Effects of quasi-periodic training signals on acoustic echo canceller performance

Filtered white noise is the training signal typically used to test the performance of echo cancellers. However, a system optimized for operation with such training signals may perform poorly when trained instead with voiced speech. This performance difference is a consequence of the quasi-periodic nature of voiced speech. In this paper, we explore the effects of quasi-periodic training signals on echo-canceller performance. In particular, we show that unlike filtered white noise, quasi-periodic training signals can actually limit the asymptotic echo-return loss of echo cancellers. In addition, we examine the ability of several existing techniques to improve performance when quasi-periodic signals are used to train echo cancellers.     

TMN implementation strategy based on OSI management standards

In order to facilitate the implementations ofTMN interface protocols/services studied inITU-T, it is very important to define profiles for supportingTMN management service. This paper proposes a concrete method for achieving this based on osi management standards as a promisingTMN implementation method. It proposes an idea of structuring theTMN ISP’S based on the structure of the osi managementISP’S. The paper discusses aTMN based on the osi managementISP’S. Finally the implementation as software is discussed and a software architecture for efficient application development is proposed.     

A management architecture for the international B-ISDN

This paper relates to basic management aspects of the interworking of national?-ISDNX. A general architecture of such an interconnection is proposed which is based on concepts of the worldwide intelligent network (WIN) and CCITT telecommunication management network (TMN). Various models supporting theTMN design are considered. Communication aspects of management information transfer within theTMN are discussed in more detail.     

Simplified weighted least squares design of 2-D FIR digital filters

A simple easy to use and very flexible approach is presented for the design of 2-D fir digital filters. It is illustrated by two examples, a rectangular filter and a diamond filter. Appropriate selection of the weighting function which is used in the weighted least squares method can yield very diverse characteristics, like the possibility to meet specifications on both frequency response and step response.     

Migration strategy towards Q3 interfaces of switching systems : a look at France Telecom switches

Future telecommunication management network (TMN) will be based upon object-oriented modelling of interfaces. Nevertheless, in case of public switched telephone network, this process will be long and difficult. However, as current interfaces are not completely satisfactory, some evolution steps from today’s interfaces towards tomorrow’sTMN ones are planned. This paper presents the different steps chosen by France Telecom. At first, introduction of a new protocol as close as possible to the target Q3 is detailed. Unification of the different heterogeneous commands is then shown through creation of standard commands, last, integration toTMN through Q-adaptor is presented.     

The information layered architecture model,layers, methodology and their application for service activation

The authors propose a model (information layered architecture ILA) forTMN layering, that is based on grouping managed objects that share a common information perspective. The model provides an interfacing mechanism between layers that is based on referencing techniques between managed objects. We identified four levels ofTMN abstractions that can be used to represent a wide range of telecommunication processes. These are : customer, service, network and network element management. We then use the ILA to develop a model for the process of service activation. By doing so we demonstrate how modelling can be used as a technique to manage complicated processes.     

Dynamique des lasers à fibre dopée à I’erbium

The dynamics of the erbium doped fiber laser is analysed both experimentally and theoretically. This laser operates spontaneously in self-pulsing, in sinusoidal or in CW regime. Experiments have allowed to determine the control parameters : the cavity losses, the pumping rate and the ion pairs concentration. An antiphase dynamics is observed when the laser operates simultaneously at 1.55 µn and 1.536 µm. The self-pulsing behavior is attributed to a fast energy transfer which occurs between two neighboring ions (pair induced quenching). A simple model considering the active medium as a mixture of isolated ions and ions pairs is developped and leads to a good qualitative agreement with the experimental results.     

A fast exact FTF adaptive algorithm

In this paper we present a new method for the efficient implementation of the fast transversal filter(ftf) algorithm. Reduction of the arithmetic complexity is obtained by making use of the redundancy in the successive computations of the forward prediction error and the filtering error in the joint process. The resulting algorithm is exactly equivalent to the originalftf algorithm, hence retaining the same theoretical convergence characteristics and offering the least squares(ls) estimate at each recursion step without delay. Furthermore, the algorithm can be numerically stabilized by using a simple and effective stabilization measure which needs only one additional multiplication per recursion step. The equivalence of the proposed algorithm to the originalftf algorithm is demonstrated by simulations of an acoustic room impulse response identification.     

Synthesis methods of array antenna excitation laws

Standard numerical optimization methods become very computational time consuming when dealing with synthesis of array antenna excitation laws, especially when number of elements exceeds a few tenths. Array antenna designer needs a fast and efficient synthesis tool, since the design phase implies several reoptimizations within the definition phase, in order to satisfy tradeoffs on antagonistic requirements : performances, size, cost,… We propose the use of an efficient, from the computational time point of view, and secured, from the convergence point of view, synthesis method illustrated by two examples. The optimized solution satisfies a sta-tionarity condition, which leads to steady solutions in frequency but also with regard to errors (manufacturing, …). Based on analytical calculus performed before any numerical resolution and together with the use of fast Fourier transform algorithms, this method leads to reduced optimization times (in the order of a minute) for arrays of some tenths up to some hundred feeds, computations being performed on a pc. Two theoretical applications are presented and for each of them a computed example is also shown; moreover possible extensions are considered.     

Modeling of spontaneous emission noise in erbium-doped fibre amplifiers using the Green functions method

This article describes a study of the main characteristics of erbium-doped fibre amplifiers based on a non-quantum approach to the phenomenon of amplification in spatially inhomogeneous structures. The noise figure in particular is assessed theoretically and numerically on the basis of various amplifier parameters, including the reflection coefficients at the fibre input and output. The results show the importance of monitoring reflectivity at the amplifier input and the power of the incoming signal.     

Transmission and distribution of a multichannel AM-VSB TV signal on a single mode optical fiber for CATV videocommunication networks

First of all, in this paper, we situate and recall the advantage of single mode fiber for transmission of a multichannel AM-VSB TV signal. Then we describe the prototype of a multichannel transmission system studied under a France Télécom contract and discuss the experimental results obtained on this system. Today the equipment issued from that study, is in current industrial development in order to be soon installed in networks. A few rules are given in terms of range, number of channels per fiber and number of subscribers to help the design of networks. After having identified the system limitations, solutions are proposed to improve the performances. Particularly the concept of a am-vsb optical repeater and the linearity improvement with that procedure are shown. We will also discuss the use of 1.54 μm dfb laser and compare the performances at both wavelengths. The interest of predistortion network to escape from the previously identified linearity constraint is justified. Then the extension of transport network to optical distribution network closer to the subscriber (fttc) is envisaged. We show how the trade-off between less stringent performances and higher distribution potentiality can be handled. At last the interest of external modulation in order to escape from the noise limitation and to extend the coverage of the network at both wavelengths is discussed.     

The fast subsampled-updating fast transversal filter (FSU FTF) RLS algorithm

We present a new fast algorithm for Recursive Least-Squares(rls) adaptive filtering that uses displacement structure and subsampled updating. Thefsu ftf algorithm is based on the Fast Transversal Filter(ftf) algorithm, which exploits the shift invariance that is present in therls adaptation of afir filter. Theftf algorithm is in essence the application of a rotation matrix to a set of filters and in that respect resembles the Levinson algorithm. In the subsampled updating approach, we accumulate the rotation matrices over some time interval before applying them to the filters. It turns out that the successive rotation matrices themselves can be obtained from a Schur type algorithm which, once properly initialized, does not require inner products. The various convolutions that thus appear in the algorithm are done using the Fast Fourier Transform(fft). For relatively long filters, the computational complexity of the new algorithm is smaller than the one of the well-known lms algorithm, rendering it especially suitable for applications such as acoustic echo cancellation.     

A static study of multilayered multiconductor transmission lines

Accurate knowledge of passive MIES and MMIGS has become as important as knowledge of active components. A rather classical method of analysis of multi-dielectric multistrip transmission lines is presented. This analysis, restricted to the quasi-static approach, is based on a variational method in the discrete spectral domain, combined with the transverse transmission line method devoted to the calculation of Green’s functions. This work, devoted only to the study of the distributed parameters of the structures, i.e. in any transverse section, is followed by a complete modal analysis with respect to the direction of propagation and the possible boundary conditions at the terminals. These calculations consist actually mostly of an eigenvalue problem. They are carried out analytically as far as possible by means of a block matrix algebraic formalism particularly well suited to the problem. The mode impedances and the impedance matrix are obtained analytically in a general manner. All the parameters of new structures with four conductors are completely made explicit.     

Codage CRC parallèle pour cellules ATM

In this paper, it is shown that a parallel approach for CRC coding can be derived for a number of bits less than, or equal to, the degree of the code generator polynomial, or for larger values as well. In both cases if formulas are different, the logic diagrams are very similar and corresponding circuits are prone to efficient VLSI integration.     

Comparison of virtual path bandwidth assignment and routing methods

The virtual path (VP) can simplifyAtm network management by minimizing connection routing and admission costs, and by facilitating the layered control of resources. However, fully exploiting these advantages may lead to a large number of relatively low capacity virtual paths travelling on each physical link. If each VP is treated as a separate unit, as is commonly assumed, low path capacities will lead to low network utilisation. This paper carefully examines the trade-off between simplification through traffic separation and improved efficiency due to traffic consolidation. We review existing vp bandwidth assignment and control techniques, and propose a new vp tagging control method. A comparison shows that by permitting resource sharing between paths it is possible to influence significantly the trade-off between simplified network management and multiplexing gain from traffic consolidation.     

The hands-free telephone problem: an annotated bibliography update

Providing means for a full-duplex hands-free telephone conversation is still a research topic in many communication and signal processing laboratories around the world. More than ninety publications mainly from the last two and a half years support this statement. The reason for these activities lies in the complexity of the problem: full-duplex telephone communication without hand-sets requires electronic replicas of the loudspeaker-enclosure-microphone systems used by the communicating parties. The impulse response of such a system is typically several 100 ms long and time varying. Therefore, the replica has to be adaptive while the adaptation is controlled by speech signals. Regarding these requirements the hands-free telephone problem may currently be considered as one of the most challenging signal processing problems. This bibliography supplements a bibliography composed approximately two and a half years ago [29]. Brief summaries are given on papers published since then.     

On the stochastic timed Pétri nets model and its application to the DQDB protocol

The goal of this paper is twofold : at first, to present the stochastic timed Pétri nets model by emphasizing its ability to express the main characteristics of real time distributed systems (parallelism, synchronization by message exchange, time attributes and in particular time constraints) and to allow both qualitative and quantitative analysis (concept of randomized states graph); second, to show a modeling methodology based on this model that allows a rigorous modeling of theDqdb protocol (Qa access, bwb mechanism role,Pa access influence). A new contribution to theQa mechanism has been done (1).     

Transport layer for cooperative multimedia applications

This paper presents work carried out within the Cesame project for the design of a multimedia highspeed distributed transport provider. The term transport provider is used to target a reference framework but does not preclude any standardized reference model. It gives the objectives and relates the main contributions obtained during the first year of the project. The goal of this task is to monitor, select and design services, protocols and mechanisms for defining a transport provider to support distributed multimedia services. As far as architectural aspects are concerned, a refinement of reference models is mandatory, and work on this topic has already been started within several institutions. Moreover, it is worth to mention that the transport provider will have first to run on top ofAtm/Aal networks and is expected on top of other networks.     

Perfect sequence excitation of the NLMS algorithm and its application to acoustic echo control

Adaptive echo cancellers are currently being studied for applications such as audio teleconference systems or hands-free telephone sets with high speech quality. The purpose of the echo control is to eliminate the acoustic feedback from the loudspeaker to the microphone. One problem of echo cancellers using e.g. the normalized least mean square algorithm(nlms) for the adaptation of the coefficients is that the convergence properties degrade with colored signal input such as speech signals [8, 9, 17]. One approach to accelerate the convergence speed is to introduce linear prediction filters in order to decorrelate the speech signal [1, 2, 12, 15, 20]. This paper presents a new approach, named the excited lms algorithm orelms algorithm, which prewhitens the input signal applying perfect sequences. Coincidently, the proposed algorithm can be interpreted as a combination of the conventionalnlms algorithm and a system identification approach using m-sequences or related sequences.     

Modelling buffer admission mechanisms using stochastic automata networks

The stochastic automata networks formalism is an attractive technique to model complex systems with interacting components. Each component of the system is modelled by a single automaton; interactions between components are modelled by labels on the arcs which may represent synchronization and state-dependent transitions. Every automaton is associated with some matrices which allow to build the transition matrix of the underlying Markov chain, using tensor algebra. To illustrate this methodology, we introduce two buffer policies which could be used inAtm switching node. Every policy manages two priority levels which have distinct cell loss requirements. The first buffer policy is based on the push-out mechanism : a high priority cell replaces a low priority cell when the buffer is full. The second policy causes the discarding of all the low priority cells when the user transmits a request to send a burst of cells. In both studies, we compute the loss probabilities of each type of cells under various assumptions.