An Optimization Framework for Data Collection in Software Defined Vehicular Networks.

A Software Defined Vehicular Network (SDVN) is a new paradigm that enhances programmability and flexibility in Vehicular Adhoc Networks (VANETs). There exist different architectures for SDVNs based on the degree of control of the control plane. However, in vehicular communication literature, we find that there is no proper mechanism to collect data. Therefore, we propose a novel data collection methodology for the hybrid SDVN architecture by modeling it as an Integer Quadratic Programming (IQP) problem. The IQP model optimally selects broadcasting nodes and agent (unicasting) nodes from a given vehicular network instance with the objective of minimizing the number of agents, communication delay, communication cost, total payload, and total overhead. Due to the dynamic network topology, finding a new solution to the optimization is frequently required in order to avoid node isolation and redundant data transmission. Therefore, we propose a systematic way to collect data and make optimization decisions by inspecting the heterogeneous normalized network link entropy. The proposed optimization model for data collection for the hybrid SDVN architecture yields a 75.5% lower communication cost and 32.7% lower end-to-end latency in large vehicular networks compared to the data collection in the centralized SDVN architecture while collecting 99.9% of the data available in the vehicular network under optimized settings.

Automatic Assessment of Aphasic Speech Sensed by Audio Sensors for Classification into Aphasia Severity Levels to Recommend Speech Therapies.

Aphasia is a type of speech disorder that can cause speech defects in a person. Identifying the severity level of the aphasia patient is critical for the rehabilitation process. In this research, we identify ten aphasia severity levels motivated by specific speech therapies based on the presence or absence of identified characteristics in aphasic speech in order to give more specific treatment to the patient. In the aphasia severity level classification process, we experiment on different speech feature extraction techniques, lengths of input audio samples, and machine learning classifiers toward classification performance. Aphasic speech is required to be sensed by an audio sensor and then recorded and divided into audio frames and passed through an audio feature extractor before feeding into the machine learning classifier. According to the results, the mel frequency cepstral coefficient (MFCC) is the most suitable audio feature extraction method for the aphasic speech level classification process, as it outperformed the classification performance of all mel-spectrogram, chroma, and zero crossing rates by a large margin. Furthermore, the classification performance is higher when 20 s audio samples are used compared with 10 s chunks, even though the performance gap is narrow. Finally, the deep neural network approach resulted in the best classification performance, which was slightly better than both K-nearest neighbor (KNN) and random forest classifiers, and it was significantly better than decision tree algorithms. Therefore, the study shows that aphasia level classification can be completed with accuracy, precision, recall, and F1-score values of 0.99 using MFCC for 20 s audio samples using the deep neural network approach in order to recommend corresponding speech therapy for the identified level. A web application was developed for English-speaking aphasia patients to self-diagnose the severity level and engage in speech therapies.

Performance Analysis of Distributed Estimation for Data Fusion Using a Statistical Approach in Smart Grid Noisy Wireless Sensor Networks.

Internet of Things (IoT) can significantly enhance various aspects of today's electric power grid infrastructures for making reliable, efficient, and safe next-generation Smart Grids (SGs). However, harsh and complex power grid infrastructures and environments reduce the accuracy of the information propagating through IoT platforms. In particularly, information is corrupted due to the measurement errors, quantization errors, and transmission errors. This leads to major system failures and instabilities in power grids. Redundant information measurements and retransmissions are traditionally used to eliminate the errors in noisy communication networks. However, these techniques consume excessive resources such as energy and channel capacity and increase network latency. Therefore, we propose a novel statistical information fusion method not only for structural chain and tree-based sensor networks, but also for unstructured bidirectional graph noisy wireless sensor networks in SG environments. We evaluate the accuracy, energy savings, fusion complexity, and latency of the proposed method by comparing the said parameters with several distributed estimation algorithms using extensive simulations proposing it for several SG applications. Results prove that the overall performance of the proposed method outperforms other fusion techniques for all considered networks. Under Smart Grid communication environments, the proposed method guarantees for best performance in all fusion accuracy, complexity and energy consumption. Analytical upper bounds for the variance of the final aggregated value at the sink node for structured networks are also derived by considering all major errors.

A study in University of Ruhuna for investigating prevalence, risk factors and remedies for psychiatric illnesses among students.

There is no comprehensive study on the mental health of Sri Lankan undergraduate in higher education, as most existing studies have been done for medical students only. It is unknown how academic and environmental factors contribute for the prevalence of psychiatric illnesses. Further, there is no sufficient information on the student/university based remedies to reduce the psychological distress of students. This research is carried out to find the overall psychological distress, well-being, prevalence percentages of psychiatric illnesses, associated risk factors, and student/university remedies to overcome them. We use standard questionnaires to screen for psychiatric illnesses, and we analyze the responses for our own questionnaire using Binary logistic regression analysis to identify demographic factors, academic factors, and environmental factors causing each mental disorder. We use Pearson correlation coefficient to identify correlation between prevalence of each psychiatric illnesses. All 13 psychiatric illnesses were found with a moderate correlation among diseases having a mean prevalence percentage of 28 and a standard deviation of 14.36, despite the prevalence of well-being factors among students and only 8% are clinically diagnosed. 89% of the students were suffering from at least one psychiatric illness and 68% were found to be psychologically distressed. Sets of overall and individual demographic, academic, and environmental risk factors contributing for the prevalence of a psychiatric illness in general and in particular were identified respectively after a binary logistic regression analysis. 61% of the students don't receive psychiatric help from the university and are using their own remedies. The universities must consider the environmental and academic risk factors associated with psychiatric illnesses and design curriculum, expand resources, and provide counseling services to reduce the impact of risk factors.