The H.264 standard achieves much higher coding efficiency than the MPEG-2 standard, due to its improved inter-and intra-prediction modes at the expense of higher computational complexity. Transcoding MPEG-2 video to H.264 is important to enable gradual migration to H.264. However, given the significant differences between the MPEG-2 and the H.264 coding algorithms, transcoding is a much more complex task and new approaches to transcoding are necessary. The main problems that need to be addressed in the design of an efficient heterogeneous MPEG-2/H.264 transcoder are: the inter-frame prediction, the transform coding and the intra-frame prediction. In this paper, we focus our attention on the inter-frame prediction, the most computationally intensive task involved in the transcoding process. This paper presents a novel macroblock (MB) mode decision algorithm for P-frame prediction based on machine learning techniques to be used as part of a very low complexity MPEG-2 to H.264 video transcoder. Since coding mode decisions take up the most resources in video transcoding, a fast MB mode estimation would lead to reduced complexity. The proposed approach is based on the hypothesis that MB coding mode decisions in H.264 video have a correlation with the distribution of the motion compensated residual in MPEG-2 video. We use machine learning tools to exploit the correlation and construct decision trees to classify the incoming MPEG-2 MBs into one of the several coding modes in H.264. The proposed approach reduces the H.264 MB mode computation process into a decision tree lookup with very low complexity. Experimental results show that the proposed approach reduces the MB mode selection complexity by as much as 95% while maintaining the coding efficiency. Finally, we conduct a comparative study with some of the most prominent fast inter-prediction methods for H.264 presented in the literature. Our results show that the proposed approach achieves the best results for video transcoding applications.

The objective of this study was to evaluate the potential use of Near-Infrared Spectroscopy (NIRS) for the analysis of oil content, moisture and fatty acids composition in intact olive fruit. A total of 287 samples, each from a single plant from an olive breeding program, were scanned by NIRS between 400 and 1700 nm. Partial least squares (PLS) regression was used to create calibration models (with 70 % of samples) relating laboratory reference values to spectral data (original, first and second derivative spectral data). The best equations obtained were validated (with 30% of samples) showing values of r2 of 0.88 % for the moisture, 0.83 % for oil content, 0.77 % for oleic acid content and 0.81 % for linoleic acid content. Therefore a reliable and accurate preselection can be made by using NIRS for both oil content and oleic acid content, with a nondestructive analysis, in a few seconds and without use neither production of chemical reagents.

The ever-increasing demand for mobile content delivery and multimedia services is bringing renewed interest in multicast communications in Wi-Fi based WLANs. Nevertheless, multicast over Wi-Fi raises several challenges including low data rates and coexistence issues with other unicast streams. Some amendments to the Wi-Fi standard, such as 802.11aa, have introduced new delivery schemes for multicast traffic as well as finer control on the low-level aspects of the 802.11 medium access scheme. However, the logic for using such features is left to the implementer of the standard. In this paper, we present SDN@Play Mobile, a novel SDN-based solution for joint mobility management and multicast rate-adaptation in Wi-Fi networks. The solution builds upon a new abstraction, named Transmission Policy, which allows the SDN controller to reconfigure a multicast transmission policy when its optimal operating conditions are not met. An experimental evaluation carried out over a real-world testbed shows that our approach can deliver significant improvements in terms of both throughput and channel utilization compared to the legacy 802.11 multicast scheme. Finally, we release the entire software implementation under a permissive APACHE 2.0 license for academic use.

Most of the reported distributed video coding (DVC) architectures have a serious limitation that hinders its practical application. The uses of a feedback channel between the encoder and the decoder require an interactive decoding procedure which is a limitation for applications such as offline processing. On the other hand, the decoder needs an efficient way to estimate the probability of error without assuming the availability of the original video at the decoder. In this paper we continue with our previous works into a more practical DVC architecture which solves both problems based on the use of machine learning. The proposed approach is based on extracting the relationships that exist between the residual frame and the number of requests over this feedback channel. We apply these concepts to pixel-domain Wyner-Ziv coding demonstrating significant savings in bitrates with a little loss of quality.

The original version of the IEEE 802.11 standard is not able to provide the required Quality of Service (QoS) for real-time applications. The IEEE 802.11e amendment was developed to overcome this situation, introducing the Enhanced Distributed Channel Access (EDCA) as a new channel access method. This method makes it possible to prioritize the different types of traffic through a group of Medium Access Control (MAC) parameters. The most important role of these MAC parameters is played by the Arbitration Inter-Frame Space Number (AIFSN). Although the AIFSN can be modified during a transmission, the commercial Access Points (APs) only use the combination defined in the standard. Therefore, we propose a new adaptation scheme for network traffic priorities involving the construction of a J48 decision tree classifier with the main goal of enhancing the voice and video communications. This classifier calculates a new set of AIFSN values by taking into account the current network conditions, maintaining the interoperability with the legacy Distributed Coordination Function (DCF) applications. The results show that our proposal improves upon the voice+video performance results obtained by the AIFSN standard values by up to 20%. Furthermore, this scheme is fully compatible with the commercial network cards available on the market.