Дисертації з теми "Modulation en phase continue (CPM)"

Dans le cadre de cette thèse nous nous intéressons à la liaison d'un terminal utilisateur vers un satellite géostationnaire pour les futurs systèmes d'applications multimédia en bande Ka. Le but est de concevoir une forme d'onde qui permet de s'adapter aux conditions de propagation. L'adaptation des performances en spectre et en puissance de la forme d'onde, notamment dans un contexte avec une forte fluctuation du niveau du signal, comme celui du canal satellite en bande Ka, permet à la fois d'augmenter la capacité et d'améliorer la disponibilité du système. Dans ces travaux, les formes d'onde basées sur les modulations CPM (Continuous Phase Modulation) ont été adoptées. L'enveloppe constante, les différents paramètres ayant un impact sur les performances de la forme d'onde, ainsi que les bonnes performances en puissance lors d'un processus de décodage itératif sont les principales motivations de notre choix. Concernant l'adaptativité nous montrons qu'une forme d'onde CPM basée sur la variation de l'indice de modulation (et éventuellement la réponse en fréquence) au sein d'une famille d'indices ayant un même dénominateur est une solution particulièrement attractive. Une telle solution ne nécessite pas une très grande complexité lors d'un passage d'un mode de transmission à un autre. Elle nécessite essentiellement le changement des coefficients des filtres adaptés. Cette technique permet aussi d'obtenir une marge de performance assez importante vu que l'indice de modulation est le paramètre qui affecte le plus les performances de la CPM. La CPM octale 2RC est un schéma particulièrement attractif d'un point de vue performance.

Dans cette thèse, nous étudions la modulation par déplacement de fréquence à bande latérale unique (SSB-FSK), un schéma de modulation de phase continue (CPM) ayant, par essence, la caractéristique original du spectre à bande latérale unique (BLU). Tout d’abord, nous présentons l’origine du signal à partir de la physique quantique. Ensuite, nous proposons un détecteur de séquence à maximum de vraisemblance (MLSD) simplifié pour les schémas CPM conventionnels, basé sur le réarrangement montré dans le modèle de signal SSB-FSK. Pour exploiter pleinement les performances de la SSB-FSK, nous examinons la probabilité d’erreur du signal, l’occupation de la bande passante et la complexité du récepteur. Étant donné que différentes mesures de performance sont prises en compte, nous avons utilisé une optimisation multi-objectifs pour obtenir de nouveaux schémas SSB-FSK plus performants que les schémas CPM classiques. En outre, nous proposons une solution pour simplifier la complexité des signaux SSB-FSK en utilisant la décomposition de la modulation d’amplitude d’impulsion (PAM). Les impulsions PAM ont été obtenues à partir d’un algorithme que nous avons développé. En outre, nous proposons une séquence d’entraînement générique optimale pour l’estimation conjointe de la synchronisation des symboles, du décalage de fréquence et de la phase de la porteuse pour la synchronisation en mode burst. La séquence d’entraînement a été obtenue en utilisant les bornes de Cramér-Rao
In this PhD thesis, we investigate the single-sideband frequency shift keying (SSB-FSK), a continuous phase modulation (CPM) scheme having, by essence, the original feature of the single-sideband (SSB) spectrum. First, we present the origin of the signal from quantum physics. Then, we propose a simplified Maximum likelihood sequence detection (MLSD) detector for conventional CPM schemes based on the rearrangement shown in the SSB-FSK signal model. To fully exploit the SSB-FSK performance, we examine the signal error probability, bandwidth occupancy, and receiver complexity. Since different performance metrics are considered, we employed a multi-objective optimization to achieve new SSB-FSK schemes that outperform conventional CPM schemes. Moreover, we propose a solution to simplify the complexity of SSB-FSK signals using the pulse amplitude modulation (PAM) decomposition. The PAM pulses were achieved from an algorithm we developed. Furthermore, we offer an optimum generic training sequence for the joint estimation of symbol timing, frequency offset, and carrier phase for burst mode synchronization. The training sequence was obtained using the Cramér-Rao bounds

La télémesure aéronautique est un système utilisé pendant la phase des essais en vol qui consiste à transmettre en temps-réel les données mesurées à bord de l'avion vers une station sol par le biais d'une liaison radiofréquence. Elle est une opération critique visant à surveiller le comportement de l’avion et à garantir la sécurité du pilote. La conception et les performances des émetteurs et des récepteurs ont été améliorées au fil des années mais tout en gardant la même famille de modulation utilisée pour transmettre les données. Ces données sont modulées par une modulation à phase continue (CPM) car cette dernière possède une enveloppe complexe constante, ce qui permet l’emploi des amplificateurs de puissance dans leur régime de saturation sans distordre le signal. Cependant, contrairement aux modulations classiques, la modulation CPM n’est pas une fonction linéaire des symboles transmis, ce qui rend la tâche de la démodulation complexe surtout quand on considère des scenarii assez compliqués tels que la présence des multi trajets ou bien l’emploi d’un système multi-antennes. Dans cette thèse, on se focalise sur une modulation appelée « Shaped Offset Quadrature Phase Shift Keying telemetry group » (SOQPSK-TG) qui commence à être de plus en plus populaire en télémesure aéronautique. Cette modulation appartient à une famille particulière de CPM car elle transmet des symboles ternaires au lieu de binaires. Dans ce travail, nous développons différents algorithmes de réception pour cette modulation en considérant différents scenarii afin de garantir un lien de télémesure permanent. Les solutions proposées offrent des performances attrayantes tout en gardant une complexité raisonnable pour une implémentation en temps-réel
Aeronautical telemetry is a system used during the flight testing phase to monitor the behaviour of the plane by transmitting in real-time the aircraft dynamics from the aircraft to the ground station over a radio-frequency link. It is a critical process that requires highly reliable systems to ensure the pilot's safety. The design and the performance of the transmitting/ receiving equipment have been progressively upgraded over the years, and they mainly rely on continuous phase modulations (CPM) to convey the data. CPM is transmitter friendly because it has a constant complex envelope and therefore power amplifiers can be used at their saturation mode without distorting the signal. However, due to the nonlinear nature of this modulation, it can make the demodulation a hard task especially when considering complex scenarios such as the presence of multipath or the use of multiple antenna transmitters. In this PhD thesis, we focus on shaped offset quadrature phase-shift keying telemetry group (SOQPSK-TG) whose use is getting more and more popular in aeronautical. This modulation belongs to a particular CPM family since it transmits ternary symbols instead of binary ones. In this work, several reception algorithms have been developed for this modulation under different scenarios to ensure the availability of aeronautical telemetry link. These solutions offer excellent power efficiency without prohibitive complexity

Continuous phase modulation~(CPM) is a non-linear, constant-envelope modulation scheme with memory, known for its bandwidth and power efficiency. Multi-h CPM uses multiple modulation indices in successive symbol intervals to improve the error performance as compared to single-h CPM~(basic CPM that utilizes only a single modulation index). One of the major applications of multi-h CPM is in aeronautical telemetry systems. Modern aeronautical devices host an increasing number of sensors, which can transmit flight testing data to the ground station. However, this excess data transfer increases the intersymbol interference, and thus channel equalization is required at the receiver. The objective of our research is to propose low-complexity frqeuency-domain equalization~(FDE) techniques for multi-h CPM waveforms. For a modulation scheme with memory, such as CPM, the cyclic constraint on the FDE block necessitates the use of an extra segment of symbols, called intrafix or tail segment. We have used very simple geometric arguments to derive upper and lower bounds on the length of the intrafix in terms of the parameters of the modulation scheme and the Frobenius number. It is concluded that the length of the intrafix for multi-h CPM schemes is typically shorter than those required for single-h modulation schemes. We propose two receiver architectures; one uses a matched filter front end, while the other utilizes a fractional sampling front end. Various simplifications are proposed for each architecture, and the trade-off between receiver complexity and performance is analyzed and verified through detailed simulation studies.

Thesis (M.S.)--Texas A&M University, 2003.
"Major Subject: Electrical Engineering" Title from author supplied metadata (record created on Jul. 18, 2005.) Vita. Abstract. Includes bibliographical references.

Fiber-optic networks are continually evolving to accommodate the ever increasing data rates demanded by modern applications and devices. The current state-of-the art systems are being deployed with 100 Gb/s rates per wavelength while maintaining the 50 GHz channel spacing established for 10 Gb/s dense wavelength division multiplexed (DWDM) systems. Phase modulation formats (in particular quadrature phase shift keying - QPSK) are necessary to meet the spectral efficiency (SE) requirements of the application. The main challenge for phase modulated optical systems is fiber nonlinearities, where changes in intensity of the combined optical signal result in changes to the fiber's refractive index. Limiting launch power is the primary means to avoid dramatic intensity fluctuations, a strategy which in turn limits the available signal-to-noise ratio (SNR) within the channel. Continuous phase modulation (CPM) is a format in which data is encoded in the phase, while the amplitude is constant throughout all transmission (even during transitions between symbols). With the goal of reducing the impact of nonlinearities, the purpose of this research was to identify a set of CPM signals best suited for high speed fiber-optic transmission, and quantify their performance against other formats. The secondary goal was to identify techniques appropriate for demodulation of high speed fiber-optic systems and implement them for simulation and experimental research. CPM encompasses a number of variable parameters that combine to form an infinite number of unique schemes, each of which is characterized by its own SE, minimum distance, and implementation complexity. A method for computing minimum distance of DWDM-filtered CPM formats is presented and utilized to narrow down to a range of candidate schemes. A novel transmitter design is presented for CPM signal generation, as well as a number of novel reception techniques to achieve proper demodulation of the CPM signal from the coherent optical receiver. Using these methods, the identified range of candidate schemes was compared in simulation to the conventional QPSK format, showing that some modest gain can be expected from CPM. Through these and other simulations, it is revealed that fiber nonlinearities depend on the aggregate sum of all wavelengths rather than the imposition of each separate carrier on its neighbors. Therefore the constant envelope of CPM does not directly impact the nonlinearities since multiple carriers will photonically interfere and result in intensity fluctuations regardless of modulation format. Additionally, dispersive effects in fiber decompose the underlying channels so that the intensity throughout propagation is nearly Gaussian distributed, regardless of format. The benefits gained from CPM are thus limited to schemes that attain a higher minimum distance than alternative formats (in the given channel passband), and for optically compensated links in which low dispersion is maintained throughout the fiber link.

Aeronautical telemetry systems that transmit the same signal through multiple antennas from the flight vehicle suffer from severe link dropouts when the signals destructively interfere one with another at receiver. The underlying issue is a transmit array with element spacing far greater than the wavelength producing a transmit antenna pattern with significant and deep nulls. Recently, space-time coding techniques have been proven to resolve the issue for systems using linear modulations and shaped-offset quadrature shift keying (SOQPSK) modulation, a non-linear continuous phase modulation (CPM). This thesis examines application of space-time coding techniques to resolve the self-interference issue for another CPM modulation, the advanced range telemetry (ARTM) CPM. It is shown in this thesis among the two branches of space-time coding, space-time block coding (STBC) and space-time trellis coding (STTC), only the latter offers a solution for a full rate, low complexity, no hardware modification implementation. Various candidate STTCs are identified via simulation using the pair-wise error probability as a performance metric. One STTC is identified with trivial implementation costs and an error performance that is a function of code length.

International Telemetering Conference Proceedings / October 21, 2002 / Town & Country Hotel and Conference Center, San Diego, California
It is widely recognized that telemetry channels, particularly airborne channels, are afflicted by multipath propagation effects. It has also been shown that adaptive equalization can be highly effective in mitigating these effects. However, numerous other factors influence the behavior of adaptive equalization, and the type of modulation employed is certainly one of these factors. This is particularly true on modulations which exhibit different operating bandwidths. In this paper, we will examine the effect multipath and adaptive equalization for three modulation techniques which are either already in use, or have been proposed, for airborne telemetry.

La matière grasse laitière anhydre (MGLA) est composée à plus de 98 % de triglycérides. Leur diversité, liée à la nature des acides gras estérifiés, induit un comportement complexe de la MGLA lors de sa cristallisation. Par ailleurs, d’autres composés, dits mineurs, présents de manière endogène ou formés lors de procédés de transformation peuvent moduler le processus de cristallisation. Néanmoins, leurs modes d’action sont encore mal connus, rendant la maitrise du processus de cristallisation délicat. La mise en oeuvre de méthodes de caractérisation à différentes échelles (RMN, diffraction des rayons X, microscopie optique, analyse thermique différentielle, rhéologie) et à différentes températures permet de discriminer les mécanismes d’action des composés mineurs sur la cristallisation de la MGLA en phase continue ou dispersée. Ainsi, les composés mineurs modifient la cinétique de cristallisation de la MGLA, en agissant sur ses deux étapes, la nucléation et la croissance. En accélérant ou en ralentissant la vitesse de cristallisation, ils modifient la nature et/ou la structure du réseau cristallin et, éventuellement, les propriétés macroscopiques de la matière grasse. Les mécanismes d’action des composés mineurs sur la cristallisation de la MGLA dépendent de leur nature chimique (longueur de la chaine carbonée, degré d’insaturation, estérification) et de leur concentration. Lorsque la matière grasse est émulsionnée, son confinement et la création d’interfaces complexifient le processus de cristallisation : les composés mineurs modulent la cristallisation via l’interface ou la phase grasse selon leur affinité préférentielle pour l’une ou l’autre des deux phases de l’émulsion et induisent des propriétés thermiques et structurales spécifiques. Ce travail permet, donc, de mieux comprendre les facteurs qui modulent la cristallisation de la matière grasse et ainsi de mieux maitriser l’un des processus déterminants dans l’élaboration de la structure et de la fonctionnalité des produits laitiers
Anhydrous Milk Fat (AMF) comprises more than 98% triglycerides. The diversity of the esterified fatty acids induces a complex behavior of AMF during crystallization. In addition, other minor components, either endogenously present or added upon processing modulate AMF crystallization. Nevertheless, the mechanisms underlying their mode of action are still poorly understood, making the mastery of fat crystallization difficult. Several characterization methods (NMR, X-ray diffraction, optical microscopy, differential scanning calorimetry, rheology) were implemented to screen a large set of minor components at different scales and temperatures. The approach enabled to gain knowledge about the involved mechanisms, both in bulk and in the emulsified state. Minor components modify the kinetics of AMF crystallization through their impact on the two main stages of the process, nucleation and growth. By accelerating or slowing down the rate of crystallization, they modify the nature and/or the structure of the crystal lattice and, possibly, the macroscopic properties of AMF. The impact of minor components on AMF crystallization depend on their chemical nature (length of the carbon chain, degree of unsaturation, esterification) and on their concentration. When fat is emulsified, confinement and interfacial effects come into play. Minor components modulate crystallization via the oil/water interface or via the fat phase depending on their preferential solubility, which may induce specific thermal and structural properties. On the whole, this study enables a better understanding of the factors that modulate milk fat crystallization and provides useful guidances for a better control of this key process controlling the structure and function of dairy products

One of the major driving elements behind the explosive boom in wireless revolution is the advances in the field of modulation which plays a fundamental role in any communication system, and especially in cellular radio systems. Hence, the elaborate choice of an efficient modulation scheme is of paramount importance in the design and employment of any communications system. Work presented in this thesis is an investigation (study) of the feasibility of whether multilevel FSK modulation scheme would provide a viable alternative modem that can be employed in TDMA cordless communications systems. In the thesis the design and performance analysis of a non-coherent multi-level modem that offers a great deal of bandwidth efficiency and hardware simplicity is studied in detail. Simulation results demonstrate that 2RC pre-modulation filter pulse shaping with a modulation index of 0.3, and pre-detection filter normalized equivalent noise bandwidth of 1.5 are optimum system parameter values. Results reported in chapter 5 signify that an adjacent channel rejection factor of around 40 dB has been achieved at channel spacing of 1.5 times the symbol rate while the DECT system standards stipulated a much lower rejection limit criterion (25-30dB), implying that CP-QFSK modulation out-performs the conventional GMSK as it causes significantly less ACI, thus it is more spectrally efficient in a multi-channel system. However, measured system performance in terms of BER indicates that this system does not coexist well with other interferers as at delay spreads between 100ns to 200ns, which are commonly encountered in such indoor environment, a severe degradation in system performance apparently caused by multi-path fading has been noticed, and there exists a noise floor of about 40 dB, i.e. high irreducible error rate of less than 5.10-3. Implementing MRC diversity combiner and BCH codec has brought in a good gain.

Cette these presente une contribution, a la fois theorique et experimentale, a la mise en uvre des systemes optiques coherents a modulation de frequence et aux outils de modelisation et de caracterisation necessaires. Le traitement analytique des performances en presence de bruit de phase est base sur des hypotheses reductrices pas toujours satisfaites en pratique. Le developpement d'outils de simulation specifiques permet la prise en compte de l'ensemble des parametres physiques (filtrage moyenne frequence, bruit de phase et non-uniformite de l'excursion en frequence optique,. . . ) pour l'analyse et l'optimisation des performances. L'analyse du champ optique emis en regime de modulation necessite la mise en uvre d'outils de spectroscopie a haute resolution. Des techniques spectrometriques basees sur une photodetection coherente sont proposees pour la mesure de la largeur spectrale et la mesure de l'excursion dynamique en frequence optique. Un systeme experimental a ete developpe pour un debit binaire de 1 gbit/s. L'effet du filtrage fi sur la reduction de la gigue de phase, conduisant a une amelioration tres sensible des performances, a ete mis en evidence. L'influence des bruits du recepteur a ete analysee pour des lasers a faible largeur spectrale. Un seuil de sensibilite de 350 photons, soit 230 photo-electrons, est obtenu pour un taux d'erreur de 10##9, ce qui est au niveau des meilleurs resultats aujourd'hui publies

Si les drones militaires connaissent un développement important depuis une quinzaine d’année, suivi depuis quelques années par les drones civiles dont les usages ne font que se multiplier, en réalité les drones ont un siècle avec le premier vol d’un avion équipé d’un système de pilotage automatique sur une centaine de kilomètre en 1918. La question des règles d’usage des drones civiles sont en cours de développement malgré leur multiplication pour des usages allant de l’agriculture, à l’observation en passant par la livraison de colis. Ainsi, leur intégration dans l’espace aérien reste un enjeu important, ainsi que les standards de communication avec ces drones dans laquelle s’inscrit cette thèse. Cette thèse vise en effet à étudier et proposer des solutions pour les liens de communications des drones par satellite.L’intégration de ce lien de communication permet d’assurer la fiabilité des communications et particulièrement du lien de Commande et Contrôle partout dans le monde, en s’affranchissant des contraintes d’un réseau terrestre (comme les zones blanches). En raison de la rareté des ressources fréquentielles déjà allouées pour les futurs systèmes intégrant des drones, l’efficacité spectrale devient un paramètre important pour leur déploiement à grande échelle et le contexte spatiale demande l’utilisation d’un système de communication robuste aux non-linéarités. Les Modulations à Phase Continue permettent de répondre à ces problématiques. Cependant, ces dernières sont des modulations non-linéaire à mémoire entraînant une augmentation de la complexité des récepteurs. Du fait de la présence d’un canal multi-trajet (canal aéronautique par satellite), le principal objectif de cette thèse est de proposer des algorithmes d’égalisation (dans le domaine fréquentiel pour réduire leur complexité) et de synchronisation pour CPM adaptés à ce concept tout en essayant de proposer une complexité calculatoire raisonnable. Dans un premier temps, nous avons considéré uniquement des canaux sélectifs en fréquence et avons étudier les différents égaliseurs de la littérature. En étudiant leur similitudes et différences, nous avons pu développer un égaliseur dans le domaine fréquentiel qui proposant les mêmes performances a une complexité moindre. Nous proposons également des méthodes d’estimation canal et une méthode d’estimation conjointe du canal et de la fréquence porteuse. Dans un second temps nous avons montré comment étendre ces méthodes à des canaux sélectifs en temps et fréquence permettant ainsi de conserver une complexité calculatoire raisonnable.

International Telemetering Conference Proceedings / October 20-23, 2003 / Riviera Hotel and Convention Center, Las Vegas, Nevada
One of the charters of the Advanced Range Telemetry (ARTM) program was to develop more spectrally efficient waveforms while trying to maintain similar performance to the legacy waveform, Pulse Code Modulation/Frequency Modulation (PCM/FM). The first step toward this goal was the ARTM Tier I family of waveforms which include Feher patented, quadrature phase shift keying, -B version (FQSPKB) and shaped offset quadrature phase shift keying, Telemetry Group version (SOQPSK-TG). The final step was development of Tier II, an even more spectrally efficient waveform, multi-h Continuous Phase Modulation (CPM). This paper characterizes the performance of this waveform when applied in an airborne telemetry environment and, where appropriate, comparisons are made with existing Tier 0 and Tier I waveforms. The benefits, drawbacks, and trade-offs when applying this waveform in an airborne environment will also be discussed.

In this thesis, noncoherent differential demodulation of CPM signals with joint carrier frequency offset and symbol timing estimation is investigated. CPM is very attractive for wireless communications owing to major properties: good spectral efficiency and a constant envelope property. In order to demodulate the received CPM signal differentially, the symbol timing and the carrier frequency offset have to be estimated accurately. There are numerous methods developed for the purpose. However, we have not encountered studies (which are based on autocorrelation estimation and hence suitable for blind synchronization) that give expectable performance for both M-ary and partial response signaling. Thus, in this thesis we analyze a feedforward blind estimation scheme, which recovers the symbol timing and the frequency offset of M-ary CPM signals and partial response CPM signals. In addition, we surveyed low complexity symbol detection methods for CPM signals. Reduced state Viterbi differential detector incorporated to the joint frequency offset and symbol timing estimator is also examined. The performance of the examined demodulator scheme is assessed for the AWGN channel by computer simulations.

Recently, a study on obtaining better Euclidean distance for CPM (Continuous Phase Modulation) signals that fit well-known GSM spectral envelope has been carried out, and significant performance improvements were obtained. Two new pulse shapes, which are represented using 8th degree polynomials, were optimized to give the best error performance under the constraint that the PSD stays below GSM spectral standards. However, the approach uses parameters that cause the number of states to increase considerably, and thus yielding high complexity for receiver implementation. In this thesis, a study on finding a feasible receiver design that can provide a performance with acceptable degradation but affordable complexity is carried out for those new pulse shapes. After a survey about complexity reduction techniques, a decision is made to go on with a receiver structure based on Laurent Decomposition (LD) of phase modulated signals. Unlike other complexity reduction techniques, usage of LD based receivers permits reduction in both the number of matched filters and trellis states. Throughout the study, different numbers of matched filters and trellis states were used in LD based receivers for the new pulse shapes, and good results are obtained. For the pulse shape with pulse length L = 3, about a gain of 0.93dB in power is achieved by only 2 matched filters and 14 trellis states. For the case where L = 7, approximately a gain of 2.25dB is achieved with only 8 matched filters and 56 states, whereas 896 matched filters and 448 states are needed in the optimum case without complexity reduction.

碩士
國立海洋大學
電機工程學系
91
The multimedia data which includes image, audio and video is increasing with accompanying the progress of digital technology and the development of internet. If we want to transmit those multimedia data, we will need more transmissive bandwith and more storage space. For this reason, it is necessary to find a effective transmission technology to transmit data. OFDM using several subchannels to transmit data . The bandwidth of multicarrier communication system compare with single carrier system is much narrow. An OFDM sigansl consists of a number of independently modulated subcarriers, which can cause a large peak-to-average power. Large PAPR is a main problem of OFDM technique. Large PAPR make the system difficult to resist the nonlinear distortion, and limites it’s application. In thesis we utilize multiamplitude M-ary CPM in OFDM system. CPM is a modulation technique which hides information in the difference of phases. Using M-ary CPM can increase the data rate in wireless communication. Using multiamplitude technique in CPM-OFDM system can reduce the PAPR of OFDM signals, and avoiding the exceed peak signal affected by power amplifier’s nonlinear distortion. To reduce the calculation in the choice of multiamplitude, we have an ideal of real-time stop. This thesis also used Gray Code coding in CPM-OFDM system to reduce the bit-error-rate. CPM hides information in the difference of phase, so using multiamplitude M_ary CPM-OFDM in non-coherence system is better then using in coherence system. Multiamplitude CPM OFDM signal has lower PAPR, and it can resist the nonlinear distortion of the power amplifier, avoiding the spreading of external bandwidth when signals was clipped, and making the system fit in the standard.