Eduardo Parente Ribeiro

Minimum delay path selection in multi-homed systems with path asymmetry

As mobile and fixed hosts are increasingly equipped with multiple network interfaces that enable multi-homing, there is significant current interest to seek appropriate method for access network selection. To address the presence of asymmetric forward-return path combinations in practice, we propose a simple but novel scheme to determine the one-way path with the lowest delay independently for both directions. This method provides delay sensitive multimedia applications with the lowest possible round-trip time in a multi-homed environment, compared to existing methods which assume path symmetry.

Total harmonic distortion calculation by filtering for power quality monitoring

Measuring and monitoring quality parameters of AC power systems requires several calculations, such the total harmonic distortion (THD) of voltage and current. This calculation is performed with samples of the monitored waveforms, at sample frequency equal to a power of two multiple of the frequency of waves. The samples are converted to digital values by analog-to-digital converters, with a finite number of bits. Numeric algorithms applied to these digital values insert some errors in the final results, due to the number of bits used in calculations. The most used algorithm to obtain the THD is FFT. Conventional 16-bit DSP FFT algorithms use only 16-bit calculations, despite the DSP 40-bit hardware accumulator. The results are errors in amplitude of each harmonic order. Using these amplitudes to calculate the THD, these errors are summed, and the final error is very greater. The purpose of this article is to present an algorithm to calculate the THD using the DSP 40-bit hardware accumulator and resulting more accurate results.

Zero crossing determination by linear interpolation of sampled sinusoidal signals

Measuring and monitoring quality parameters of AC power systems requires several calculations, such the actual frequency and relative phase angles between each one of the three phases. These calculations must result in accurate values to comply with the function requirements; for synchronism function, frequency inaccuracy must be less than 0.01 Hz, and phase angle inaccuracy must be less than one degree. To perform these calculations, the zero crossing instants must be determined for each monitored signal. Conventional methods use a hardware circuit to perform these determinations resulting a high-cost system, because each monitored signal must have its own circuit. Averaging is not a good alternative due to the long time required besides there is no improvement in phase angle accuracy. The purpose of this article is to introduce a low-cost one cycle method to perform these calculations by software in a DSP using samples of the monitored signals.

The application of neural networks to improve the quality of experience of video transmission over IP networks

The transmission of real-time multimedia streams requires service guarantees, such as limited packet loss, minimum bandwidth and low delay and jitter, to ensure a good quality of experience (QoE) for viewers. The spatial and temporal redundancy of videos is addressed by coding algorithms that reduce the amount of information necessary to represent the images. As a consequence, multimedia traffic commonly presents variable bit rate behavior and self-similar characteristics. Although the reduction in bandwidth requirements is highly desirable, the burstiness of traffic leads to problems in network design and performance prediction. Even a low level of packet loss could severely affect the viewer QoE. In this paper, we propose a real-time packet payload classifier, implemented with artificial neural network (ANN) to be used at network routers. A priority packet discard strategy can be implemented to avoid discarding packets that carry the most relevant information for image reconstruction, thus improving the perceived quality. This approach does not require changes at the video source to classify outgoing packets. The ANN was employed because of its good capacity in temporal series recognition and the possibility of its implementation in real-time systems due to its low computational complexity. The video traces used for training and validation were encoded with H.264/MPEG-4 Advanced Video Coding and are publicly available. The priority packet discard strategy was tested through computational simulations. The QoE was estimated comparing the peak signal-to-noise ratio (PSNR) of original and the received frames of video, and the results indicate that the proposed method improves the QoE. The implementation does not require packet payload processing and can be performed with network layer information only. HighlightsPacket loss in video transmission severely degrades video quality.We propose a real-time packet classifier for video MPEG-4 implemented with neural networks.Priority discard strategy can be used to avoid discarding important packets.We show that MPEG-4 video quality can be improved by discarding classified packets.

Asymmetric path delay optimization in mobile multi-homed SCTP multimedia transport

Multimedia applications are very demanding of quality parameters from network provision, especially in a mobile wireless scenario. End-to-end delay is a primary concern for real-time applications. Much work have been done recently in order to fully exploit the multi-homing character of emerging wireless mobile systems addressing both communication performance as well as seamless handover. We propose an algorithm for path selection based on smallest delay that takes into account the dynamic asymmetry over the available routes. An example is provided showing how a multimedia application could benefit from this approach.

Predictive delay-centric handover for video streaming over SCTP

Video streaming is one of the most important applications of Internet Protocol (IP) networks nowadays. The growth in demand for video streaming services with higher resolution and quality pushes the development of techniques to deliver the best possible quality to users. The Stream Control Transmission Protocol (SCTP), with its ability to support multi-homing systems, has been proposed as an alternative to the transport layer protocols in use. In this article, we propose a new handover strategy using partial reliability extension of SCTP and Moving Average Convergence Divergence (MACD). This strategy is designed to preserve the quality of video streaming over IP networks. The proposed method is based on forecasting the end-to-end delay and switching the primary path before an increase in delay impairs the quality of the video. The performance of the proposed method is evaluated through computer simulations which considers handover between ADSL, WiFi, and 3G networks. The results show that the handover scheme provides better quality of experience for low- and medium-motion videos and no worse performance for high-motion videos.

Preventing quality degradation of video streaming using selective redundancy

Video streaming has already become the main portion of data traffic over the Internet. However, it is hard to guarantee the user-perceived video quality in single access networks due to the limitation in available bandwidth. Multihomed systems enable concurrent multipath data transfer over multiple network paths. Path control can be implemented at the transport layer such as in Stream Control Transport Protocol (SCTP). SCTP has built-in support for multihoming, but offers only a limited standard path selection mechanism. In this paper a new method for video streaming over SCTP is proposed. Secondary paths are used to send redundant information to prevent degradation of viewers perceived quality. The efficiency of the proposed method is demonstrated by computer simulations using video coded with H.264/MPEG-4 AVC (Advanced Video Coding). The results show that the proposed method can prevent degradation of video quality at the cost of small extra bandwidth usage.

On the impact of adaptive RED in IP networks transporting H.264/MPEG-4 AVC video streams

This paper investigate the impact on the QoS offered by IP networks that transport real-time H.264/MPEG-4 AVC video streams when applying adaptive random early detection (ARED) as an active queue management technique instead of traditional drop-tail. Simulation results indicate that the ARED technique reduces the loss of video packets but degrades a higher number of video frames if compared to the drop-tail technique, while showing a small qualitative gain regarding delay and jitter. From these results it may be reasonably concluded that H.264/MPEG-4 AVC video streams do not benefit from employing ARED technique as VoIP streams do.

One cycle AC RMS calculations for power quality monitoring under frequency deviation

Measuring and monitoring quality parameters of AC power systems requires several calculations, such as RMS values of voltage and current. These calculations are performed with samples of the monitored waveforms, at a sample frequency equal to N times the nominal frequency of waves. Conventional algorithms regard N as a constant integer. When the actual frequency has a deviation from its nominal value, errors are introduced in the calculations, and averages are used over several cycles to reduce these errors. However, for instantaneous quality parameters, these calculations must be performed every cycle for quick disturbance detection; averages are impracticable. The purpose of this article is to introduce a method to minimize these errors making N variable and a noninteger.

An unified VoIP model for workload generation

This paper presents a new model for VoIP workload generation. The novelty of our proposal consists in modeling the sessions by characterizing both the user behavior (session level) and the packet generation for an active call (intra-session level) with easily measured parameters and low computational complexity. This approach also facilitates systematic study of changes in user behavior and voice codec. The session level was modeled by analysis of call-holding time and time interval between successive calls. The model for call-holding time, characterizing the individual user behavior, uses the Pareto type 2 probability distribution. The time interval between calls is obtained from aggregate traffic and can be modeled by exponential probability distribution. Aggregate traffic is obtained by superposition of simultaneous sessions. The data used to characterize the session level were collected at the backbone of two Brazilian telecommunication carriers. The model for intra-session level comprises the characterization of the packet size and the packet inter-arrival time. The intra-session model was based on data generated in a laboratory environment, in order to properly characterize the codec influence on packet generation and to avoid the effects of delay, jitter and loss commonly present in an operational network. Models for constant bit rate and variable bit rate codecs were considered. A simulator was implemented and the results indicate that our model properly mimics the characteristics observed in real traffic and can be used for VoIP modeling and workload generation. Additionally, an application to automate the performance analysis was developed.