Many sub-Saharan African languages are categorized as tone languages, and for the most part, they are classified as low-resource languages due to the limited resources and tools available to process these languages. Identifying the tone associated with a syllable is therefore a key challenge for speech recognition in these languages. We propose models that automate the recognition of tones in continuous speech that can easily be incorporated into a speech recognition pipeline for these languages. We have investigated different neural architectures as well as several feature extraction algorithms in speech (FBs (Filter Banks), LEAF (Learnable Frontend), CS (Cestrogram), MFCC (Mel-Frequency Cepstral Coefficients)). In the context of low-resource languages, we also evaluated W2V (Wav2vec 2.0) models for this task. In this work, we use a public speech recognition dataset on Yoruba. As for the results, using the combination of features obtained from CS and FBs, we obtain a minimum TER (Tone Error Rate) of 19.54%, whereas for the evaluations of the models using W2V, we have a TER of 17.72%, demonstrating that the use of W2V provides better performance than the models used in the literature for tone identification on low-resource languages.

Adaptive video streaming improves users' quality of experience (QoE), while using the network efficiently. In the last few years, adaptive video streaming has seen widespread adoption and has attracted significant research effort. We study a dynamic system of random arrivals and departures for different classes of users using the adaptive streaming industry standard DASH (Dynamic Adaptive Streaming over HTTP). Using a Markov chain based analysis, we compute the user QoE metrics: probability of starvation, prefetching delay, average video quality and switching rate. We validate our model by simulations, which show a very close match. Our study of the playout buffer is based on client adaptation scheme, which makes efficient use of the network while improving users' QoE. We prove that for buffer-based variants, the average video bit-rate matches the average channel rate. Hence, we would see quality switches whenever the average channel rate does not match the available video bit rates. We give a sufficient condition for setting the playout buffer threshold to ensure that quality switches only between adjacent quality levels.

HTTP Adaptive Streaming is able to dynamically match video quality to variable network conditions. This is a key feature for multimedia delivery when quality of service cannot be granted network-wide. For instance, the end-to-end throughput towards mobile terminals may suffer short term fluctuations due to fading. Hence, robust bitrate adaptation schemes become crucial in order to avoid degraded video reproduction. The objective, in this context, is to control the filling level of the playback buffer, maximize the video quality, and avoid unnecessary quality variations which may also impair the perceived quality of experience. In this work we study bitrate adaptation algorithms leveraging on Backward-Shifted Coding (BSC), a scalable video coding scheme able to cope with the effects of end-to-end throughput fluctuations. We have proposed a new adaptation scheme able to balance video rate smoothness and high network capacity utilization. Both the throughput-based and buffer-based variants of such scheme have been designed. Extensive simulations using synthetic and real-world video traffic traces show that the proposed solution performs remarkably well even under challenging network conditions.

In future networks, Radio Resource Management (RRM) could benefit from Geo-Localized Measurements (GLM) thanks to the Minimization of Drive Testing (MDT) feature introduced in Long Term Evolution (LTE). Such measurements can be processed by the network and be used to optimize its performance. The purpose of this paperais to use GLM to significantly improve scheduling. We introduce the concept of forecast scheduler for users in high mobility that exploits GLM. It is assumed that a Radio Environment Map (REM) can provide interpolated Signal to Interference plus Noise Ratio (SINR) values along the user trajectories. The diversity in the mean SINR values of the users during a time interval of several seconds allows to achieve a significant performance gain. The forecast scheduling is formulated as a convex optimization problem namely the maximization of an α-fair utility function of the cumulated rates of the users along their trajectories. Numerical results for thee different mobility scenarios illustrate the important performance gain achievable by the forecast scheduler.

The goal of this paper is to study the behaviour of viewcount in YouTube. We first propose several bio-inspired models for the evolution of the viewcount of YouTube videos. We show, using a large set of empirical data, that the viewcount for 90% of videos in YouTube can indeed be associated to at least one of these models, with a Mean Error which does not exceed 5%. We derive automatic ways of classifying the viewcount curve into one of these models and of extracting the most suitable parameters of the model. We study empirically the impact of videos' popularity and category on the evolution of its viewcount. We finally use the above classification along with the automatic parameters extraction in order to predict the evolution of videos' viewcount.

Our purpose in this paper is to characterize buffer starvations for streaming services. The buffer is modeled as a FIFO queue with exponential service time and Poisson arrivals. When the buffer is empty, the service restarts after a certain amount of packets are prefetched. With this goal, we propose two approaches to obtain exact distribution of the number of buffer starvations, one of which is based on Ballot theorem, and the other uses recursive equations. The Ballot theorem approach gives an explicit result. We extend this approach to the scenario with a constant playback rate using Tàkacs Ballot theorem. The recursive approach, though not offering an explicit result, allows us to obtain the distribution of starvations with non-independent and identically distributed (i.i.d.) arrival process in which an ON/OFF bursty arrival process is considered. We further compute the starvation probability as a function of the amount of prefetched packets for a large number of files via a fluid analysis. Among many potential applications of starvation analysis, we show how to apply it to optimize objective quality of experience (QoE) of media streaming, by exploiting the tradeoff between startup/rebuffering delay and starvations.

Social scientists have observed that human behavior in society can often be modeled as corresponding to a threshold type policy. A new behavior would propagate by a procedure in which an individual adopts the new behavior if the fraction of his neighbors or friends having adopted such behavior exceeds some threshold. In this paper we study the question of whether the emergence of threshold policies may be modeled as a result of some rational process which would describe the behavior of non-cooperative rational members of some social network. We focus on situations in which individuals take the decision whether to access or not some content, based on the number of views that the content has. Our analysis aims at understanding not only the behavior of individuals, but also the way in which information about the quality of a given content can be deduced from view counts when only part of the viewers that access the content are informed about its quality. In this paper we present a game formulation for the behavior of individuals using a meanfield model: the number of individuals is approximated by a continuum of atomless players and for which the Wardrop equilibrium is the solution concept. We derive conditions on the problem's parameters that result indeed in the emergence of threshold equilibria policies. But we also identify some parameters in which other structures are obtained for the equilibrium behavior of individuals.

Our purpose in this paper is to obtain the exact distribution of the number of buffer starvations within a sequence of N consecutive packet arrivals. The buffer is modeled as an M/M/1 queue. When the buffer is empty, the service restarts after a certain amount of packets are prefetched. With this goal, we propose two approaches, one of which is based on Ballot theorem, and the other uses recursive equations. The Ballot theorem approach gives an explicit solution, but at the cost of the high complexity order in certain circumstances. The recursive approach, though not offering an explicit result, needs fewer computations. We further propose a fluid analysis of starvation probability on the file level, given the distribution of file size and the traffic intensity. The starvation probabilities of this paper have many potential applications. We apply them to optimize the quality of experience (QoE) of media streaming service, by exploiting the tradeoff between the start-up delay and the starvation.

This paper studies distributed choice of retransmission probabilities in slotted ALOHA. Both the cooperative team problem as well as the noncooperative game problem is considered. In previous work that has focused on the maximization of throughput, it was shown that in heavy load, this maximization is obtained at the cost of a huge delay of backlogged packets. This motivates us to investigate the delay minimization problem as well as the multicriterion problem of minimizing the average expected delay (or maximizing the throughput) subject to constraints on the expected delay of backlogged packets. A Markov chain analysis is used to obtain optimal and equilibrium retransmission probabilities and expected delays analysis.

In a previous paper, BonaId and Roberts (June 2001) studied non-persistent TCP connections in transient overload conditions, under the assumption that all connections have the same round-trip times. In this paper our goal is to develop theoretical tools that will enable us to relax this assumption and obtain explicit expressions for the rate of growth of the number of connections at the system, the rate at which TCP connections leave the system, as well as the time needed for the completion of a connection. To that end, we model the system as a DPS (discriminatory processor sharing) system which we analyze under very mild assumptions on the probability distributions related to different classes of arrivals: we only assume that the arrival rates of connections exist, and that the amount of information transmitted during a connection of a given type forms a stationary ergodic sequence. We then proceed to obtain explicit expressions for the growth rate of the number of connections at the DPS system for several specific probability distributions. We check through simulations the applicability of our queueing results for modeling TCP connections sharing a bottleneck.

In this paper, we consider call admission control of multiple classes without waiting room. We use event-based dynamic programming for our model. We show that sometimes the customer classes can be ordered: if it is optimal to accept a class, then to accept a more profitable class is optimal too. We demonstrate submodularity of the minimum cost for the 2-classes problem and establish some properties of optimal policies. Then we formulate a fluid model that allows us to study the optimal control for the large-capacity case. We show that in the case of same service time distributions, the control problem can be reduced to a model with a one-dimensional (1-D) state space, and a trunk reservation policy is optimal. We present numerical examples that validate our results.

We study a class of noncooperative general topology networks shared by N users. Each user has a given flow which it has to ship from a source to a destination. We consider a class of polynomial link cost functions, adopted originally in the context of road traffic modeling, and show that these costs have appealing properties that lead to predictable and efficient network flows. In particular, we show that the Nash equilibrium is unique, and is moreover efficient, i.e., it coincides with the solution of a corresponding global optimization problem with a single user. These properties make the cost structure attractive for traffic regulation and link pricing in telecommunication networks. We finally discuss the computation of the equilibrium in the special case of the affine cost structure for a topology of parallel links.

Currently, PVM constitutes a widely used software for developing parallel applications in workstation and parallel environments. In this paper we propose a processors management system for PVM which allows to assign the PVM tasks over a computers system. The Processors Management System uses two task assignment heuristics. These heuristics are based on Neural Networks and Genetic Algorithms. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.