Showing papers on "Compression ratio published in 2014"

Gzip on a chip: high performance lossless data compression on FPGAs using OpenCL

Abstract: Hardware implementation of lossless data compression is important for optimizing the capacity/cost/power of storage devices in data centers, as well as communication channels in high-speed networks. In this work we use the Open Computing Language (OpenCL) to implement high-speed data compression (Gzip) on a field-programmable gate-arrays (FPGA). We show how we make use of a heavily-pipelined custom hardware implementation to achieve the high throughput of ~3 GB/s with more than 2x compression ratio over standard compression benchmarks. When compared against a highly-tuned CPU implementation, the performance-per-watt of our OpenCL FPGA implementation is 12x better and compression ratio is on-par. Additionally, we compare our implementation to a hand-coded commercial implementation of Gzip to quantify the gap between a high-level language like OpenCL, and a hardware description language like Verilog. OpenCL performance is 5.3% lower than Verilog, and area is 2% more logic and 25% more of the FPGA's available memory resources but the productivity gains are significant.

On the Performance of Lossy Compression Schemes for Energy Constrained Sensor Networking

Abstract: Lossy temporal compression is key for energy-constrained wireless sensor networks (WSNs), where the imperfect reconstruction of the signal is often acceptable at the data collector, subject to some maximum error tolerance. In this article, we evaluate a number of selected lossy compression methods from the literature and extensively analyze their performance in terms of compression efficiency, computational complexity, and energy consumption. Specifically, we first carry out a performance evaluation of existing and new compression schemes, considering linear, autoregressive, FFT-/DCT- and wavelet-based models , by looking at their performance as a function of relevant signal statistics. Second, we obtain formulas through numerical fittings to gauge their overall energy consumption and signal representation accuracy. Third, we evaluate the benefits that lossy compression methods bring about in interference-limited multihop networks, where the channel access is a source of inefficiency due to collisions and transmission scheduling. Our results reveal that the DCT-based schemes are the best option in terms of compression efficiency but are inefficient in terms of energy consumption. Instead, linear methods lead to substantial savings in terms of energy expenditure by, at the same time, leading to satisfactory compression ratios, reduced network delay, and increased reliability performance.

Development of a Gasoline Direct Injection Compression Ignition (GDCI) Engine

Abstract: In previous work, Gasoline Direct Injection Compression Ignition (GDCI) has demonstrated good potential for high fuel efficiency, low NOx, and low PM over the speed-load range using RON91 gasoline. In the current work, a four-cylinder, 1.8L engine was designed and built based on extensive simulations and single-cylinder engine tests. The engine features a pent roof combustion chamber, central-mounted injector, 15:1 compression ratio, and zero swirl and squish. A new piston was developed and matched with the injection system. The fuel injection, valvetrain, and boost systems were key technology enablers. Engine dynamometer tests were conducted at idle, part-load, and full-load operating conditions. For all operating conditions, the engine was operated with partially premixed compression ignition without mode switching or diffusion controlled combustion. At idle and low load, rebreathing of hot exhaust gases provided stable combustion with NOx and PM emissions below targets of 0.2g/kWh and FSN 0.1, respectively. The coefficient of variation of IMEP was less than 3 percent and the exhaust temperature at turbocharger inlet was greater than 250 C. BSFC of 280 g/kWh was measured at 2000 rpm-2bar BMEP. At medium-to-higher loads, rebreathing was not used and cooled EGR provided NOx, PM, and combustion noise below targets. MAP was reduced to minimize boost parasitics. At full load operating conditions, near stoichiometric mixtures were used with up to 45 percent EGR. Maximum BMEP was about 20 bar at 3000 rpm. For all operating conditions, injection quantities and timings were used to control mixture stratificaton and combustion phasing. Transient co-simulations of the engine system were conducted to develop control strategies for boost, EGR, and intake air temperature control. Preliminary transient tests on a real engine with high rate of load increase demonstrated potential for very good control. Cold start simulations were also conducted using an intake air heating strategy. Preliminary cold start tests on a real engine at room temperature demonstrated potential for very good cold starting. More work is needed to calibrate the engine over the full operating map and to further develop the engine control system. CITATION: Sellnau, M., Foster, M., Hoyer, K., Moore, W. et al., \"Development of a Gasoline Direct Injection Compression Ignition (GDCI) Engine,\" SAE Int. J. Engines 7(2):2014, doi:10.4271/2014-01-1300. 2014-01-1300 Published 04/01/2014 Copyright © 2014 SAE International doi:10.4271/2014-01-1300 saeeng.saejournals.org developed with significant improvements. Delphi [23, 24, 25, 26, 27] reported single-cylinder and multi-cylinder engine test results with various injectors using single, double, and triple injection strategies. Engine tests have also been performed using naphtha fuels on both modified spark-ignited engines [28] and modified diesel engines [29]. Naphtha has significantly lower octane number (RON 70 to 84) and significantly lower processing cost compared to current market gasoline. This work has shown compatibility of the PPCI combustion process with lower octane fuels in a longer term perspective. PPCI has demonstrated very good potential for very high fuel efficiency with low engine-out NOx and PM emissions using a range of gasoline-like fuels. However, towards a production solution, significant issues remain. Due to the lower exhaust enthalpy of low temperature engines using PPCI, it is difficult to produce intake boost with acceptable boost system parasitics. A practical powertrain system with robust PPCI combustion is needed, including injection, valvetrain, boost, and exhaust subsystems. The engine must also meet vehicle packaging requirements under hood and satisfy cold start and transient response requirements. As part of a US Department of Energy funded program, Delphi has been developing a multi-cylinder engine concept for PPCI combustion with current US market gasoline (RON91). The engine has four cylinders with a 1.8L displacement and was designed based on extensive simulations and single-cylinder engine tests. A multiple-late-injection (MLI) strategy with GDi-like injection pressures was selected without use of a premixed charge. The absence of classic knock and preignition limits in this process enabled a higher compression ratio of 15. The engine operates “full time” [25] over the entire operating map with partially premixed compression ignition. No combustion mode switching, diffusion controlled combustion, or spark plugs were used. Delphi uses the term Gasoline Direction Injection Compression Ignition (GDCI) in reference to this combustion process. One program objective was to build a practical GDCI engine that achieves diesel-like fuel efficiency using current market gasoline (RON91) with engine-out NOx and PM emissions below that needed for NOx or PM aftertreatment. Table 1 lists initial targets for engine testing and development. Combustion noise level (CNL) limits are shown in Figure 1. Other program objectives include demonstration of good transient load response and room temperature cold starts. In the current work, analysis and simulation tools were used to design and fabricate a new multi-cylinder GDCI engine. Design tools were used to package the powertrain in a D-class vehicle. Engine dynamometer tests were conducted over a range of operating conditions and included preliminary calibration mapping. With this data, a competitive assessment of brake specific fuel consumption (BSFC) was performed using published data for gasoline, diesel, and hybrid vehicle engines. Finally, aggressive transients with high rate of load increase were simulated and then tested on the real engine. Cold starts at various ambient temperatures were also simulated and tested on the real engine. Table 1. Preliminary Targets for Engine Testing. Figure 1. Combustion Noise Level (CNL) Limits used for Testing. GDCI CONCEPT AND INJECTION STRATEGY The GDCI engine concept combines the best of diesel and spark-ignited engine technology. Like diesel engines, the compression ratio is high, there is no intake throttling, and the mixture is lean for improved ratio of specific heats. GDCI uses a new low-temperature combustion process for partiallypremixed compression-ignition. Multiple-late-injections of gasoline (RON91) vaporize and mix very quickly at low injection pressure typical of direct injected gasoline engines. Low combustion temperatures combined with low mixture motion and reduced chamber surface area result in reduced heat losses. A schematic of the GDCI combustion chamber concept is shown in Figure 2. The engine features a shallow pent roof combustion chamber, central-mounted injector, and 15:1 compression ratio. A quiescent, open chamber design was chosen to support injection-controlled mixture stratification. Swirl, tumble, and squish were minimized since excessive mixture motion may destroy mixture stratification created during the injection process. The piston and bowl shape were matched with the injection system and spray characteristics. The bowl is a symmetrical shape and was centered on the cylinder and injector axes. The GDCI injection strategy is central to the overall GDCI concept and is depicted in the Phi-T (equivalence ratiotemperature) diagram shown in Figure 3. The color contours in Figure 3 show simulated CO emissions concentration. The injection process involves one, two, or three injections during Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)

Robust Bayesian Compressive Sensing for Signals in Structural Health Monitoring

Abstract: In structural health monitoring (SHM) systems for civil structures, massive amounts of data are often generated that need data compression techniques to reduce the cost of signal transfer and storage, meanwhile offering a simple sensing system. Compressive sensing (CS) is a novel data acquisition method whereby the compression is done in a sensor simultaneously with the sampling. If the original sensed signal is sufficiently sparse in terms of some orthogonal basis (e.g., a sufficient number of wavelet coefficients are zero or negligibly small), the decompression can be done essentially perfectly up to some critical compression ratio; otherwise there is a trade-off between the reconstruction error and how much compression occurs. In this article, a Bayesian compressive sensing (BCS) method is investigated that uses sparse Bayesian learning to reconstruct signals from a compressive sensor. By explicitly quantifying the uncertainty in the reconstructed signal from compressed data, the BCS technique exhibits an obvious benefit over existing regularized norm-minimization CS methods that provide a single signal estimate. However, current BCS algorithms suffer from a robustness problem: sometimes the reconstruction errors are very large when the number of measurements K are a lot less than the number of signal degrees of freedom N that are needed to capture the signal accurately in a directly sampled form. In this article, we present improvements to the BCS reconstruction method to enhance its robustness so that even higher compression ratios N/K can be used and we examine the trade-off between efficiently compressing data and accurately decompressing it. Synthetic data and actual acceleration data collected from a bridge SHM system are used as examples. Compared with the state-of-the-art BCS reconstruction algorithms, the improved BCS algorithm demonstrates superior performance. With the same acceptable error rate based on a specified threshold of reconstruction error, the proposed BCS algorithm works with relatively large compression ratios and it can achieve perfect loss-less compression performance with quite high compression ratios. Furthermore, the error bars for the signal reconstruction are also quantified effectively.

Optimization of operational parameters on performance and emissions of a diesel engine using biodiesel

Abstract: This work investigates the influence of compression ratio on the performance and emissions of a diesel engine using biodiesel (10, 20, 30, and 50 %) blended-diesel fuel. Test was carried out using four different compression ratios (17.5, 17.7, 17.9 and 18.1). The experiments were designed using a statistical tool known as design of experiments based on response surface methodology. The resultant models of the response surface methodology were helpful to predict the response parameters such as brake specific fuel consumption, brake thermal efficiency, carbon monoxide, hydrocarbon and nitrogen oxides. The results showed that best results for brake thermal efficiency and brake specific fuel consumption were observed at increased compression ratio. For all test fuels, an increase in compression ratio leads to decrease in the carbon monoxide and hydrocarbon emissions while nitrogen oxide emissions increase. Optimization of parameters was performed using the desirability approach of the response surface methodology for better performance and lower emission. A compression ratio 17.9, 10 % of fuel blend and 3.81 kW of power could be considered as the optimum parameters for the test engine.

A Si-Micromachined 162-Stage Two-Part Knudsen Pump for On-Chip Vacuum

Abstract: This paper investigates a two-part architecture for a Knudsen vacuum pump with no moving parts This type of pump exploits the thermal transpiration that results from the free-molecular flow in nonisothermal channels For a high compression ratio, 162 stages are serially cascaded The two-part architecture uses 54 stages designed for the pressure range from 760 to ≈ 50 Torr, and 108 stages designed for lower pressures This approach provides greater compression ratio and speed than using a uniform design for each stage Finite element simulations and analytical design analysis are presented A five-mask single-wafer fabrication process is used for monolithic integration of the Knudsen pump that has a footprint of 12 × 15 mm2 The pressure levels of each stage are measured by integrated Pirani gauges Experimental evaluation shows that, using an input power of ≈ 039 W, the evacuated chamber is reduced from 760 to ≈ 09 Torr, resulting in a compression ratio of ≈ 844 The vacuum levels are sustained during 37 days of continuous operation

Data Compression for the Exascale Computing Era - Survey

Abstract: While periodic checkpointing has been an important mechanism for tolerating faults in high performance computing HPC systems, it is cost-prohibitive as the HPC system approaches exascale. Applying compression techniques is one common way to mitigate such burdens by reducing the data size, but they are often found to be less effective for scientific datasets. Traditional lossless compression techniques that look for repeated patterns are ineffective for scientific data in which high-precision data is used and hence common patterns are rare to find. In this paper, we present a comparison of several lossless and lossy data compression algorithms and discuss their methodology under the exascale environment. As data volume increases, we discover an increasing trend of new domain-driven algorithms that exploit the inherent characteristics exhibited in many scientific dataset, such as relatively small changes in data values from one simulation iteration to the next or among neighboring data. In particular, significant data reduction has been observed in lossy compression. This paper also discusses how the errors introduced by lossy compressions are controlled and the tradeoffs with the compression ratio.

Development Approach of a Spark-Ignited Free-Piston Engine Generator

Abstract: Copyright © 2014 SAE International.The free-piston engine generator (FPEG) is a novel type of energy conversion device; it integrates a two stroke combustion engine and a linear electric machine into a single unit. As an alternative to conventional engines, the FPEG is a promising power generation system due to its simplicity and high thermal efficiency and has attracted considerable research interests recently. This paper presents the development for a spark-ignited free-piston engine generator prototype, and the design of major sub-systems is introduced. The electrical linear machine is operated as a motor to start the engine and switched to a generator after successful ignition. Ignition is one of the most crucial problems for the generating process, thus a unique control sub-system to generate ignition signals at the correct ignition timing based on the piston position was completed. Then experiments of the starting process were carried out with the prototype. The results indicate that with a fixed motor force of 110N, the maximum in-cylinder gas pressure can reach 12 bar and the compression ratio can reach 8:1. Moreover, the experiment results show a good agreement with the simulation prediction. According to the performance of the starting process, the development of the prototype is acceptable and further research on the generating process will be undertaken soon.

Real-time lossless ECG compression for low-power wearable medical devices based on adaptive region prediction

Abstract: A real-time lossless compression technique for ECG signals, which benefits wearable medical devices with stringent low-power requirements, is presented. The real-time ECG waveform is automatically classified into four regions according to its fluctuation features and the most suitable prediction method is adaptively selected from several linear prediction methods for different regions. Further proposed is the use of a modified variable length code to encode the prediction difference for a simpler transmit format. Experimental results based on three publically available test databases show that the proposed method achieves a better compression ratio with a lower prediction difference than existing state-of-the-art approaches. A very large-scale integration implementation is also demonstrated which can be used as an intellectual property core with a core area of 25 809 μm2 and which achieves a power consumption of 127 μW at 100 MHz in a 0.18 μm CMOS technology.

Effects of compression ratio, swirl augmentation techniques and ethanol addition on the combustion of CNG-biodiesel in a dual-fuel engine

Abstract: The diminishing resources and continuously increasing cost of petroleum in association with their alarming pollution levels from diesel engines has led to an interest in finding alternative fuels to diesel. Emission control and engine efficiency are two of the most important parameters in current engine design. The impending introduction of emission standards such as Euro IV and Euro V has forced research towards developing new technologies for combating engine emissions. This paper examines the effects of compression ratio, swirl augmentation techniques and ethanol addition on the combustion of compressed natural gas (CNG) blended with Honge oil methyl esters (HOME) in a dual fuel engine. The present results show that the combustion of HOME and 15% ethanol blend with CNG induction in a dual-fuel engine operated in optimized parameters at an injection timing of 27° Before Top Dead Centre and a compression ratio of 17.5 resulted in acceptable combustion emissions and improved brake thermal efficiencies. Th...