Robust indoor speaker recognition in a network of audio and video sensors

Situational awareness is achieved naturally by the human senses of sight and hearing in combination. Automatic scene understanding aims at replicating this human ability using microphones and cameras in cooperation. In this paper, audio and video signals are fused and integrated at different levels of semantic abstractions. We detect and track a speaker who is relatively unconstrained, i.e., free to move indoors within an area larger than the comparable reported work, which is usually limited to round table meetings. The system is relatively simple: consisting of just 4 microphone pairs and a single camera. Results show that the overall multimodal tracker is more reliable than single modality systems, tolerating large occlusions and cross-talk. System evaluation is performed on both single and multi-modality tracking. The performance improvement given by the audio–video integration and fusion is quantified in terms of tracking precision and accuracy as well as speaker diarisation error rate and precision–recall (recognition). Improvements vs. the closest works are evaluated: 56% sound source localisation computational cost over an audio only system, 8% speaker diarisation error rate over an audio only speaker recognition unit and 36% on the precision–recall metric over an audio–video dominant speaker recognition method.

Multifunctional anthraquinone-based sensors: UV, O2and time

An anthraquinone dye, Remazol brilliant blue R, RBBR, is used to create an indicator which can function as: (i) a UV dosimeter, (ii) an O2 indicator and (iii) a ‘Consume within’ indicator, CWI, for fresh, refrigerated foods. The dye is encapsulated in an ink containing a polymer, glycerol and a UV-activated semiconductor photocatalyst, titanium dioxide. When cast as a film, the dye is readily reduced by the TiO2 photocatalyst nanoparticles, thereby changing the colour of the film from blue to yellow, via a transitional green colour. The RBBR indicator is appropriately formulated, and covered with a film of Sellotape, which acts as an O2 barrier, so as to act as a sunburn warning indicator for people with skin type II. In the absence of the layer of Sellotape the RBBR indicator is used as an, albeit slow, sensor for measuring ambient levels of O2. Finally, by keeping the Sellotape layer, a UV-activated, yellow-coloured, RBBR indicator film is found to take ca. 42 h at 5 °C in ambient air to attain a green colour, and, on this basis, it is demonstrated as a possible CWI for refrigerated fresh foods.

A secure authentication and key management scheme for Wireless Sensors Network

In recent years, the adaptation of Wireless Sensor Networks (WSNs) to application areas requiring mobility increased the security threats against confidentiality, integrity and privacy of the information as well as against their connectivity. Since, key management plays an important role in securing both information and connectivity, a proper authentication and key management scheme is required in mobility enabled applications where the authentication of a node with the network is a critical issue. In this paper, we present an authentication and key management scheme supporting node mobility in a heterogeneous WSN that consists of several low capabilities sensor nodes and few high capabilities sensor nodes. We analyze our proposed solution by using MATLAB (analytically) and by simulation (OMNET++ simulator) to show that it has less memory requirement and has good network connectivity and resilience against attacks compared to some existing schemes. We also propose two levels of secure authentication methods for the mobile sensor nodes for secure authentication and key establishment.

Bridge weigh-in-motion system and Structural Health Monitoring using fiber optic sensors

There have been over 3000 bridge weigh-in-motion (B-WIM) installations in 25 countries worldwide, this has led vast improvements in post processing of B-WIM systems since its introduction in the 1970’s. This paper introduces a new low-power B-WIM system using fibre optic sensors (FOS). The system consisted of a series of FOS which were attached to the soffit of an existing integral bridge with a single span of 19m. The site selection criteria and full installation process has been detailed in the paper. A method of calibration was adopted using live traffic at the bridge site and based on this calibration the accuracy of the system was determined.

Experimental Study of Interfacial Stress Distribution of Bonded-in BFRP Rod Glulam Joints Using Fibre Optic Sensors (FOS)

In this study, fibre optic sensors (FOS) were used to investigate the interfacial stress-strain behaviour of bonded-in basalt fibre reinforced polymer (BFRP) rods loaded into glulam members. Pull-out tests were conducted to examine the effect of bonded length and load-to-grain direction on the distribution of stress at the BFRP rod/adhesive zone. It was observed that the stress concentration at the loaded end of the BFRP rod of the samples was significantly the highest while the unloaded end showed the lowest. Increasing the bonded length at the same loading configuration resulted in a decrease in stress concentration at the loaded end. The stress concentration at the loaded end of the perpendicular to the grain samples was relatively higher than that of the corresponding parallel to the grain samples.