AI Based Digital Twin Model for Cattle Caring.

In this paper, we develop innovative digital twins of cattle status that are powered by artificial intelligence (AI). The work is built on a farm IoT system that remotely monitors and tracks the state of cattle. A digital twin model of cattle based on Deep Learning (DL) is generated using the sensor data acquired from the farm IoT system. The physiological cycle of cattle can be monitored in real time, and the state of the next physiological cycle of cattle can be anticipated using this model. The basis of this work is the vast amount of data that is required to validate the legitimacy of the digital twins model. In terms of behavioural state, this digital twin model has high accuracy, and the loss error of training reach about 0.580 and the loss error of predicting the next behaviour state of cattle is about 5.197 after optimization. The digital twins model developed in this work can be used to forecast the cattle's future time budget.

Enabling Real-Time Quality-of-Service and Fine-Grained Aggregation for Wireless TSN.

Wireless Time-Sensitive Networking (WTSN) has emerged as a promising technology for Industrial Internet of Things (IIoT) applications. To meet the latency requirements of WTSN, wireless local area network (WLAN) such as IEEE 802.11 protocol with the time division multiple access (TDMA) mechanism is shown to be a practical solution. In this paper, we propose the RT-WiFiQA protocol with two novel schemes to improve the latency and reliability performance: real-time quality of service (RT-QoS) and fine-grained aggregation (FGA) for TDMA-based 802.11 systems. The RT-QoS is designed to guarantee the quality-of-service requirements of different traffic and to support the FGA mechanism. The FGA mechanism aggregates frames for different stations to reduce the physical layer transmission overhead. The trade-off between the reliability and FGA packet size is analyzed with numerical results. Specifically, we derive a critical threshold such that the FGA can achieve higher reliability when the aggregated packet size is smaller than the critical threshold. Otherwise, the non-aggregation scheme outperforms the FGA scheme. Extensive experiments are conducted on the commercial off-the-shelf 802.11 interface. The experiment results show that compared with the existing TDMA-based 802.11 system, the developed RT-WiFiQA protocol can achieve deterministic bounded real-time latency and greatly improves the reliability performance.

Delay-Sensitive NOMA-HARQ for Short Packet Communications.

This paper investigates the two-user uplink non-orthogonal multiple access (NOMA) paired with the hybrid automatic repeat request (HARQ) in the finite blocklength regime, where the target latency of each user is the priority. To limit the packet delivery delay and avoid packet queuing of the users, we propose a novel NOMA-HARQ approach where the retransmission of each packet is served non-orthogonally with the new packet in the same time slot. We use a Markov model (MM) to analyze the dynamics of the uplink NOMA-HARQ with one retransmission and characterize the packet error rate (PER), throughput, and latency performance of each user. We also present numerical optimizations to find the optimal power ratios of each user. Numerical results show that the proposed scheme significantly outperforms the standard NOMA-HARQ in terms of packet delivery delay at the target PER.

Significant Low-Dimensional Spectral-Temporal Features for Seizure Detection.

Absence seizure as a generalized onset seizure, simultaneously spreading seizure to both sides of the brain, involves around ten-second sudden lapses of consciousness. It common occurs in children than adults, which affects living quality even threats lives. Absence seizure can be confused with inattentive attention-deficit hyperactivity disorder since both have similar symptoms, such as inattention and daze. Therefore, it is necessary to detect absence seizure onset. However, seizure onset detection in electroencephalography (EEG) signals is a challenging task due to the non-stereotyped seizure activities as well as their stochastic and non-stationary characteristics in nature. Joint spectral-temporal features are believed to contain sufficient and powerful feature information for absence seizure detection. However, the resulting high-dimensional features involve redundant information and require heavy computational load. Here, we discover significant low-dimensional spectral-temporal features in terms of mean-standard deviation of wavelet transform coefficient (MS-WTC), based on which a novel absence seizure detection framework is developed. The EEG signals are transformed into the spectral-temporal domain, with their low-dimensional features fed into a convolutional neural network. Superior detection performance is achieved on the widely-used benchmark dataset as well as a clinical dataset from the Chinese 301 Hospital. For the former, seven classification tasks were evaluated with the accuracy from 99.8% to 100.0%, while for the latter, the method achieved a mean accuracy of 94.7%, overwhelming other methods with low-dimensional temporal and spectral features. Experimental results on two seizure datasets demonstrate reliability, efficiency and stability of our proposed MS-WTC method, validating the significance of the extracted low-dimensional spectral-temporal features.