Abstract: Provided is an operation method for a memory device, the operation method comprising: performing an erase operation; performing a verify-read operation on a memory cell to generate a cell current, the memory cell including a first transistor and a second transistor; checking whether the cell current is lower than a first cell current threshold; when the cell current is not lower than the first cell current threshold, increasing a memory gate voltage until the cell current is lower than the first cell current threshold, wherein the memory gate voltage is applied to the first transistor; fixing the memory gate voltage and increasing a drain voltage; checking whether the cell current is lower than a second cell current threshold; and if the cell current is not lower than the second cell current threshold, increasing the drain voltage until the cell current is lower than the second cell current threshold.

Abstract: An inference engine for a neural network uses a compute-in-memory array storing a kernel coefficients. A clamped input matrix is provided to the compute-in-memory array to produce an output vector representing a function of the clamped input vector and the kernel. A circuit is included receiving an input vector, where elements of the input vector have values in a first range of values. The circuit clamps the values of the elements of the input vector a limit of a second range of values to provide the clamped input vector. The second range of values is more narrow than the first range of values, and set according to the characteristics of the compute-in-memory array. The first range of values can be used in training using digital computation resources, and the second range of values can be used in inference using the compute-in-memory array.

Abstract: An in-memory computation device including a memory array, p×q analog to digital converters (ADCs) and a ladder adder is provided. The memory array is divided into p×q memory tiles, where p and q are positive integers larger than 1. Each of the memory tiles has a plurality local bit lines coupled to a global bit line respectively through a plurality of bit line selection switches. The bit line selection switches are turned on or cur off according to a plurality of control signals. The memory array receives a plurality of input signals. The ADCs are respectively coupled to a plurality of global bit lines of the memory tiles. The ADCs respectively convert electrical signals on the global bit lines to generate a plurality of sub-output signals. The ladder adder is coupled to the ADCs, and performs an addition operation on the sub-output signals to generate a calculation result.

Abstract: A method of fabricating a semiconductor device includes the steps of: providing a semiconductor structure including a memory region and a logic region. The semiconductor structure includes a first interlayer dielectric and at least one magnetoresistive random access memory (MRAM) cell disposed on the first interlayer dielectric, and the MRAM cell is disposed in the memory region; depositing a second interlayer dielectric covering the first interlayer dielectric and the at least one MRAM cell; depositing a mask layer conformally covering the second interlayer dielectric; perform a planarization process to remove the mask layer in the memory region; after the step of performing the planarization process, removing the mask layer in the logic region.

Abstract: A method for fabricating a semiconductor device includes the steps of: forming a first inter-metal dielectric (IMD) layer on a substrate; forming a first metal interconnection and a second metal interconnection in the first IMD layer; forming a channel layer on the first metal interconnection and the second metal interconnection; forming a magnetic tunneling junction (MTJ) stack on the channel layer; and removing the MTJ stack to form a MTJ.

Abstract: A method for fabricating a semiconductor device includes the steps of: forming a magnetic tunneling junction (MTJ) on a substrate; forming a first inter-metal dielectric (IMD) layer around the MTJ; forming a first metal interconnection adjacent to the MTJ; forming a stop layer on the first IMD layer; removing the stop layer to form an opening; and forming a channel layer in the opening to electrically connect the MTJ and the first metal interconnection.

Abstract: A method for fabricating a semiconductor device includes the steps of: forming a first metal interconnection on a substrate; forming a stop layer on the first metal interconnection; removing the stop layer to form a first opening; forming an electromigration enhancing layer in the first opening; and forming a second metal interconnection on the electromigration enhancing layer. Preferably, top surfaces of the electromigration enhancing layer and the stop layer are coplanar.

Abstract: A semiconductor device includes: a substrate comprising a magnetic tunneling junction (MTJ) region and a logic region, a MTJ on the MTJ region, a top electrode on the MTJ, a connecting structure on the top electrode, and a first metal interconnection on the logic region. Preferably, the first metal interconnection includes a via conductor on the substrate and a trench conductor, in which a bottom surface of the trench conductor is lower than a bottom surface of the connecting structure.

Abstract: A memory apparatus and compensation method for a computation result thereof are provided. The memory apparatus includes a memory sub-block, a reference memory sub-block and a control circuit. During a computation phase, the memory sub-block receives an input signal, and generates a computation result by a multiply-accumulate operation according to the input signal. The reference memory sub-block includes a plurality of memory cells pre-programmed with a reference weight value. The reference memory sub-block receives a reference input signal during a calibration phase, and generates a reference computation value by a multiply-accumulate operation according to the reference input signal and the reference weight value. The control circuit generates an adjustment value according to the reference computation value and a standard computation value, and during the computation phase, adjusts the computation result according to the adjustment value to generate an adjusted computation result.

Abstract: An artificial intelligence accelerator receives a binary input data set and a selected layer of layers of overall weight pattern. The artificial intelligence accelerator includes processing tiles and a summation output circuit. Each processing tile receives one of input data subsets of the input data set and performs a convolution operation on weight blocks of each sub weight pattern of the overall weight pattern to obtain weight operation values and then obtains a weight output value expected from a direct convolution operation on the input data subset with the sub weight pattern through performing a multistage shifting and adding operation on the weight operation values. The summation output circuit sums up the plurality of weight output values through a multistage shifting and adding operation, so as to obtain a sum value expected from a direct convolution operation performed on the input data set with the overall weight pattern.

Abstract: A MRAM structure, which is provided with multiple source lines between active areas, each source line has multiple branches electrically connecting with the active areas at opposite sides in alternating arrangement. Multiple word lines traverse through the active areas to form transistors. Multiple storage units are disposed between the word lines on the active areas in staggered array arrangement, and multiple bit lines electrically connect with storage units on corresponding active areas, wherein each storage cell includes one of the storage unit, two of the transistors respectively at both sides of the storage unit, and two branches of the source line.

Abstract: A method for fabricating semiconductor device includes the steps of: forming a substrate having a magnetic tunneling junction (MTJ) region and a logic region; forming a MTJ on the MTJ region; forming a top electrode on the MTJ; forming an inter-metal dielectric (IMD) layer around the MTJ; removing the IMD layer directly on the top electrode to form a recess; forming a first hard mask on the IMD layer and into the recess; removing the first hard mask and the IMD layer on the logic region to form a contact hole; and forming a metal layer in the recess and the contact hole to form a connecting structure on the top electrode and a metal interconnection on the logic region.

Abstract: A method for fabricating semiconductor device includes the steps of: forming a magnetic tunneling junction (MTJ) stack on a substrate; forming a top electrode on the MTJ stack; performing a first patterning process to remove the MTJ stack for forming a first MTJ; forming a first inter-metal dielectric (IMD) layer around the first MTJ; and performing a second patterning process to remove the first MTJ for forming a second MTJ and a third MTJ.

Abstract: An in-memory computing device including a plurality of memory cell arrays and a plurality of sensing amplifiers are provided. The memory cell arrays respectively receive a plurality of input signals. The input signals are divided into a plurality of groups. The groups respectively have at least one partial input signal. The at least one partial input signal of each of the groups has a same value. Numbers of the at least one partial input signal in the groups sequentially form a geometric sequence with a common ration equal to 2. The memory cell arrays respectively provide a plurality of weightings, and perform multiply-add operations respectively according to the received input signals and the weightings to generate a plurality of computation results. The sensing amplifiers respectively generate a plurality of sensing results according to the computation results.

Abstract: A memory layout structure, which is provided with multiple source lines between active areas, each source line has multiple branches electrically connecting with the active areas at opposite sides in alternating arrangement. Multiple word lines traverse through the active areas to form transistors. Multiple storage units are disposed between the word lines on the active areas in staggered array arrangement, and multiple bit lines electrically connect with all storage units on a corresponding active area, wherein each storage cell includes one of the storage unit, two of the transistors respectively at both sides of the storage unit, and two branches of the source line.

Abstract: A semiconductor device includes a first magnetic tunneling junction (MTJ) and a second MTJ on a substrate and a dummy MTJ between the first MTJ and the second MTJ, in which a bottom surface of the dummy MTJ is not connected to any metal. Preferably, the semiconductor device further includes a first metal interconnection under the first MTJ, a second metal interconnection under the second MTJ, and a first inter-metal dielectric (IMD) layer around the first metal interconnection and the second metal interconnection and directly under the dummy MTJ.

Abstract: A method of fabricating a semiconductor device includes the steps of: providing a semiconductor structure including a memory region and a logic region. The semiconductor structure includes a first interlayer dielectric and at least one magnetoresistive random access memory (MRAM) cell disposed on the first interlayer dielectric, and the MRAM cell is disposed in the memory region; depositing a second interlayer dielectric covering the first interlayer dielectric and the at least one MRAM cell; depositing a mask layer conformally covering the second interlayer dielectric; perform a planarization process to remove the mask layer in the memory region; after the step of performing the planarization process, removing the mask layer in the logic region.

Abstract: A method for fabricating semiconductor device includes the steps of: forming a substrate having a magnetic tunneling junction (MTJ) region and a logic region; forming a MTJ on the MTJ region; forming a top electrode on the MTJ; forming an inter-metal dielectric (IMD) layer around the MTJ; removing the IMD layer directly on the top electrode to form a recess; forming a first hard mask on the IMD layer and into the recess; removing the first hard mask and the IMD layer on the logic region to form a contact hole; and forming a metal layer in the recess and the contact hole to form a connecting structure on the top electrode and a metal interconnection on the logic region.

Abstract: A memory layout structure, which is provided with multiple source lines between active areas, each source line has multiple branches electrically connecting with the active areas at opposite sides in alternating arrangement. Multiple word lines traverse through the active areas to form transistors. Multiple storage units are disposed between the word lines on the active areas in staggered array arrangement, and multiple bit lines electrically connect with all storage units on a corresponding active area, wherein each storage cell includes one of the storage unit, two of the transistors respectively at both sides of the storage unit, and two branches of the source line.

Abstract: The present disclosure provides a back cover assembly of a portable electronic device. The back cover assembly includes a substrate structure and a coil structure. The substrate structure includes a metal substrate and a first non-metal substrate connected with the metal substrate. The coil structure is matched with an IC chip for generating an antenna magnetic field that passes through the first non-metal substrate without matching with the metal substrate. The coil structure has a first coil portion and a second coil portion connected to the first coil portion, the first coil portion is disposed above the metal substrate, the second coil portion is disposed above the first non-metal substrate, and the percentage of the first coil portion to the coil structure is larger than that the percentage of the second coil portion to the coil structure.