摘要:
In the present invention, two improved variants of the reliability-based iterative majority-logic decoding algorithm for regular low-density parity-check (LDPC) codes are presented. The new algorithms are obtained by introducing a different reliability measure for each check-sum of the parity-check matrix, and taking it into account in the computation of the extrinsic information that is used to update the reliability measure of each received bit in each iteration. In contrast to the first algorithm, the second algorithm includes check reliability that changes at each iteration. For the tested random and structured LDPC codes, both algorithms, while requiring very little additional computational complexities, achieve a considerable error performance gain over the standard one. More importantly, for short and medium block length LDPC codes of relatively large column weight, both algorithms outperform or perform just as well as the iterative decoding based on belief propagation (IDBP) with less decoding complexity.
摘要:
In the present invention, two improved variants of the reliability-based iterative majority-logic decoding algorithm for regular low-density parity-check (LDPC) codes are presented. The new algorithms are obtained by introducing a different reliability measure for each check-sum of the parity-check matrix, and taking it into account in the computation of the extrinsic information that is used to update the reliability measure of each received bit in each iteration. In contrast to the first algorithm, the second algorithm includes check reliability that changes at each iteration. For the tested random and structured LDPC codes, both algorithms, while requiring very little additional computational complexities, achieve a considerable error performance gain over the standard one. More importantly, for short and medium block length LDPC codes of relatively large column weight, both algorithms outperform or perform just as well as the iterative decoding based on belief propagation (IDBP) with less decoding complexity.
摘要:
A porous discriminating layer is formed on a ceramic support having at least one porous wall by (a) establishing a flow of a gas stream containing agglomerates of particles and (b) calcining said deposited layer to form the discriminating layer. At least a portion of the particles are of a sinter-resistant material or a sinter-resistant material precursor. The particles have a size from 0.01 to 5 microns and the agglomerates have a size of from 10 to 200 microns. This method is an inexpensive and effective route to forming a discriminating layer onto the porous wall.
摘要:
A fuel injector is provided and includes a member defining a flowpath through which a first fluid flows, the flowpath having a cross-section with transverse elongate and short axes, a head defining a plenum storing a supply of a second fluid and a system fluidly coupled to the flowpath and the plenum to inject the second fluid from the plenum and into the flowpath at first and second locations along the elongate axis. The injected second fluid is formed into jets at the first and second locations, the first fluid entrains the jets such that the injected second fluid flows through the flowpath and mixes with the first fluid, and the short axis has a sufficient dimension such that the jets remain spaced from a sidewall of the member.
摘要:
A system and method for detecting an anode pressure sensor failure in a fuel cell system. The system and method include a controller that sets an initial minimum anode pressure sensor value and an initial maximum anode pressure sensor value. The controller determines a desired time interval for sampling anode pressure measurements and determines a total number of samples of anode pressure measurements to be collected by the controller from an anode pressure sensor. The controller also compares a pressure difference between the initial or a measured minimum anode pressure and the initial or a measured maximum anode pressure to a predetermined pressure difference threshold and sets a pressure sensor fault if the pressure difference between the initial or measured minimum anode pressure and the initial or maximum anode pressure is less than the predetermined pressure difference threshold.
摘要:
A skin is applied to a ceramic honeycomb. The skin is formed by applying a skin-forming composition and drying it. The skin-forming composition includes a carrier liquid, colloidal silica and/or colloidal alumina, and an inorganic filler. The filler includes an inorganic fiber. The filler may contain low aspect ratio particles that have the same or nearly the same CTE as the inorganic fiber. The filler may include a small proportion of a low aspect ratio filler particle that has a different CTE than the inorganic fiber.
摘要:
A system and method for detecting an anode pressure sensor failure in a fuel cell system. The system and method include a controller that sets an initial minimum anode pressure sensor value and an initial maximum anode pressure sensor value. The controller determines a desired time interval for sampling anode pressure measurements and determines a total number of samples of anode pressure measurements to be collected by the controller from an anode pressure sensor. The controller also compares a pressure difference between the initial or a measured minimum anode pressure and the initial or a measured maximum anode pressure to a predetermined pressure difference threshold and sets a pressure sensor fault if the pressure difference between the initial or measured minimum anode pressure and the initial or maximum anode pressure is less than the predetermined pressure difference threshold.
摘要:
The information transfer method includes: determining an indicator of a start bit in a protection time slot or a preamble time slot of a laser, and establishing a communication channel according to the indicator of the start bit; inserting Operation Administration and Maintenance (OAM) information of a Long Reach-Passive Optical Network (LR-PON) device into an upstream frame sent to an Optical Line Terminal (OLT) device by the LR-PON device at a position indicated by the indicator of the start bit, and transporting the upstream frame to the OLT device through the communication channel. Therefore, OAM functions of the LR-PON device are extended, so that the OAM information can be transported by the LR-PON device independently, and OAM characteristics of the LR-PON device are enhanced. For example, PON protection is better supported, and the speed and success ratio of protection switching of a PON system are increased.
摘要:
A semiconductor device including: a low threshold PMOS device formed over an N-type region, the source and drain of the low threshold PMOS formed in P-regions surrounded by N-regions; a low threshold NMOS device formed in a P-type region, the source and drain of the low threshold NMOS formed in N-regions surrounded by P-regions; first and second substrate bias generators, each connected to one of the low threshold devices for generating a substrate bias; a voltage source for generating substrate bias during a standby mode to reduce leakage current; wherein a low voltage threshold is established by the source and drain regions of the low threshold devices and their respective surrounding regions of opposite polarity.
摘要:
A ceramic honeycomb structure comprised of at least two separate smaller ceramic honeycombs that have been adhered together by a cement comprised of inorganic fibers and a binding phase wherein the smaller honey-combs and fibers are bonded together by the binding phase which is comprised of an silicate, aluminate or alumino-silicate. The fibers have a multi-modal size distribution in which some fibers have lengths of up to 1000 micons and other fibers have lengths in excess of 1 mm. The cement composition may be made in the absence of other inorganic and organic additives while achieving a shear thinning cement, for example, by mixing oppositely charged inorganic binders in water together so as to make a useful cement composition for applying to the smaller honeycombs to be cemented.