• Journal of the Microelectronics and Packaging Society
• /
• v.20 no.4
• /
• pp.47-52
• /
• 2013

Gold wire has long been used as a proven method of connecting a silicon die to a substrate in wide variety of package types, delivering high yield and productivity. However, with the high price of gold, the semiconductor packaging industry has been implementing an alternate wire material. These materials may include silver (Ag) or copper (Cu) alloys as an alternative to save material cost and maintain electrical performance. This paper will analyze and compare the electrical characteristics of several wire types. For the study, typical 0.6 mil, 0.8 mil and 1.0 mil diameter wires were selected from various alloy types (2N gold, Palladium (Pd) coated/doped copper, 88% and 96% silver) as well as respective pure metallic wires for comparison. Each wire model was validated by comparing it to electromagnetic simulation results and measurement data. Measurements from the implemented test boards were done using a vector network analyzer (VNA) and probe station setup. The test board layout consisted of three parts: 1. Analysis of the diameter, length and material characteristic of each wire; 2. Comparison between a microstrip line and the wire to microstrip line transition; and 3. Analysis of the wire's cross-talk. These areas will be discussed in detail along with all the extracted results from each type the wire.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.7B
• /
• pp.528-537
• /
• 2008

This paper proposes a dynamic job distribution algorithm in a hybrid NoC structure which can improve system network performance by reducing deadlock and packet drop rate for various multimedia applications. The proposed job distribution algorithm schedules every job to the sub-cluster where packet drop rate can be minimized for each multimedia application program. The proposed Job distribution algorithm and network topology targets multimedia applications frequently used in modern embedded systems, such as MPEG4 and MP3 decoder, GPS positioning system, and OFDM demodulator. Experimental results show that packet drop rate was reduced by about 13.0%, and chip area was increased by about 2.7% compared to the APSRA algorithm. When compared to the XY algorithm popularly used for benchmarking, the packet drop rate was reduced by about 23.9%, while chip area was increased by about 3.0%.

• Proceedings of the KIEE Conference
• /
• 2004.07d
• /
• pp.2386-2388
• /
• 2004

Recently, rapid process of network technology has an effect on the field of control engineering. And it gives effect to network control system(NCS) research that grafts existing control system and network has been studied. According to one-chip development of TCP/IP that is the most useful in LAN server for hardware, it can usefully and conveniently apply to network control system before. This paper construct network control system that can work close-loop control takes advantage of Ethernet that is the most for general data transmission network. Here, protocol uses for data transmission makes use for TCP/IP. This work the basic data transmitter-receive experiment to take advantage of network node was produced. And then, DC-motor can be control plant. Finally, I will prove to build speed control system of multiple DC-motor through Ethernet.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.44 no.8
• /
• pp.74-79
• /
• 2007

In general, the high cost, long time, and complex steps are required in the design and implementation of MPSoC(Multi-Processor System on a Chip), therefore a platform is used to test the functionality and performance of IPs(Intellectual Properties). In this paper, we study a platform architecture to verify IPs based on Interconnect Network among processors, and show that the MPSoC platform gives better performance than a single processor for an application program.

• Journal of Advanced Navigation Technology
• /
• v.19 no.5
• /
• pp.423-432
• /
• 2015

This paper shows that design, manufacture and the performance of FC-NIC (fibre channel network interface card) for network unit configuration which is based on one of the 5 main configuration items of the common functional module for IMA (integrated modular Avionics) architecture. Especially, FC-NIC uses zynq SoC (system on chip) for host load reductions. The host merely transmit FC destination address, source memory location and size information to the FC-NIC. After then the FC-NIC read the host memory via DMA (direct memory access). FC upper layer protocol and sequence process at local processor and programmable logic of FC-NIC zynq SoC. It enables to free from host load for external communication. The performance of FC-NIC shows average 5.47 us low end-to-end latency at 2.125 Gbps line speed. It represent that FC-NIC is one of good candidate network for IMA.

• Journal of the Microelectronics and Packaging Society
• /
• v.14 no.2 s.43
• /
• pp.41-47
• /
• 2007

Today lots of investigations on Embedded Passive Technology using materials and chip components have been carried out. We fabricated diplexers with 1005 sized-passives, which were made by burying chips in PCB substrate and surface mounting chip on PCB. 6 passive chips (inductors and capacitors) were used for the frequency divisions of $880\;MHz{\sim}960\;MHz(GSM)$ and $1.71\;GHz{\sim}1.88\;GHz(DCS)$. Two types of diplxer were characterized with Network analyzer. The chip buried diplexer showed extra 5db loss and a little deviation of 0.6GHz at aimed frequency areas, whereas the chip mounted diplexer showed man. 0.86dB loss within GSM field and max. 0.68dB within DCS field respectively. But few degradations were observed after $260^{\circ}C$ for 80min baking and $280^{\circ}C$ for 10sec solder floating.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09b
• /
• pp.122-125
• /
• 2003

By advent of NNC(Neural Network Chip), it is possible that process in parallel and discern the importance of signal with learning oneself by experience in external signal. So, the design of general purpose operation unit using VHDL(VHSIC Hardware Description Language) on the existing FPGA(Field Programmable Gate Array) can replaced EN(Expert Network) and learning algorithm. Also, neural network operation unit is possible various operation using learning of NN(Neural Network). This paper present general purpose operation unit using hierarchical structure of EN EN of presented structure learn from logical gate which constitute a operation unit, it relocated several layer The overall structure is hierarchical using a module, it has generality more than FPGA operation unit.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.6 no.3
• /
• pp.410-415
• /
• 2002

This paper describes the implementation of a SoC(System-on-a-Chip) and an ISDN communication terminal by the SoC in ISDN network. The SoC has been developed with the functions of 32-bit ARM7TDMI RISC core processor, network connection with S/T interface, TDM--bus interface and voice codec, user interface. And we also review the developed software structure and the ISDN service protocol procedures which are working on the SoC. And finally this paper describers a structure of an ISDN terminal equipment using the implemented SoC and terminal software.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.2
• /
• pp.93-100
• /
• 2008

In this paper, we address a novel simulated annealing(SA)-based test scheduling method for testing network-on-chip (NoC)-based systems-on-chip(SoCs), on the assumption that the test platform proposed in [1] is installed. The proposed method efficiently mixed the rectangle packing method with SA and improved the scheduling results by locally changing the test access mechanism(TAM) widths for cores and the testing orders. Experimental results using ITC'02 benchmark circuits show that the proposed algorithm can efficiently reduce the overall test time.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.8
• /
• pp.671-677
• /
• 2016

A Controller Area Network (CAN) is a serial communication protocol that is highly reliable and efficient in many aspects, such as wiring cost and space, system flexibility, and network maintenance. Therefore, it is chosen for the communication protocol between a single chip controller based on Field Programmable Gate Array (FPGA) and peripheral devices. In this paper, the design and implementation of CAN IP, which is written in VHSIC Hardware Description Language (VHDL), is presented. The implemented CAN IP is based on the CAN 2.0A specification. The CAN IP consists of three processes: clock generator, bit timing, and bit streaming. The clock generator process generates a time quantum clock. The bit timing process does synchronization, receives bits from the Rx port, and transmits bits to the Tx port. The bit streaming process generates a bit stream, which is made from a message received from a micro controller subsystem, receives a bit stream from the bit timing process, and handles errors depending on the state of the CAN node and CAN message fields. The implemented CAN IP is synthesized and downloaded into SmartFusion FPGA. Simulations using ModelSim and chip test results show that the implemented CAN IP conforms to the CAN 2.0A specification.