Sungryul Kim

Planar cone-beam computed tomography with a flat-panel detector

For a dedicated x-ray inspection of printed-circuit boards (PCBs), a bench-top planar cone-beam computed tomography (pCT) system with a flat-panel detector has been built in the laboratory. The system adopts the tomosynthesis technique that can produce cross-sectional images parallel to the axis of rotation for a limited angular range. For the optimal operation of the system and further improvement in the next design, we have evaluated imaging performances, such as modulation-transfer function, noise-power spectrum, and noise-equivalent number of quanta. The performances are comparatively evaluated with the coventional cone-beam CT (CBCT) acquisition for various scanning angular ranges, applied tube voltages, and geometrical magnification factors. The pCT scan shows a poorer noise performance than the conventional CBCT scan because of less number of projection views used for reconstruction. However, the pCT shows a better spatial-resolution performance than the CBCT. Because the image noise can be compensated by an elevated exposure level during scanning, the pCT can be a useful modality for the PCB inspection that requires higher spatial-resolution performance.

SLSMP: Time Synchronization and Localization Using Seawater Movement Pattern in Underwater Wireless Networks

Time synchronization and localization in underwater environment are challenging due to high propagation delay, time measurement error, and node mobility. Although synchronization and localization depend on each other and have the similar process, they have been usually handled separately. In this paper, we suggest time synchronization and localization based on the semiperiodic property of seawater movement, called SLSMP. Firstly, we analyze error factors in time synchronization and localization and then propose a method to handle those errors. For more accurate synchronization, SLSMP controls the transmission instant by exploiting the pattern of seawater movement and node deployment. Then SLSMP progressively decreases the localization errors by applying the Kalman filter or averaging filter. Finally, INS (inertial navigation system) is adopted to relieve localization error caused by node mobility and error propagation problem. The simulation results show that SLSMP reduces time synchronization error by 2.5 ms and 0.56 ms compared with TSHL and MU-Sync, respectively. Also localization error is lessened by 44.73% compared with the single multilateration.

In-vitro and in-vivo antibacterial activity evaluation of a polyurethane matrix.

Various in‐vitro and in‐vivo methods for evaluation of the duration of antibacterial activity were compared using a controlled‐release polyurethane matrix developed for the prevention of surface bacterial adhesion and growth. Cefadroxil was incorporated into this polyurethane matrix by a solvent casting method before the matrix was coated with polyurethane in tetrahydrofuran solution. The release of cefadroxil from the matrix into distilled water at 37 °C was measured by HPLC. The morphological change of matrices before and after release studies was investigated by scanning electron microscopy (SEM). The duration of antimicrobial activity of the matrix against Escherichia coli and Staphylococcus aureus was evaluated by measuring the diameters of the inhibition zone and the optical density of the broth. The matrices were also implanted subcutaneously in rats and the duration of the antibacterial activity was determined by measuring the inhibition zone. The results showed that duration of antibacterial activity of the polyurethane matrix was successfully determined in‐vitro by these methods, and the results differed from the conventional in‐vitro release study. It was also possible to determine the duration of action of the matrix in‐vivo by implanting the matrix in rats, and then measuring the antibacterial activity of the matrix at predetermined time intervals. While a good correlation was observed between the in‐vitro and in‐vivo methods used in this study to evaluate the duration of the antibacterial activity of the polymeric matrix, the conventional in‐vitro release study did not coincide with these results.

Development of oil-mist-detector with capacitance sensor

Low speed and high load operating conditions of large diesel engines can cause structural damage which results in frictional heat. When the frictional heat increases in the crank case of a diesel engine it causes the lubricants to evaporate and thus the oil mist density increases. Flames from a damaged structure or heat from the engine starts ignites easily under a high density condition of the oil mist. The resultant explosion due to a heavy oil mist causes not only engine fracture but can also cause casualties. The oil-mist-detector (OMD) is used to prevent engine explosions by checking the density of the oil mist in the diesel engine. Current OMDs use the light scatter method and an absorption method. The first detects the oil mist by checking the quantity of light scattered due to oil mist and the absorption method detects the density by sensing the quantity of light penetrated through oil mist from a transmitter to a receiver. However, these two methods are not conducive to long-term use because the transmitter and receiver become polluted by cohesion or electrification of the oil mist. In this study, the OMD equipped with a capacitance sensor is proposed to detect the oil mist density and the performance is investigated. The capacitance sensor is proposed and designed for effective detection of oil mist and the experiments are conducted under various oil mist size and density conditions. The results show the capacitance sensor detects the oil mist density effectively and the OMD with a capacitance sensor shows the linear response according to density changes even in polluted conditions.

Dynamic transmission power control based on exact sea surface movement modeling in underwater acoustic sensor networks

Prediction of sea surface movement can be an important tool for the estimation of time-variant acoustic channel because signal attenuation caused by reflection occupies a large proportion in path loss. Although a number of researches have proposed resource allocation schemes based on the channel modeling, they did not consider reflection loss and time-variant characteristic. This paper suggests a transmission power control based on the prediction of time-variant channel by using the RMS (Root Mean Square) wave-height for low power consumption and stable throughput. The proposed scheme adopts transfer function including reflection coefficient overlooked in other papers using the Kirchhoff approximation. In addition, it defines the transmission power needed to guarantee a pre-specified SNR (Signal-to-Noise Ratio) threshold using the transfer function. The BELLHOP and WAFO simulators were utilized to build simulation environment similar to actual ocean. The simulation results show that the proposed method is practical by considering the reflection impact on the power control and reduces energy consumption by 32.79% compared with the existing methods which do not use the adaptive power control based on channel condition.

Container security device chain network for safe railway transportation

Container Security Device (CSD) has been adopted for the security of containers during the sea transportation. On the other hand, the railway is also one of the most important facilities for freight logistics. This paper suggests the use of the CSD tags to enhance the security of railway freight transportation. The proposed protocol, CSD Chain Network Protocol (CCNP), is designed for a chain-form network of CSDs each of which is attached to a container. In CCNP, unauthorized access to freight is reliably reported to the reader deployed at the locomotive. CCNP also cares about the fairness in energy consumption. In a chain network, usually, a tag closer to the reader consumes more energy since it ought to forward packets coming from farther tags as well. However, in CCNP, the tag energy consumption for packet forwarding is always the same regardless of the position in a chain by adopting an efficient data fusion scheme. In our simulation using QualNet, CCNP reduces the number of disconnections between tags by 59.5% compared to general chain networks. The energy consumption also decreases by 91.3% of non-aggregation methods.

A bit collision detection based Query Tree protocol for sensor tags in logistics management

The advantages of wireless sensors and RFID technology have been widely acknowledged. Based on the two advanced technologies, we improved our current RFID system by replacing passive tags with active sensor tags. Toward better performance on new system demands, in this paper, we proposes an improved QT protocol called SQT (Sensor tag based Query Tree protocol) which combines the basic principle of QT protocol and a new bit collision detection mechanism for sensor tags identification by a mobile reader. Analysis shows that the SQT protocol reduces identification collisions, which accelerates the identification process. Simulation results substantiate the significant performance improvement by the SQT protocol for sensor tags identification.

MIMO-HFM: Acoustic system design for reliability and high data rate in underwater sensor networks

Accomplish of reliability and high data rate over the time-varying multipath acoustic channel is very challenging. In this paper, we propose Multiple-Input and Multiple-Output (MIMO) system using Hyperbolic Frequency Modulation (HFM) waveform. Our system combines the advantages of both system, the spatial multiplexing of MIMO and Doppler-tolerance of HFM. To increase spectral efficiency, we use M-ray HFM, and overlapped sub-channel by leveraging the high temporal resolution characteristic of HFM. Also, massive operation overhead in the receiver side is drastically reduced using Fast Fourier Transform (FFT). Based on our acoustic simulator, the performance of MIMO-HFM is evaluated to show its multipathfading resolution and Doppler-invariant. The simulation results show that the MIMO-HFM is robust in error-prone acoustic channel while increasing data rate as compared with single antenna system.

Container Security Device Chain Network for Safe Railway Transportation

The railway is one of the most important facilities for freight logistics. This paper suggests the use of the container security devices (CSDs), which has been adopted for the marine transportation so far, to enhance the security of railway freight transportation. The proposed protocol, CSD Chain Network Protocol (CCNP), was designed for a chain-form network of CSDs, each of which is attached to a container. In CCNP, unauthorized access to freight is reliably reported to the reader deployed at the locomotive. CCNP also cares about the fairness in energy consumption. In a usual chain network, a tag closer to the reader consumes more energy since it ought to forward packets coming from rear tags as well. However, in CCNP, the tag energy consumption for packet forwarding is always the same regardless of the position in a chain owing to an efficient data fusion scheme. In our simulation using QualNet, CCNP reduces the number of disconnections between tags by 59.5% as compared to general chain networks. The energy consumption also decreases by 91.3% of other networks not using an aggregation method.

High-Precision and Practical Localization Using Seawater Movement Pattern and Filters in Underwater Wireless Networks

Localization in underwater is challenging due to harsh communication channel. Although a number of researches have proposed localization for underwater environment, they did not fully consider practical issues affecting the localization accuracy. In particular, our research focus on the unreliable reference location caused by mobility of underwater sensor nodes and unpredicted time measurement error in TDoA(Time Difference of Arrival) calculation. Unfortunately, none of related papers dealing with those factors, they significantly reduce the localization performance though. For the more accurate and practical localization, we suggest a protocol called Localization with Seawater Movement Pattern and Filter (LSMF), for underwater sensor networks. Especially, LSMF utilizes feature of seawater movement and node deployment. Also, LSMF more accurately estimates the location with iterative data processing using Kalman filter or Averaging filter. Furthermore, Adopting of INS (Inertial Navigation System) relieves localization error caused by node mobility. As a result, the localization accuracy will be improved and error propagation problem in the multi-hop networks is naturally weaken. Finally, simulation results show that LSMF reduced localization error by 44.73% compared with previous multilateration in the same simulation environment.