This paper presents a hybrid architecture for a neural network (NN) trained by a particle swarm optimization (PSO) algorithm. The NN is implemented on the hardware side while the PSO is executed by a processor on the software side. In addition, principal component analysis (PCA) is also applied to reduce correlated information. The PCA module is implemented in hardware by the SystemVerilog programming language to increase operating speed. Experimental results showed that the proposed architecture had been successfully implemented. In addition, the hardware-based NN trained by PSO (NN-PSO) program was faster than the software-based NN trained by the PSO program. The proposed NN-PSO with PCA also obtained better recognition rates than the NN-PSO without-PCA.

A novel spur cancelled clock generator (SCCG) capable of recovering RX sensitivity degradations caused by digital clocks in wireless SoCs is presented. Clock spurs that degrade RX sensitivities are canceled by applying the SCCG to digital circuits or ADCs. The SCCG is integrated into a Bluetooth Low Energy (BLE) SoC fabricated in a 65 nm CMOS process. A measured clock spur reduction of 34 dB and an RX sensitivity recovery of 5 dB are achieved by the proposed SCCG. The power consumption and occupied area of the SCCG is only 18 µW and 40 μm × 120 μm, respectively.

This paper presents the design of a second-order and a fourth-order bandpass filter (BPF) for 60 GHz millimeter-wave applications in 0.18 µm CMOS technology. The proposed on-chip BPFs employ the folded open loop structure designed on pattern ground shields. The adoption of a folded structure and utilization of multiple transmission zeros in the stopband permit the compact size and high selectivity for the BPF. Moreover, the pattern ground shields obviously slow down the guided waves which enable further reduction in the physical length of the resonator, and this, in turn, results in improvement of the insertion losses. A very good agreement between the electromagnetic (EM) simulations and measurement results has been achieved. As a result, the second-order BPF has the center frequency of 57.5 GHz, insertion loss of 2.77 dB, bandwidth of 14 GHz, return loss less than 27.5 dB and chip size of 650 µm810 µm (including bonding pads) while the fourth-order BPF has the center frequency of 57 GHz, insertion loss of 3.06 dB, bandwidth of 12 GHz, return loss less than 30 dB with chip size of 905 µm810 µm (including bonding pads).

This paper proposes an Integer Linear Programming (ILP)-based power minimization method by partitioning into regions, first, with three different VDD's(PM3V), and, secondly, with two different VDD's(PM2V). To reduce the solving time of triple-VDD case (PM3V), we also proposed a partitioned ILP method(p-PM3V). The proposed method provides 29% power saving on the average in the case of triple-VDD compared to the case of single VDD. Power reduction of PM3V compared to Clustered Voltage Scaling (CVS) was about 18%. Compared to the unpartitioned ILP formulation(PM3V), the partitioned ILP method(p-PM3V) reduced the total solution time by 46% at the cost of additional power consumption within 1.3%.

The voltage margin of an embedded DRAM's sense operation has been shrinking with the scaling of process technology. A method to estimate this margin would be a key to optimizing the memory array configuration and the size of the sense transistor. In this paper, the voltage margin of the sense operation is theoretically analyzed. The accuracy of the proposed voltage margin model was confirmed on a 0.13-µm eDRAM test chip, and the results of calculation were generally in agreement with the measured results.

A large memory is typically designed with multiple identical memory blocks for reducing delay and power. The circuit verification of individual memory blocks can be effectively handled by the Symbolic Trajectory Evaluation (STE) approach. However, if multiple memory blocks are integrated into a single system, the STE approach cannot verify it economically. This paper introduces algorithms for verifying block-level connectivity of memories. The verification time of a large memory can be reduced drastically by using bottom-up verification scheme. That is, a memory block is first verified thoroughly, and then only the interconnection between memory blocks of the large memory needs to be verified. The proposed verification algorithms require (3n+2(log2n+1)+3log2m) Read/Write operations for a 2nm-bit memory, where n and m are the address width and data width, respectively. Also, the algorithms can verify 100% of the inter-port and intra-port signal misplaced faults of the address, data input, and data output ports.

A low power 16 Mb embedded DRAM (eDRAM) macro is fabricated using 0.15 µm logic -based embedded DRAM process technology. A 0.5 µm2 CUB (apacitor nder it-line) DRAM cell is newly developed for this process. Novel start-up and dynamic fuse-data loading circuit are developed to realize easy customization of memory capacities with minimum area penalty. A new write-mask control circuit using write-gate sense-amplifier is adopted in order to apply column shift-redundancy circuit. Various low power technologies including unique "non-precharge read-data bus" method are applied. In the test-chip adopting new process-technology and three original circuit-design techniques, random column operation of 166 MHz and data retention power of 123 µW are demonstrated at 1.5 V power supply.

A 23.3 mm2 32 Mb embedded DRAM (eDRAM) macro has been fabricated using 0.18 µm triple-well 4-metal embedded DRAM process technology to realize an accelerated 3-D graphics controller. The array architecture, using a dual-port sense amplifier, achieves the column access latency of two cycles at 222 MHz and a peak data rate of 14.2 4 GB/s at 4 macros. The process cost has been kept low by using VT-MOS circuit technology and taking advantage of a characteristic of dual-gate oxide process technology. A tRAC of 11.6 ns at 2.0 V is achieved using a 'pre-detect redundancy' circuit.

The features for the integration of 1Tr/1C DRAM and logic for graphic and multimedia applications are surveyed. The key circuit/process technology for large scale embedded DRAM cores is described. The methods to improve transistor performance and gate density are shown. Noise immunity design and easy customization techniques are also introduced.

Entire systems embedded in a chip and consisting of a processor, memory, and system-specific peripheral hardware are now commonly contained in commodity electronic devices. Cost minimization of these systems is of paramount economic importance to manufactures of these devices. By employing a variable configuration processor in conjunction with a multi-precision compiler generator, we show that there are situations in which considerable system cost reduction can be obtained by synthesizing a CPU that is narrower than the largest variable in the application program.

Integrating the power supply and signal processing circuit into one chip is an important step towards achieving a system-on-chip. This paper reviews and looks at the current technologies and their trends for power supply components such as DC-DC converters, intelligent power LSIs, and thin-film magnetic devices for the system-on-chip. A device structure has been proposed for the system-on-chip that is based on a quasi-SOI technique, in which the buried oxide layer is partially removed from the SOI substrate. In this structure, the CMOS devices for the digital signal-processing circuit and the bipolar transistors are formed in a conventional SOI region, and the CMOS analog devices and high-voltage devices are formed in a quasi-SOI region.

Digital-switching noise coupled into sensitive analog circuits is a critical problem in large-scale integration of mixed analog and digital circuits. This paper describes noise coupling of this kind, especially, through the substrate in CMOS integrated circuits, and reviews recent technical solutions to this noise problem. Simplified models have been developed to simulate the substrate coupling rapidly and accurately. A method using a CMOS comparator was proposed for measuring the effects of substrate noise, and equivalent waveforms of actual substrate noise were obtained. A circuit tecnique, called active guard band filtering, that controls the noise source is a new approach to substrate noise decoupling. CAD methods for handling substrate-coupled switching noise are making design verification possible for practical mixed-signal LSIs.