations which generaten-level sequences having an autocorrelation function which remains nonzero throughout its domain in order to insure a narrow spectrum. In addition, any transformation of the class attains maximum redundancy and can be implemented by a simple one-bit memory stochastic machine. Most importantly, the autocorrelation function is obtained analytically which permits establishing for any odd value ofnthe set of level values which achieve optimum bandwidth compression.

A hybrid information carrier is defined as a carrier for partly analog and partly digital source information. Some of the reported schemes transmit 1.5 Mbits/s and 1200 frequency division multiplexing (FDM) channels, others 6.3 Mbits/s and 600 FDM channels, in the shared spectrum of the microwave baseband (BB). Baseband diversity (BBD) is frequently used in this type of system. As hybrid transmission is getting wider popularity, it is important to determine the requirements on hybrid BBD systems. Different block diagrams are considered and performance of realizable hybrid BBD units is discussed. The BBD has different configurations for different data rates. We discuss the reasons for this within the theoretical performance and practical limit of technology. The important functions of signal presence and noise evaluation occur in the BBD. A novel approach to noise evaluation in the BB of microwave systems is feasible with digital techniques. The analog noise and pilot detection are in its threshold for high noise condition and fast pilot sensing time requirement. The proposed scheme evaluates digitally the error probabilityP(e), based on error detections, and works well in high noise conditions.

Control, supervisory, and order wire signal transmission, i.e., service channel (SC) is a requirement for most microwave systems. In digital radio the data spectrum extends to the lowest baseband (BB) frequencies. Frequency translation of the SC above the data spectrum is one possibility. Another method, in frequent use, is the double modulation technique, e.g., the SC amplitude modulates the carrier that is already phase modulated by the data spectrum. A novel approach that results in significant equipment cost savings and good performance is proposed. The low-frequency power density spectrum of the random multilevel channel coded data is removed. The vacated spectrum is available for the SC. For different bit rates, the error probabilityP(e)degradation due to SC is calculated. The analysis is performed for various channel coded messages and SC spectrum utilizations. The high-pass filter that assures low-frequency data spectrum removal is optimized for the SC signal to noise ratio and dataP(e)degradation. The analysis is verified with computer simulations and actual multilevel frequency-shift keying (FSK) 6-GHz microwave hop measurements. The proposed approach is also applicable to analog microwave systems when carrying hybrid information. The hybrid system carries an SC, 6.3-Mb/s data, and 900 frequency division multiplexed voice channels in a shared BB.

Performance evaluations and measurement principles of hybrid information carriers, that is, carriers for partly analog and partly digital source infomation, are established. Simulation requirements, which are independent from a particular hybrid configuration or system capacity, for the analog and digital sources are presented. In order to perform the analysis of multilevel digital streams, a simple formula for the power spectrum density of the randomn-level Gray-coded stream is derived. With normalization of the formula, the same power density spectrum for any number of levels is achieved. The required length of the pseudorandom sequence as a function of the transmission characteristics and hybrid environment is established. The measurement difficulties, when the noise source is not Gaussian, are discussed.

The relation between time-limiting and sequency-band-limiting in finite Walsh transforms is considered. It is found that optimum solutions (energy ratio equal to unity) are a nontrivial linear combination of[TM/N]successive orthogonal block pulses of durationR = NM^{-1}. Optimum solutions for bandpass signals are pointed out. Extension to the continuous-time domain is also discussed.

A general expression for oo such that cal (k,θ+θ0)= sal (k,θ) is obtained and a proof is given. This new form of θ0 is more direct and thus easier to use than other existing formulas.