Abstract:
The apparatus a multiphase clock pulse generator (201) from which a pulse to synchronise incoming transmission signals is selected. Re-synchronisation is performed by a clock pulse selector (202). The selector finds clock pulses which are near to the edges of the transmission pulse sequences and a re-synchronisation synthesiser (203) synchronises the incoming transmissions using the selected pulses. The re-synchronised words are output by an output buffer.
Abstract:
A bit synchronizer for NRZ data wherein a loop gain of a phase locked loop in the bit synchronizer is not varied sensitively to bit pattern and rate of the NRZ data and a voltage control led oscillator in the bit synchronizer oscillates synchronously with a multiple of a frequency of an external reference clock pulse even in the absence of NRZ data transitions or over a wide range of variation of a clock frequency of the voltage controlled oscillator, so that the NRZ data and clock can be recovered stably, According to the invention, the bit synchronizer comprises a phase comparator, a first gain controller, a frequency comparator, a second gain controller, a N-frequency divider, a low pass filter and a voltage controlled oscillator.
Abstract:
A bit synchronizer for NRZ data wherein a loop gain of a phase locked loop in the bit synchronizer is not varied sensitively to bit pattern and rate of the NRZ data and a voltage controlled oscillator in the bit synchronizer oscillates synchronously with a multiple of a frequency of an external reference clock pulse oven in the absence of NRZ data transitions or over a wide range of variation of a clock frequency of the voltage controlled oscillator, so that the NRZ data and clock can be recovered stably. According to the invention, the bit synchronizer comprises a phase comparator (21), a first gain controller (22), a frequency comparator (23), a second gain controller (24), a N-frequency divider (25), a low pass filter (26) and a voltage controlled oscillator (27). The first and second gain controllers limit the gains of the place and frequency comparators, respectively, to predetermined values.
Abstract:
PURPOSE: A distributed processing method and apparatus for multi-layer data using priority processing of layer 7 payload are provided to efficiently process a higher layer by properly distributing and arranging a higher layer throughput and a lower layer throughput. CONSTITUTION: A distributed processing apparatus(100) for multi-layer data integrally processes data having a multi-layer structure. The distributed processing apparatus prioritizes a higher layer prior to a lower layer in order to efficiently process data flow consisting of multi-layers. The distributed processing apparatus for multi-layer includes a higher layer processing unit(110) and a lower layer processing unit(120). The higher layer processing unit processes upper layer data flow generated by using higher layer information. The lower layer processing unit processes data flow received by using lower layer information into lower layer data flow.
Abstract:
PURPOSE: A packet inspection apparatus and method thereof are provided to effectively execute parallel processing by minimizing numbers of a bottleneck phenomenon in the communication of a pattern processor. CONSTITUTION: A pattern providing processor(110) manages plural patterns through connection lists including priority. The pattern providing processor provides plural patterns. Plural pattern inspection processors(120~140) inspects a correlation between inputted patterns and plural patterns. The pattern providing processor divides the patterns into a pattern groups. The pattern providing processor successively allocates the divided pattern group to the pattern providing processors according to the connection lists.
Abstract:
PURPOSE: A packet scheduling device and a method thereof are provided to connect tuple state information with a processor identifier without a hash function, thereby increasing speed of scheduling packet data in a processor. CONSTITUTION: A processor allocating unit(210) allocates a packet to one of processors according to a comparison result of an information comparing unit. An information statistical unit(240) stores the number of packets which are processed per at least one processor. The information statistical unit selects one processor which minimally processes the packet. The information statistical unit provides a result to the processor allocating unit.