Search

As a promising use-case of the Internet of Things (IoT), wildlife tracking and monitoring applications greatly benefit the ecology-related research both commercially and scientifically. In literature, a forward-wait-deliver strategy has been researched to facilitate energy-efficient dissemination of delay-tolerant information, which penitentially contributes to long-term tracking and monitoring. However, this strategy is not directly applicable for wildlife tracking and monitoring applications, as the movement trajectory of animals cannot be precisely predicted for relay selection. To this end, further studies are required to utilize partially predictable mobility based on more generalized navigational information such as the movement direction. In this paper, the feasible exploitation of directional movement in path-unconstrained mobility is investigated for strategic forwarding. Our proposal is an advance to the state-of-the-art because the directional correlation of destination movement is considered to dynamically exploit the node mobility for the optimal selection of a stationary relay. Simulation results show that higher delivery utility can be achieved by the proposed fuzzy path model compared with a forwarding scheme without contact prediction or one based on linear trajectory model.

This paper proposes a position-based routing scheme called Speed Up-Greedy Perimeter Stateless Routing Protocol for Wireless Sensor Networks (SU-GPSR). The speed up mode works together with greedy mode to provide a solution for data transmission in WSN. Unlike other variants of greedy perimeter stateless routing protocols, SU-GPSR not only embraces both still nodes and mobile nodes in one unified scheme, but provide a complete energy consumption consideration when evaluating the potential next hop. The simulation results reveal that the SU-GPSR outperforms EGPSR and FA-GPSR from various aspects in both dense and loose network size.

This paper proposes a cross-layer wireless routing protocol called REAR (Radio Environment Adaptive Routing Protocol), which operates with Cognitive Radio (CR) Mobile Ad Hoc Networks (CRMANET). Unlike conventional on-demand routing protocols for ad hoc networks, the design of REAR allows nodes along the route to adjust the next hop according to real-time radio environment conditions during the transmission. This flexibility benefits the communication by reducing the risk of link breaks in the CRMANET and so improves the overall network performance and spectrum utilization. Simulation results show that, compared to DSR [2] and CRMBR [8], REAR provides higher robustness in a rapid changing ad hoc environment. Therefore the overall network throughput is improved with reduced end-to-end delay.

This paper proposes an adaptive packet aggregation scheme with QoS awareness in CR (Cognitive Radio) enabled Mobile Ad Hoc Networks (CRMANET). Instead of directly introducing the aggregation scheme from the 802.11n standard into CRMANET, the proposed adaptive scheme inherits the design advantages from the work given in [1] in regard to packet scheduling to further utilize each sub-channel before every transmission. Two types of service (Retransmission and Non-Retransmission) are considered in the scheduling scheme. Meanwhile three aspects (buffer structure, time measurement and back-off) are carefully designed in relation to the aggregation frame. Simulation results show that the aggregated scheduling scheme provides much higher throughput and less delay in a CRMANET environment compared to the performance of the 802.11n scheduling scheme.

Due to the miniaturization trend and functional demand in high-density microelectronic packaging, the thermomigration in flip chip solder joints owing to the joule heating becomes a serious reliability issue. In this study, a novel apparatus which can provide a sufficient temperature gradient cross the solder joint in the specimen was used to carry on the thermomigration experiment separated from electromigration behavior. A constant temperature gradient above 2000°C/cm was applied on the eutectic SnPb solder layer of specimen and the load duration were 20h, 40h, and 80h respectively. The result reveals that in eutectic SnPb solder layer of specimen, an obvious mass diffusion from hot side to cold side occurs and Pb is estimated as the primary diffusion element under a high enough temperature gradient. EDX result also proves that the percentage of Pb element in the cold side is higher than that in the hot side. With the temperature gradient load time goes up from 20h to 80h, the phenomenon of Pb migration to cold side turns to be much severer. Due to the redistribution of both Pb and Sn element, the morphology and thickness of IMC in both hot side and cold side changes obviously comparing that in as-reflowed solder layer. Meanwhile, the tensile test also shows that the mechanical strength of eutectic SnPb solder layer becomes weaker with the thermomigration load duration increases gradually, and this may be owing to the defects existing in the solder caused by mass diffusion.

Clustering a data set that exhibits arbitrary shapes is an important research area in unsupervised data labeling. The paper is concerned with using a shape-matched curve as the prototype of that cluster. Data points are then labeled by those curve-represented clusters without any preknowledge and priori assumptions of hidden structures in the data set. Since the shapes of these curves are often dendritic, we call this learning process dendritic curve clustering (DCC). DCC makes use of fuzzy c-means, and each shape-matched cluster curve is achieved through cluster growing. Experimental results demonstrate the DCC approach is feasible.