Abstract
In movie/TV collaborative recommendation approaches, ratings users gave to already visited content are often used as the only input to build profiles. However, users might have rated equally the same movie but due to different reasons: either because of its genre, the crew or the director. In such cases, this rating is insufficient to represent in detail users' preferences and it is wrong to conclude that they share similar tastes. The work presented in this paper tries to solve this ambiguity by exploiting hidden semantics in metadata elements. The influence of each of the standard description elements (actors, directors and genre) in representing user's preferences is analyzed. Simulations were conducted using Movielens and Netflix datasets and different evaluation metrics were considered. The results demonstrate that the implemented approach yields significant advantages both in terms of improving performance, as well as in dealing with common limitations of standard collaborative algorithm.

Abstract
Efficient test and diagnosis methods are required to ensure high levels of dependability of the electronic systems deployed to the market. These methods involve a trade-off in terms of accessibility to test nodes, test stimuli complexity, area overhead, and data processing that, altogether determine the impact that the involved operations have in the final cost, performance, and reliability presented by these systems. The work presented here describes preliminary results obtained with the application of correntropy as a means to efficiently analyse test responses in the fault detection decision process. © 2017 IEEE.

Abstract
The increasing trend in wireless Internet access has been boosted by IEEE 802.11. However, the application scenarios are still limited by its short radio range. Stub Wireless Multi-hop Networks (WMNs) are a robust, flexible, and cost-effective solution to the problem. Yet, typically, they are formed by single radio mesh nodes and suffer from hidden node, unfairness, and scalability problems. We propose a simple multi-radio, multi-channel WMN solution, named Wi-Fi network Infrastructure eXtension-Dual-Radio (WiFIX-DR), to overcome these problems. WiFIX-DR reuses IEEE 802.11 built-in mechanisms and beacons to form a Stub WMN as a set of self-configurable interconnected Basic Service Sets (BSSs). Experimental results show the improved scalability enabled by the proposed solution when compared to single-radio WMNs. © 2017 IEEE.

Abstract
This paper presents a novel low-power analog circuit, with n-type IGZO TFTs that can function as an adder operator or be designed to operate as a driver. Experiments were set to show summation of up to four signals. However, the design can easily be expanded to add higher number of signals, by appending a single TFT at the input per each additional signal. The circuit is simple, uses a single power supply irrespective to the number of input voltage signals, and shows good accuracy over a reasonable range of input values. By choosing proper TFT dimensions, the topology can replace the typical output drivers of TFT amplifiers, namely the common-drain with current source biasing, or the common-source with diode connected load. The circuit was fabricated with a temperature that does not exceeds 200 degrees C. Its performance is characterized from measurements at room temperature and normal ambient, with a power supply voltage of 12 V and a load of approximate to 4 pF. The proposed circuit has shown a linearity error less than 3.2% (up to an input signal peak-to-peak value of 2 V), a power consumption of 78 mu W and a bandwidth of approximate to 115 kHz, under worst case condition (when it is adding four signals with the same frequency). It has shown superior performance in terms of linearity when compared to the typical drivers considered in this study. In addition, it has shown almost the same behavior when measurements were repeated after one year. Therefore, the proposed circuit is a robust viable alternative to conventional approaches, being more compact, and contributes to increase the functionality of large-area flexible electronics.

Abstract
This paper presents a low-power binary phase shift keying (BPSK) pulse generator for ultra-wide-band transmitters. The circuit has been designed based on LC-tank resonators using 0.13 um CMOS technology. Simulation shows -10dB bandwidth of around 3 GHz and power consumption of 2 mW at 100 MHz PRF. Peak-peak amplitude voltage for both symbols '1' and '0' are approximately as large as 1.2V supply voltage and can radiate enough energy to satisfy the FCC mask only by one pulse. Thus, the energy consumption is 20 pJ/pulse/bit. Pulse duration is 1.5 ns and the transmitter can reach data rates of 660 Mbps

Abstract
Deep transfer learning emerged as a new paradigm in machine learning in which a deep model is trained on a source task and the knowledge acquired is then totally or partially transferred to help in solving a target task. In this paper, we apply the source-target-source methodology, both in its original form and an extended multi-source version, to the problem of cross-sensor biometric recognition. We tested the proposed methodology on the publicly available CSIP image database, achieving state-of-the-art results in a wide variety of cross-sensor scenarios.