Speech transmission using DPCM coding, partial response continuous phase modulation and diversity

The performance of constant envelope digital partial response continuous phase modulation (PRCPM) with two-bit differential detection and offset receiver diversity is theoretically analyzed in fast Rayleigh fading AWGN (additive white Gaussian noise) channels. A simple closed-form expression for the probability of error is derived and evaluated for cases of practical interest to researchers and designers of land mobile radio systems. It is shown that the dynamic BER performance is considerably improved using the offset diversity scheme. Thus, many new PRCPM signals having compact power spectra can be used in future mobile radio systems. >

The complexity of the multiple-symbol detection structures for partial response continuous phase modulation (CPM) signals goes exponentially with the length of the pulse shaping filter and the observation length as well. Simple no coherent trellis receivers for partial response CPM signals are proposed in this paper. Parallel to the simple coherent detectors, where CPM signals are approximately represented by a sum of Phase Amplitude Modulation waveforms, Laurent decomposition is applied to reduce the complexity. The proposed no coherent receiver is structured with only fixed K matched filters in the front-end and a processor which uses Viterbi algorithm with a 2N-1 states trellis to perform the sequence estimation. It is attractive because the structure does not involve any matrix calculations. The good error performance is confirmed by simulation results.

A differentially coherent receiver which uses the Viterbi algorithm is presented. This receiver compensates for intersymbol interference (ISI) and avoids noise enhancement. A phase estimate which takes into account the presence of inherent ISI in the signal is derived. This phase estimate is then used in the decision metric of a coherent receiver. The resulting decision rule can be implemented using the Viterbi algorithm. The complexity of the resulting receiver increases with the duration of observation used by the implicit phases estimator. A simple four-state Viterbi receiver with good performance is obtained when the phase estimate is based on one observable. Simulation results are given. >

The joint optimization of partial response continuous phase modulation (CPM) parameters is addressed. For fixed modulation order and memory length, the modulation index, the symbol rate and the phase pulse shape are considered as degrees of freedom. In particular, we drop the restriction of “classical” phase pulses and we allow for custom shapes, which are parametrized through Bezier curves. In order to deal with the many variables involved, an evolutionary algorithm is employed for numerical optimization. The optimization metric is related to the extrinsic information transfer (EXIT), under the assumption that CPM is serially concatenated with an outer convolutional encoder, as it is the case for several practical systems. The presence of high adjacent channel interference (ACI) is explicitly accounted for in the optimization process. We show that the proposed optimization strategy can effectively exploit the power-bandwidth trade-off. Nevertheless, only moderate improvements are achievable in comparison to state-of-the-art serially concatenated systems.

Summary form only given. Future mobile radio systems are expected to provide high-quality voice and high-speed data transmission and to be integrated into the digital communication networks in several countries. The performance of some digital modulation techniques that are relevant for future mobile radio systems is studied. These techniques are characterized by having a constant envelope signal to reduce the effect of nonlinearity and fading and a compact RF power spectrum to reduce adjacent channel interference. Most digital partial-response continuous-phase modulation (PRCPM) techniques such as 2 RC, 2 SRC, 2 RC, 3 SRC, GMSK, and GTFM have the above properties. Discriminator detection of these techniques is analyzed in fast Rayleigh fading characterizing land mobile radio channels. Approximate closed-form expressions for the probability of error are derived and numerical results are presented for land mobile radio systems. >

The performance of cellular digital mobile radio systems is investigated using quadrature amplitude modulation (4-QAM) and differential phase shift keying (DPSK) in a frequency nonselective Rayleigh fading environment with lognormal shadowing, including the effects of (white) Gaussian noise, narrow-band impulsive noise/interference (Middleton's Class A model), and cochannel interference. The performance is analyzed in terms of probability of bit error. The UHF groundwave propagation law is introduced by considering the local mean power to be inversely proportional to the fourth power of propagation distance. To investigate the improvement in performance the influence of selection diversity and maximal ratio combining on the bit error probability is analyzed. It has been found that shadowing significantly affects the bit error performance. Application of selection diversity or maximal ratio combining results in important improvement of system performance. Selection diversity gives better results than maximal ratio combining for low signal-to-noise ratios. Results of our analysis show maximal ratio combining is not that sensitive to unequal branch signal-to-noise-plus-interference ratio.

This paper considers the application of constant-amplitude, partial-response, continuous-phase modulation with simple near-optimum receivers to cellular digital mobile radio systems. These smoothed modulation schemes have low spectral sidelobes and narrow main lobes. A combiner for time-division retransmission systems with space diversity is given for continuous-phase modulation. Cochannel interference and adjacent-channel interference are calculated for continuous-phase modulation in cellular systems with frequency reuse. The efficiency of 3RC and 4RC modulations with space diversity is evaluated for conventional cellular systems.

An integrated circuit for the Pan-European Groupe Speciale Mobile (GSM) digital radio systems is described which performs the modulation and front-end functions for both base and mobile stations. The circuit, fabricated using a 2- mu m CMOS technology, fully meets the GSM requirements. The system uses Gaussian minimum-shift keying (GMSK) modulation, with a 0.3 modulation index, which is characterized by a narrowband spectrum. The circuit dissipates 180 mW with a single 5-V supply. The die size is 6.7 mm*5.3 mm.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Pan-European speech coding standard for digital mobile radio

Direct sequence (DS) code-division multiple-access (CDMA) is a modulation and multiple access technique currently being implemented in digital cellular mobile radio communication systems. The conventional quasisynchronous (QS) DS-CDMA communication system is generalized, and the problem of optimal spreading pseudonoise (PN) sequence design for new QS DS-CDMA communication systems is formulated. The classes of optimal PN sequences which have good pseudo-randomness and very small cross-correlation values for some range of offsets are designed by using a matrix-based approach. In contrast to the case of the other sequences, the optimal PN sequences produce small multiple-access interference (MAI) which is not affected by the number of users if there are sufficiently large number of users in the link. The new results are compared with the commonly used PN sequences to prove the superiority of our designed sequences, and the system performance of the QS-CDMA systems using different PN sequences are also evaluated.

A digital cellular mobile radio system has been under development in Europe since 1982 under the coordination of the working group CEPT GSM (groupe speciale mobile). In a recent coordinated experiment, listening opinion tests were performed on the speech output of six candidate 16 kb/s speech coding schemes for this system: one regular-pulse excited coder, one multiple-excited coder, and four subband coders. For comparison purposes, test conditions from a companded cellular FM system currently in operation were included in the experiment. The six codecs were companded in terms of subjective quality, transmission delay, and ease of implementation. In this overall comparison, no single codec was superior in all respects. However, the regular-phase-excited linear predictive coder, which provided the best speech quality, had acceptable complexity and delay and was singled out for further improvement. Ultimately, an improved version of this codec, a regular-pulse-excited/long-term-prediction LPC coder was selected. >

Designing a spectrally efficient wireless channel requires a comprehensive understanding of its time and frequency varying characteristics, making it a stochastic medium of communication. These characteristics become more challenging to cater at the receiving terminal in a multipath transmission. This is because of the fading effect and Doppler shift of the transmitted frequency, specifically in cellular mobile radio systems and vehicle to vehicle communications. This paper presents the modeling, simulation, and then characterization of a cellular mobile radio multipath channel over its time and frequency varying fading gain. For this purpose, a discrete-time Finite Impulse Response (FIR) type filter of such a channel is designed, modeled, and simulated using time and frequency varying characteristics of the received signal. The simulated channel response is further analyzed in terms of coherence bandwidth, coherence time, delay spread, Doppler spread, and symbol time.

In cellular mobile radio systems using coded modulations, the spectrum efficiency of the system is related to the cutoff rate and the cluster size. We evaluate the cutoff rate for M-ary QAM considering a coherent detection and a memoryless flat fading channel. Simulation results relate the cluster size to the signal-to-noise ratio and to the cutoff rate. We present the spectrum efficiency of the system as a function of the cutoff rate and show its limits in the classical cellular system. We apply the same procedure to examine the performance of systems using directional antennas with and without sectorization of the original cell surface. With our proposed representation, we can always determine the optimum spectrum efficiency of the modulation that maximizes the spectrum efficiencies of the different systems.

There are two types of strategies to assign the channels to a cell: fixed channel assignment (FCA) and dynamic channel assignment (DCA). The basic characteristics of DCA are presented in comparison with FCA. The authors introduce various DCA strategies and discuss some of most important DCA research topics currently being pursued. In particular, the authors focus on microcellular systems, including a proposed MD (moving direction) strategy. It uses the information a mobile station's (MS's) moving direction with the DCA strategy. A channel is selected among those occupied by MSs moving in the same direction with the concerned MS. In this manner, the MS can still use the same channel at the handover with high probability. DCA requires more radio equipment at each base station than FCA, and it requires elaborate channel assignment control. However, it exploits the limited frequency spectrum efficiently and has a lower forced call-termination rate than FCA when the blocking rates are equal.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

This paper describes a modification of the Free-Space and Hata formulae for the prediction of received signal power, PR and propagation path loss, LP, in two cellular mobile radio systems (CMRS), in the Northern Nigeria. Measurements of PRs were taken with a Cellular Mobile Radio test Receiver (Sagem OT 160), in some selected open/rural environments, when the receiver was being moved away from the base stations along the propagation paths. LPs were then obtained from values of measured PRs using an appropriate expression. A close comparison of measurement values and computed values from the free-space and Hata formulae revealed that direct application of these formulae is inappropriate for the prediction of these parameters in the region of investigation, as computed values fell short significantly from the corresponding measured values. Consequently, some correction factors have been introduced to both models and these have produced results which closely matched the measured values.