Collaborators

• P. Holloway National Semiconductor

Graduate Students

• S. Chung, Res. Asst., EECS
• J. Dong, Res. Asst., EECS
• P. Godoy, Res. Asst. EECS
• T. Khanna, Res. Asst., EECS
• Z. Li, Res. Asst., EECS
• O. Ogunnika, Res. Asst., EECS
• W. Sanchez, Res. Asst., EECS
• J. Spaulding, Res. Asst., EECS

Support Staff

• C. Kinsella, Administrative Assistant II

Publications

J. L. Bohorquez, M. Yip, A. P. Chandrakasan, J. L. Dawson, “A Digitally-Assisted Sensor Interface for Biomedical Applications,” VLSI Symposium on Technology and Circuits, Honolulu, HI, June 15-18, 2010.

W. Sanchez, C. G. Sodini, J. L. Dawson, “An Energy Management IC for Bio-Implants Using Ultracapacitors for Energy Storage,” VLSI Symposium on Technology and Circuits, Honolulu, HI, June 15-18, 2010.

S. Chung, P.A. Godoy, T.W. Barton, D.J. Perrault, and J.L. Dawson, “Asymmetric Multilevel Outphasing Transmitter Using Class-E PAs with Discreet Pulse Width Modulation,” International Microwave Symposium. Anaheim, CA, May 23-28, 2010.

N. Verma, A. Shoeb, J. L. Bohorquez, J. L. Dawson, J. Guttag, and A. P. Chandrakasan, “A Micro-power EEG Acquisition SoC with Integrated Feature-Extraction Procesor for a Chronic Seizure Detection System,” IEEE Journal of Solid-State Circuits, pp. 804-816, Apr. 2010.

P. Godoy, D. J. Perreault, and J. L. Dawson, “Outphasing Energy Recovery Amplifier with Resistance Compression for Improved Efficiency,” IEEE Transactions on Microwave Theory and Techniques, pp. 2895-2906, Dec. 2009.

H. Boo, S. Chung, and J.L. Dawson, “Adaptive Predistortion Using a ΔΣ Modulator for Automatic Inversion of Power Amplifier Nonlinearity,” IEEE Transactions on Circuits and Systems II: Express Briefs, pp. 901-905, Dec. 2009.

J.L. Bohorquez, A.P. Chandrakasan, and J.L. Dawson, “Frequency-Domain Analysis of Super-Regenerative Amplifiers,” IEEE Transactions on Microwave Theory and Techniques, pp. 2882-2894, Dec. 2009.

S. Chung, and J.L. Dawson, “A 73.1dB SNDR Digitally Assisted Subsampler for RF Power Amplifier Linearization Systems,” VLSI Symposium on Technology and Circuits, Japan, pp. 148-149, June 2009.

S. Chung, P.A. Godoy, T.W. Barton, E.W. Huang , D.J. Perreault, and J.L. Dawson, “Asymmetric Multilevel Outphasing Architecture for Multi-standard Transmitters,” IEEE Radio Frequency Integrated Circuits Symposium, Boston, MA, pp. 237-240, June 2009.

J.L. Bohorquez, A.P. Chandrakasan, and J. L. Dawson, “A 350μW CMOS MSK Transmitter and 400μW OOK Super-Regenerative Receiver for Medical Implant Communications,” IEEE Journal of Solid-State Circuits (VLSI Symposium Special Issue), pp. 1248-1259, Apr. 2009

With the continued communication systems trend towards high data rates come increasingly high signal bandwidths. If the carrier frequency is not increased along with the bandwidth, then the resulting system requires an ultra wideband transmitter. With a carrier frequency at mm-wave frequencies, however, a signal bandwidth of, for example, 2 GHz becomes a relatively small fractional bandwidth.

Our objective is to develop a transmitter that not only is linear and efficient but also operates at high powers at mm-wave frequencies. By building on the Asymmetric Multilevel Outphasing (AMO) architecture presented in ^[1] , we intend to employ mm-wave and architectural techniques to achieve a data rate of 2 GSymbols/second. With our particular focus on high bandwidth and high average efficiency, combined with the difficulty of designing efficient mm-wave power amplifiers (PAs), we have modified the AMO architecture to improve its data rate and theoretical average efficiency.

AMO works as a hybrid of polar and outphasing transmitters. The signal to be transmitted is decomposed into two outphased signals, as in the LINC architecture ^{[2] [3]} , but with discretely and independently varying amplitudes. Figure 1 shows a block diagram of the architecture. At high carrier frequencies it becomes necessary to use multiple parallel PA cells with their outputs combined. As shown, the cells are grouped in an advantageous way: into four PA blocks with independently driven amplitude. The additional amplitude paths on average result in reduced loss from outphasing as compared to the 2-way AMO system in ^[3] . A time-interleaved digital-to-RF phase modulator is employed for the high-bandwidth requirement. Weighted quadrature signals at the carrier frequency are combined to produce an output with an arbitrary phase shift as in ^[4] . As shown in Figure 2, with 2-way time-interleaving, the phase modulator can operate at twice the speed of the current-steering DACs, and the resulting data rate is doubled.

1. S. Chung, P. A. Godoy, T. W. Barton, Z. Li, T. W. Huang, D. J. Perreault, and J. L. Dawson, “Asymmetric multilevel outphasing architecture for multi-standard transmitters,” in Proc. IEEE Symposium on Radio Frequency Integrated Circuits, pp. 237-240, Jun. 2009, Boston, MA.
2. D. Cox, “Linear amplification with nonlinear components,” IEEE Transactions on Communications, vol. 12, pp. 1942-1945, Dec. 1974.
3. Y.-J. Chen, K.-Y. Jheng, A.-Y. Wu, H.-W. Tsao, and B. Tzeng, “Multilevel LINC system design for wireless transmitters,” in Proc. International Symposium on VLSI Design, Automation, and Test, pp. 25-27, Apr. 2007.
4. M. E. Heidari, M. Lee, and A. A. Abidi, “All digital outphasing modulator for a software-defined transmitter,” IEEE Journal of Solid State Circuits, vol. 44, pp. 1260-1271, Apr. 2009.

The proposed research program has two primary goals. The first goal is to improve the evaluation and treatment of patients with diabetes and a variety of movement disorders including Parkinson’s disease, restless leg syndrome, and essential tremor, by allowing doctors to continuously monitor relevant biomarkers over much longer time scales and with better precision than currently possible. The second goal is that the proposed implant be a platform for electronic sensory monitoring that is inexpensive and flexible and that can be used with a wide variety of sensors and for a wide variety of purposes, such as chemical sensors for monitoring blood chemistry.  In this work, we develop an energy-efficient analog-to-digital converter designed to operate with a power management scheme using ultracapacitors as opposed to a battery.

Two techniques are employed to save on energy for the entire system.  The first is to use adiabatic charging ^{[1] [2]} of the capacitors contained in the SAR ADC.  This application is ideal for adiabatic techniques because of the low frequency of operation and the ease at which we can reclaim energy from discharging the capacitors.  Figures 1 shows the single-reference, differential ADC topology for implementing adiabatic charging.  The second technique is to employ compressive sampling (CS) ^{[3] [4]} to sample at a frequency lower than the Nyquist rate.  Leveraging the fact that tremor data is sparse in the frequency domain, we can implement the CS technique before the ADC to save energy and decrease the size of the memory.  Figure 2 shows recovery of real tremor data with only 25% of the Nyquist required samples.  In our application, both energy and size are important bottlenecks because the ADC is the dominant power consumer, and memory size is a significant factor for long-term data storage.

1. L. J. Svensson and J. G. Koller, “Driving a capacitive load without dissipating fCV²,” in IEEE Symposium on Low Power Electronics, pp. 100-101, Oct. 1994.
2. J. G. Koller and L. J. Svensson, “Adiabatic charging without inductors,” USC/ISI Technical Report ACMOSTR-3a, Feb. 8, 1994.
3. E. Candes, “Compressive sampling,” Int. Congress of Mathematics, Madrid, Spain, 2006, pp. 1433-1452.
4. P. Bofounos, J. Romberg, and R. Baraniuk, “Compressive sensing – theory and applications,” presented at IEEE International Conference on Acoustics, Speech, and Signal Processing, Las Vegas, NV, 2008.

This project aims to develop a device to monitor tremor for treatment of neurological disorders such as Parkinson’s disease (PD). PD is the third most common chronic disease of elderly people ^[1] . Currently, PD evaluation is mostly done qualitatively using the unified Parkinson’s disease rating scale or the Hoehn & Yahr scale ^[2] . Such evaluation is subjective to the doctors’ experience and impressions. Current approach makes it difficult to accurately monitor the disease’s progression, evaluate the efficacy of the drugs used for treatment, or determine whether deep brain stimulation has been effective. A long-term continuous tremor monitoring device that can collect movement data to assist quantitative analysis proves to be useful and needed.

Figure 1 shows the overview of the reader tag communication system. The initial tag prototype consists of the analog frontend, an Altera FPGA board, and the ADC sensor circuitry. The analog frontend includes a LC matching network, 5-stage voltage multiplier, demodulator, and the backscatter modulator. The FPGA board implements control logic that conforms to the industry EPC Class 1 Gen 2 RFID communication standard at 902-928 MHz. The ADC sensor circuitry includes an 8-bit ADC and an inertial sensor that collects x, y, and z axial data.  Data is transmitted through ASK backscatter modulation, encoded in an FM0 or Miller modulated subcarrier, to the commercial RFID reader. The reader is connected to the host PC, where application software receives and processes the data.

Figure 2 shows the baseband backscattered data encoded in FM0 on a digital oscilloscope. Data 0 consists of a high and low phase whereas data 1 is the entire high phase. The maximum transmitted data rate, as shown in the figure, is around 640 Kbps. The system can achieve a read range up to 3 meters. ^[3]

1. M. Manto, M. Topping, M. Soede, J. Sanchez-Lacuesta, W. Harwin, J. Pons, J. Williams, S. Skaarup, and L. Normie, “Dynamically responsive intervention for tremor suppression,” Engineering in Medicine and Biology Magazine, IEEE, vol. 22, pp. 120-132, 2003.
2. G. Rigas, A. T. Tzallas, D. G. Tsalikakis, S. Konitsiotis, and D. I. Fotiadis, “Real-time quantification of resting tremor in the Parkinson’s disease,” in Engineering in Medicine and Biology Society, 2009. EMBC 2009. Annual International Conference of the IEEE, 2009, pp. 1306-1309.
3. U. Karthaus and M. Fischer, “Fully integrated passive UHF RFID transponder IC with 16.7-µW minimum RF input power,” IEEE Journal of Solid-State Circuits, vol. 38, no. 10, pp. 1602- 1608, Oct. 2003.

With modern digital communications applications moving towards increasingly high data rates, efficient spectral use requires modulation with dense constellations. The denser the constellation, the more important the average efficiency of the power amplifier becomes. Conventional power amplifier (PA) options such as Class A or B amplifiers have the necessary linearity for dense constellations but have poor efficiency at low output powers. Outphasing solutions, which use efficient switching PAs and power combining, seek to break the efficiency/linearity tradeoff in PA design. This work focuses on a lossless multi-way power combining and outphasing system originally presented in ^[1] that provides both high efficiency and linear output control.

Outphasing was proposed by Chireix for radio frequency (RF) transmitters using a lossless combiner and two PAs ^[2] . The effective load impedance seen by each PA during operation depends on the commanded total output power. The efficiency of each PA itself depends on the load. This effect can be described using the power factor of the load, with power factor defined as the ratio of the real part of the load impedance to the magnitude of the impedance. The PA efficiency for a typical PA design is proportional to power factor, so that the PA load reactance is a key design criterion for a combiner. A related approach to Chireix outphasing, linear amplification with nonlinear components (LINC), enforces a unity power factor for the PAs by using an isolating combiner ^[3] . The combiner input port impedance is always 50 ohms, independent of the total output power. Since the constant load means that the switching PAs are operating at constant output power, power not delivered to the load must be dissipated in an isolation resistor. As a result the system efficiency is degraded for low output power levels.

The goal of this work is to develop a new lossless multi-way outphasing system that overcomes the limitations of previous outphasing systems. Based on the system proposed in ^[1] , the outphasing system shown in Figure 1 uses four switching PAs and an outphasing control strategy that provides linear control over a wide dynamic range while limiting the reactance of the loads seen by each PA. By providing ideally lossless combining and nearly resistive loading, this system is expected to allow high average efficiency for amplification of signals even with large peak-to-average power ratios. A combiner design parameter determines the relationship between branch reactances and determines the tradeoff between dynamic range and maximum load reactance. Figure 2 shows the magnitude and phase of the load impedances of the new power combiner and outphasing system for one choice of design parameters. The figure shows that the new system has small a reactive load component over a wide power range.

1. D. J. Perreault, “A new power combining and outphasing modulation system for high-efficiency power amplification,” IEEE Transactions on Circuits and Systems I: Regular Papers, Oct. 2011.
2. H. Chireix, “High power outphasing modulation,” Proc. IRE, Nov. 1935, vol. 23, no. 11, pp. 1370-1392.
3. D. Cox, “Linear amplification with nonlinear components,” IEEE Transactions on Communications, vol. 12, pp. 1942-1945, Dec. 1974.