Small Timing Errors Research Articles

A 16-nm Always-On DNN Processor With Adaptive Clocking and Multi-Cycle Banked SRAMs

Always-on subsystems in mobile/Internet of Things (IoT) SoCs process a variety of real-time sensor data deep neural network (DNN) classification workloads in a heavily constrained energy budget. This can be achieved with robust, low-voltage circuits, and specialized hardware accelerators. We present a 16-nm always-on DNN processor, which consists primarily of a microcontroller and a DNN accelerator with on-chip SRAM for the model weights. The design operates robustly from 0.4 to 1-V, with calibration-free automatic voltage/frequency tuning provided by tracking small non-zero razor timing error rates. A novel timing error-driven synchronization-free adaptive clocking scheme significantly reduces the adaptation latency to provide resilience to fast on-chip supply noise and reduce margins. To accommodate the tight energy constraints of always-on IoT workloads, we implement a multi-cycle SRAM read scheme that allows the memory voltage to scale at iso-throughput, improving energy efficiency across the entire operating range. The wide operating range allows for high performance at 1.36 GHz, low-power consumption downs to 750 μW, and stateof-the-art raw efficiency at 16-bit precision of 750 GOPS/W dense or 1.81 TOPS/W sparse.

Time Variations in Ocean Bottom Pressure from a Few Hours to Many Years: In Situ Data, Numerical Models, and GRACE Satellite Gravimetry

AbstractIn situ ocean bottom pressure (OBP) obtained from 154 different locations irregularly scattered over the globe is carefully processed to isolate signals related to the ocean general circulation and large‐scale sea level changes. Comparison against a global numerical ocean model experiment indicates poor correspondence for periods below 24 hr, possibly related to residual tidal signals and small timing errors in the atmospheric forcing applied to the ocean model. Correspondence increases rapidly for periods between 3 and 10 days, where wind‐driven dynamics are already well understood and consequently well implemented into numerical models. Coherence decreases again for periods around 30 days and longer, where processes not implemented into ocean general circulation models as barystatic sea level changes become more important. Correspondence between in situ data and satellite‐based OBP as obtained from the Gravity Recovery and Climate Experiment (GRACE) German Research Centre for Geosciences RL05a gravity fields critically depends on the postprocessing of Level‐2 Stokes coefficients that also includes the selection of appropriate averaging regions for the GRACE‐based mass anomalies. The assessment of other available GRACE Level‐2 products indicates even better fit of more recent solutions as ITSG‐Grace2016 and the Center for Space Research and Jet Propulsion Laboratory RL05 mascons. In view of the strong high‐frequency component of OBP, however, a higher temporal resolution of the oceanic GRACE products would be rather advantageous.

Gravimetric system using high-speed double switching valves for low liquid flow rates

This paper presents a gravimetric system developed to perform the static weighing with flying-start-and-stop (SW-FSS) calibration method at low liquid flow rates using a pair of identical high-speed switching valves as a flow diverter. Features of the gravimetric system comprise three main components: a pair of switching valves that divert the working liquid between two symmetrical flow paths; a weighing vessel equipped with an overflow inner vessel and enclosed in a weighing chamber; and a liquid discharge mechanism comprising a discharge tube and a discharge pump, used with a multi-purpose bin. These are described with an explanation of the design considerations behind each feature. The overflow inner vessel is designed with a notch in its wall and is positioned so that it does not come into contact with the liquid surface of the accumulated liquid in the weighing vessel or the side wall of the weighing vessel to obtain a good repeatability of the interactive effects between the feeding tube and the submerging working liquid, thus ensuring a correct mass reading of the liquid collection. A performance test showed that, in terms of contribution to the overall uncertainty of the standard flow rate, the pair of switching valves is capable of performing SW-FSS satisfactorily with small relative timing errors within %. However, the mass loss due to evaporation is considered a major source of error of the gravimetric system, showing a maximum error of 0.011% under the most evaporative condition tested for the longest liquid collection time of the gravimetric system.

Quantification of the Impact of Photon Distinguishability on Measurement-Device- Independent Quantum Key Distribution

Measurement-Device-Independent Quantum Key Distribution (MDI-QKD) is a two-photon protocol devised to eliminate eavesdropping attacks that interrogate or control the detector in realized quantum key distribution systems. In MDI-QKD, the measurements are carried out by an untrusted third party, and the measurement results are announced openly. Knowledge or control of the measurement results gives the third party no information about the secret key. Error-free implementation of the MDI-QKD protocol requires the crypto-communicating parties, Alice and Bob, to independently prepare and transmit single photons that are physically indistinguishable, with the possible exception of their polarization states. In this paper, we apply the formalism of quantum optics and Monte Carlo simulations to quantify the impact of small errors in wavelength, bandwidth, polarization and timing between Alice’s photons and Bob’s photons on the MDI-QKD quantum bit error rate (QBER). Using published single-photon source characteristics from two-photon interference experiments as a test case, our simulations predict that the finite tolerances of these sources contribute ( 4.04 ± 20 / N sifted ) % to the QBER in an MDI-QKD implementation generating an N sifted -bit sifted key.

Groups clapping in unison undergo size-dependent error-induced frequency increase

Humans clapping together in unison is a familiar and robust example of emergent synchrony. We find that in experiments, such groups (from two to a few hundred) always increase clapping frequency, and larger groups increase more quickly. Based on single-person experiments and modeling, an individual tendency to rush is ruled out as an explanation. Instead, an asymmetric sensitivity in aural interactions explains the frequency increase, whereby individuals correct more strongly to match neighbour claps that precede their own clap, than those that follow it. A simple conceptual coupled oscillator model based on this interaction recovers the main features observed in experiments, and shows that the collective frequency increase is driven by the small timing errors in individuals, and the resulting inter-individual interactions that occur to maintain unison.

Evaluation of NCAR’s AutoNowCaster for Operational Application within the National Weather Service

Abstract NCAR’s AutoNowCaster (ANC) was modified to run over a large domain that encompasses the air traffic management hubs of Chicago, Illinois; New York City, New York; and Atlanta, Georgia. ANC produces nowcasts of convective likelihood (CL), with higher values delineating areas where storms are likely to form and be sustained, and vice versa. This paper presents the results of verifying ANC’s 60-min nowcasts of CL over the study area using data collected from 11 June to 30 September 2012. To reduce the high sensitivity of statistical scores to small errors in location and timing, spatial and temporal relaxation techniques were explored. The results show that, at a spatial scale of roughly 50 km and with no temporal relaxation, a CL value of 0.6 is an optimum threshold for nowcasting the general areas both where new storms may initiate and where existing storms will be sustained. Moreover, at that same spatial scale and with temporal relaxation (45–90 min from the nowcast issuance time), a CL value of 0.7 is a good threshold for nowcasting convective initiation alone.

Performance Evaluation of a Same Directional Flow Diverter Based on Transforming Relationship Modeling of Diverting Flow Error and Timing Error

It is helpful to improve the accuracy of the flying-start-stop liquid flow calibration facilities used to calibrate different types of flowmeters by studying and decreasing the flow diverter timing error. In this paper, a transforming relationship model of diverting flow error and timing error is proposed to reveal the principle of the timing error. Meanwhile, a same directional diverter based on the transforming relationship model is also proposed, which has the following advantages: a small timing error, a high repeatability accuracy, no need to adjust the trigger position, easy to implement ISO4185 calibration test, a wide range of working flow rates, easy to be manufactured, and a good portability. To verify the advantageous characteristics of the same directional diverter, the performance comparing experiments of the same directional diverter and the traditional diverter under different flow rates, trigger positions, and diverter moving velocities are proposed. The results show that the timing error of the same directional diverter has no relationship with trigger position and has low sensitivity in both flow rate and the diverter moving velocity. While the diverter moving velocity is 0.1 m/s, the absolute value of the timing error of the same directional diverter just varies from 0.251 to 1.473 ms, which is much smaller than the traditional diverter’s (varies from 9.940 to 532.302 ms). The uncertainty caused by the timing error and the total uncertainty of measuring time of the same directional diverter vary from 0.000454% to 0.000883% and 0.000529% to 0.000924%, respectively.

Muscle activation characteristics of the front leg during baseball swings with timing correction for sudden velocity decrease.

This study aimed to clarify the activation characteristics of the vastus lateralis muscle in the front leg during timing correction for a sudden decrease in the velocity of a target during baseball swings. Eleven male collegiate baseball players performed coincident timing tasks that comprised constant velocity of 8 m/s (unchanged) and a sudden decrease in velocity from 8 to 4 m/s (decreased velocity). Electromyography (EMG) revealed that the muscle activation was typically monophasic when responding unchanged conditions. The type of muscle activation during swings in response to decreased velocity condition was both monophasic and biphasic. When biphasic activation appeared in response to decreased velocity, the impact time and the time to peak EMG amplitude were significantly prolonged and the timing error was significantly smaller than that of monophasic activation. However, the EMG onset from the target start was consistent both monophasic and biphasic activation in response to conditions of decreased velocity. In addition, batters with small timing errors in response to decreased velocity were more likely to generate biphasic EMG activation. These findings indicated that timing correction for a sudden decrease in the velocity of an oncoming target is achieved by modifying the muscle activation characteristics of the vastus lateralis muscle of front leg from monophasic to biphasic to delay reaching peak muscle activation and thus prolong impact time. Therefore, the present findings suggests that the extent of timing errors in response to decreased velocity is influenced by the ability to correct muscle activation after its initiation rather than by delaying the initiation timing of muscle activation during baseball swings.

On the Dynamics, Thermodynamics, and Forecast Model Evaluation of Two Snow-Burst Events in Southern Alberta

Abstract Two high-impact convective snowband events (“snow bursts”) that affected Calgary, Alberta, Canada, are examined to better understand the dynamics and thermodynamics of heavy snowbands not associated with lake effects or the cold conveyor belt of synoptic-scale cyclones. Such events are typically characterized by brief, but heavy, periods of snow; low visibilities; and substantial hazards to automobile and aviation interests. Previous literature on these events has been limited to a few case studies across North America, including near the leeside foothills of the U.S. Rockies. The large-scale dynamics and thermodynamics are investigated using the National Centers for Environmental Prediction (NCEP) North American Regional Reanalysis (NARR). Previously, high-resolution convection-explicit Weather Research and Forecasting Model (WRF) simulations have shown some ability to successfully reproduce the dynamics, thermodynamics, and appearance of convective snowbands, with small errors in location and timing. Therefore, WRF simulations are performed for both events, and are evaluated along with the NCEP North American Mesoscale (NAM) model forecasts. Both the NARR and WRF simulations show that while the two snow bursts are similar in appearance, they form as a result of different dynamic and thermodynamic mechanisms. The first event occurs downstream of an upper-tropospheric jet streak in a region of little synoptic-scale ascent, where ageostrophic frontogenesis helps to release conditional, dry symmetric, and inertial instability in an unsaturated environment. The inertial instability is determined to be related to fast flow over upstream high terrain. The second event occurs in a saturated environment in a region of Q-vector convergence (primarily geostrophic frontogenesis), which acts to release conditional, convective, and conditional symmetric instability (CSI).

FDG kinetic modeling in small rodent brain PET: optimization of data acquisition and analysis

BackgroundKinetic modeling of brain glucose metabolism in small rodents from positron emission tomography (PET) data using 2-deoxy-2-[18 F]fluoro-d-glucose (FDG) has been highly inconsistent, due to different modeling parameter settings and underestimation of the impact of methodological flaws in experimentation. This article aims to contribute toward improved experimental standards. As solutions for arterial input function (IF) acquisition of satisfactory quality are becoming available for small rodents, reliable two-tissue compartment modeling and the determination of transport and phosphorylation rate constants of FDG in rodent brain are within reach.MethodsData from mouse brain FDG PET with IFs determined with a coincidence counter on an arterio-venous shunt were analyzed with the two-tissue compartment model. We assessed the influence of several factors on the modeling results: the value for the fractional blood volume in tissue, precision of timing and calibration, smoothing of data, correction for blood cell uptake of FDG, and protocol for FDG administration. Kinetic modeling with experimental and simulated data was performed under systematic variation of these parameters.ResultsBlood volume fitting was unreliable and affected the estimation of rate constants. Even small sample timing errors of a few seconds lead to significant deviations of the fit parameters. Data smoothing did not increase model fit precision. Accurate correction for the kinetics of blood cell uptake of FDG rather than constant scaling of the blood time-activity curve is mandatory for kinetic modeling. FDG infusion over 4 to 5 min instead of bolus injection revealed well-defined experimental input functions and allowed for longer blood sampling intervals at similar fit precisions in simulations.ConclusionsFDG infusion over a few minutes instead of bolus injection allows for longer blood sampling intervals in kinetic modeling with the two-tissue compartment model at a similar precision of fit parameters. The fractional blood volume in the tissue of interest should be entered as a fixed value and kinetics of blood cell uptake of FDG should be included in the model. Data smoothing does not improve the results, and timing errors should be avoided by precise temporal matching of blood and tissue time-activity curves and by replacing manual with automated blood sampling.

Fast-ball sports experts depend on an inhibitory strategy to reprogram their movement timing

The purpose of our study was to clarify whether an inhibitory strategy is used for reprogramming of movement timing by experts in fast-ball sports when they correct their movement timing due to unexpected environmental changes. We evaluated the influence of disruption of inhibitory function of the right inferior frontal gyrus (rIFG) on reprogramming of movement timing of experts and non-experts in fast-ball sports. The task was to manually press a button to coincide with the arrival of a moving target. The target moved at a constant velocity, and its velocity was suddenly either increased or decreased in some trials. The task was performed either with or without transcranial magnetic stimulation (TMS), which was delivered to the region of the rIFG. Under velocity change conditions without TMS, the experts showed significantly smaller timing errors and a higher rate of reprogramming of movement timing than the non-experts. Moreover, TMS application during the task significantly diminished the expert group's performance, but not the control group, particularly in the condition where the target velocity decreases. These results suggest that experts use an inhibitory strategy for reprogramming of movement timing. In addition, the rIFG inhibitory function contributes to the superior movement correction of experts in fast-ball sports.

Assessment of validity of a high-yield surface electromyogram decomposition

BackgroundUsing recordings from a five-pin surface sensor array, a template-based surface electromyogram (sEMG) decomposition system has been developed to identify single motor unit discharge properties. However, the reliability of such template based decomposition results has not been thoroughly examined except by the developers. The focus of this study was to assess the validity of the motor unit decomposition technique, using EMG recordings from the first dorsal interosseous muscle of able-bodied human subjects.MethodsTwo tests were utilized. In the first test, a spike triggered averaging (STA) analysis was used to derive motor unit action potential (MUAP) parameters. We examined these STA derived MUAP shapes after firing times were perturbed by added timing noise. In the second test, a cross-correlation analysis was performed between the sEMG signal and MUAP trains constructed using STA estimates and their firing times.ResultsIn the first test, we found that MUAP shape features deteriorated significantly when rather small (0.6-2 ms) timing errors were added, affirming that the decomposed firing times are presumptively valid. The results of the second test reveal that the cross-correlation index between the EMG and MUAP trains increased monotonically up to 0.71 when the identified MUs were progressively added to reconstructed MUAP trains; however, this increment disappeared when the firing times or the MUAP templates were shifted randomly.ConclusionsBased on our STA selection criteria, our results suggest that the firing times and estimated MUAP shapes for each MU generated by the decomposition algorithms are presumptively valid.

Convective initiation on 19 June 2002 during IHOP: High‐resolution simulations and analysis of the mesoscale structures and convection initiation

The 19 June 2002 convective initiation (CI) case from the International H2O Project (IHOP) is simulated using the ARPS at 1 km grid spacing. It involves three distinct CI groups, CI‐A, CI‐B, and CI‐C, associated with a cold front‐dryline system. Initial condition at 1800 UTC, 19 June 2002 was created using the ARPS 3DVAR, including standard as well as special observations. The simulation captured the three groups of CIs rather well, with small timing and spatial errors. CI‐A is most typical of dryline initiation. Vertical cross sections through the initiation locations show typical dryline structures with an upward moisture bulge forced by lifting along the dryline convergence. For CI‐B, the model‐predicted mesoscale structures agree with special IHOP aircraft dropsonde observations closely. Strong low‐level convergence from opposing cold front and dryline circulations produces deep symmetric upwelling of moist air, and the convection is initiated at the crest of the moisture bulge. CI‐C is more typical of frontal convection where deep convection is first initiated at the front edge of the cold air. For CI‐B and CI‐C, north‐south moisture bands that have dryline characteristics (and can be considered multiple drylines) and contain enhanced vertical motion and upward moisture bulges intercept the dryline and cold front, respectively, providing the most favored locations on the dryline or cold front for initiation. Furthermore, the air parcels feeding the initial cells at CI‐B and CI‐C have trajectories along the moisture bands and have already experienced uplifting when they reach the dryline and cold front locations.

An Improved Data-Aided Symbol Timing Algorithm for OFDM Systems

Synchronization is very important to orthogonal frequency division multiplexing (OFDM) systems. Taking aim at the problem that the timing inaccuracy of the S&C method and the indistinct peak of the timing metric produced by Park method under the low signal-to-noise ratio (SNR) and fading channel, a new symbol synchronization algorithm based on the training symbol is proposed in this paper. The new algorithm exploits the training symbol proposed by S&C to correlate the received signal with the known training symbol again at the receiving end to identify the start point of the OFDM symbol. The result of the simulation shows that the proposed method’s timing metric is of the impulse-shaped peak under the condition of the lower SNR and multipath fading channel, and has smaller timing errors compared with the conventional method in terms of mean-square error (MSE) of the timing estimator.

Jitter Noise of Sampled Multitone Signals

Clock jitter degrades the performance of data converters in mixed-signal systems. The sampling operation inherent to analog-to-digital and digital-to-analog (D/A) conversion is inevitably subject to random timing uncertainties. These small timing errors introduce a noise component into the system that degrades the effective resolution of the data converters. Although the effect of sampling jitter on single sine waves is well understood, this classical result cannot be generalized to multitone signals, which are nowadays widely used as information carriers in digital communication systems. In this brief, an approximation for the phase noise of integrated clock sources is used to derive an analytical expression for the jitter noise power spectral density (PSD) of unmodulated multitone signals at the output of an ideal sampler. This result requires only a minimum number of basic system parameters for a quick and accurate estimation of the needed sampling clock quality for data converters used in multicarrier processing. In addition to behavioral simulation, the obtained expression is verified experimentally using a CMOS D/A converter with intentionally jittered sampling clock.

Radio Frequency Time-of-Flight Distance Measurement for Low-Cost Wireless Sensor Localization

Location-aware wireless sensor networks will enable a new class of applications, and accurate range estimation is critical for this task. Low-cost location determination capability is studied almost entirely using radio frequency received signal strength (RSS) measurements, resulting in poor accuracy. More accurate systems use wide bandwidths and/or complex time-synchronized infrastructure. Low-cost, accurate ranging has proven difficult because small timing errors result in large range errors. This paper addresses estimation of the distance between wireless nodes using a two-way ranging technique that approaches the Cramér-Rao Bound on ranging accuracy in white noise and achieves 1-3 m accuracy in real-world ranging and localization experiments. This work provides an alternative to inaccurate RSS and complex, wide-bandwidth methods. Measured results using a prototype wireless system confirm performance in the real world.

Improved Nyquist filter with an ideal piece-wise rectangular characteristic

The present approach is centered on the optimization of a piece-wise rectangular transfer characteristic in order to obtain a minimal value of the error probability when the impulse response is sampled with small timing errors. We proposed a new parametric method for the design of the Nyquist filter using constraints in the frequency characteristic construction. We studied the family of new Nyquist filters based on a composite frequency characteristic and investigated its performances in terms of ISI error probability. This method of construction allows us to derive improved Nyquist filters in terms of ISI error probability. We also observed the variation of the side lobes of the pulses when a larger number of parameters is introduced in the proposed filter frequency characteristic. An implementation example of a real filter with improved performance in terms of ISI error probability when sampled with small timing errors is presented.

A Novel Pulse Designed to Jointly Optimize Symbol Timing Estimation Performance and the Mean Squared Error of Recovered Data

A new optimal pulse is derived considering both the symbol timing estimation performance and the mean squared error of intersymbol interference corrupted data recovered in timing error. The new pulse design minimizes the mean squared error of the recovered data averaged over the timing error, accounting for the fact that the distribution of the timing error is also affected by the pulse shape. The new pulse resembles the Franks' double-jump pulse, but has different slope than the Franks' double-jump pulse in the spectral rolloff region. Although the new pulse has greater mean squared error than the Franks' pulse for a fixed small timing error, it has better symbol timing estimation performance and better overall performance as measured by the Cramer Rao lower bound and by the mean square error of the recovered data accounting for both timing recovery effects and data recovery distortion, respectively. An optimal pulse that minimizes the mean squared error of intersymbol interference corrupted data recovered in the presence of fixed but arbitrary timing error is derived as an intermediate result.

Performance of Multicode DS/CDMA With Noncoherent $M$ -ary Orthogonal Modulation in the Presence of Timing Errors

This paper derives an accurate approximation to the bit error rate (BER) of multicode direct-sequence code-division multiple access (DS/CDMA) with noncoherent <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">M</i> -ary modulation in wideband fading channels when timing errors are made at the receiver employing equal-gain combining (EGC). This reflects the practical scenario where the path delays are estimated imperfectly, leading to synchronization errors between the correlation receivers and the received signals. The analysis can be applied to any type of a fading distribution for both independent and correlated diversity branches. It is shown that the derived analytical expressions are in close agreement with the Monte Carlo system simulations, particularly in the case of small timing errors.

Optimisation of high bar circling technique for consistent performance of a triple piked somersault dismount

The dismount from the high bar is one of the most spectacular skills performed in Men's Artistic Gymnastics. Hiley and Yeadon [2005. Maximal dismounts from high bar. Journal of Biomechanics 38, 2221–2227] optimised the technique in the backward giant circle prior to release using a computer simulation model to show that a gymnast could generate sufficient linear and angular momentum to perform a triple piked backward somersault dismount with a sufficiently large release window (the period of time during which the gymnast could release the bar and successfully complete the dismount).In the present study, it was found that when the timing of the actions at the hip and shoulder joints from the optimum simulation were perturbed by 30ms the resulting simulation could no longer meet the criteria for sufficient aerial rotation and release window. Since it is to be expected that a gymnast's technique can cope with small errors in timing for consistent performance, a requirement of robustness to timing perturbations should be included within the optimisation process. When the technique in the backward giant circle was optimised to be robust to 30ms perturbations, it was found that sufficient linear and angular momentum for a triple piked dismount could be achieved with a realistic release window.