
We present a new design of an electronically tunable linear MOS resistor circuit that operates in the subthreshold saturation region, supported with mathematical derivations and simulation results using CSM0.13μm technology. For a given potential difference across the MOS resistor, its gate voltage will be automatically biased through feedback to provide the correct amount of current based on the desired resistance set through the bias current. In comparison with an existing design [1], the proposed design offers equal tunabilty with 36 less transistors for unidirectional current and 28 less transistors with one more bias current transistor for bidirectional current. A bias current ranging between 10nA to 100nA offers a tunable linear resistance between 20MΩ to 140MΩ.
This paper presents a compact 14-18 GHz 6-bit digital phase shifter based on 0.35-μm SiGe BiCOMS technology. Three different topologies are adopted to achieve the desired phase shift with low insertion loss, accurate phase shifting and small parasitic amplitude modulation. The simulated performance of all 64 states of the phase shifter demonstrates an insertion loss of 9±0.9 dB and a RMS phase error <;4° with P 1dB better than 11dBm. The input return loss and output return loss are better than -12 dB and -13.5dB over the 14-18 GHz frequency range respectively. And the chip size of this phase shifter is only 0.79×0.42 mm 2 excluding PADs.
In this paper a novel method to design a transformer based balun with isolation between output ports and matching characteristic at all ports is proposed. The main issue of transformer balun at millimeter-wave frequencies is that it has poor matching and isolation performance. In this work, an isolation circuit is introduced to place between output ports in the transformer balun design. As a validation, comparison has been made between the balun with and without isolation circuit, which has shown great performance improvement of proposed design. Moreover, design example is given for millimeter-wave on-chip implementation of this isolation circuit in standard silicon based technology with 6 metal layers.
GPU has become an important component of the high performance computing system and its principal duty is parallel computing rather than graphical display. Determining the power and energy consumption is necessary to the scaling of GPU. This paper presents a statistic model to evaluate the power and energy consumption of AMD's integrated GPU (iGPU). By collecting the data of performance counters from real hardware measurements, we apply linear regression method to estimate the energy consumed by iGPU. Our results show that the median absolute error is less than 3%. Due to the limits of profiling tool CodeXL, power sampling period is much longer than the kernel execution time. We propose a kernel extension method to lengthen the kernel execution time so that we can deal with this problem. Furthermore, we conduct a study on the importance of performance counters and explore the possibility to simplify our statistic model. The results suggest that the accuracy and stability is still acceptable when there are only 12 performance counters in the simplified model.
The design of a 40 nm CMOS technology embedded switched-mode power supply (SMPS) for low-power mobile silicon-on-chip (SoC) is presented. The embedded SMPS operates directly from a Li-Ion battery capable of a wide input voltage range of 2.3 to 5.5V while the maximum voltage rating of the transistors used is 2.75V. This is achieved using integrated low-power voltage regulators and a novel dip compensation scheme. In addition, the SMPS is designed to include PWM and PFM modes to achieve high power efficiency over a wide load current range. To take advantage of the advanced 40 nm process, the PWM controller is implemented digitally. The digital controller, consisting of a successive approximation (SAR) ADC, digital compensation filter, and digital sigma delta PWM, is implemented to minimize silicon area and achieve fast transient response. The SMPS switches at 4.5 MHz and has peak efficiency of 83%, delivering a load of 240 mW when converting from 3.6V to 1.2V. It is capable of delivering a maximum power of 960 mW.