Signal-to-noise Ratio Metrics Research Articles

Ultra-high resolution coronary CT angiography on photon-counting detector CT: bi-centre study on the impact of quantum iterative reconstruction on image quality and accuracy of stenosis measurements

PurposeTo assess the impact of different quantum iterative reconstruction (QIR) levels on objective and subjective image quality of ultra-high resolution (UHR) coronary CT angiography (CCTA) images and to determine the effect of strength levels on stenosis quantification using photon-counting detector (PCD)-CT. MethodA dynamic vessel phantom containing two calcified lesions (25 % and 50 % stenosis) was scanned at heart rates of 60, 80 and 100 beats per minute with a PCD-CT system. In vivo CCTA examinations were performed in 102 patients. All scans were acquired in UHR mode (slice thickness0.2 mm) and reconstructed with four different QIR levels (1–4) using a sharp vascular kernel (Bv64). Image noise, signal-to-noise ratio (SNR), sharpness, and percent diameter stenosis (PDS) were quantified in the phantom, while noise, SNR, contrast-to-noise ratio (CNR), sharpness, and subjective quality metrics (noise, sharpness, overall image quality) were assessed in patient scans. ResultsIncreasing QIR levels resulted in significantly lower objective image noise (in vitro and in vivo: both p < 0.001), higher SNR (both p < 0.001) and CNR (both p < 0.001). Sharpness and PDS values did not differ significantly among QIRs (all pairwise p > 0.008). Subjective noise of in vivo images significantly decreased with increasing QIR levels, resulting in significantly higher image quality scores at increasing QIR levels (all pairwise p < 0.001). Qualitative sharpness, on the other hand, did not differ across different levels of QIR (p = 0.15). ConclusionsThe QIR algorithm may enhance the image quality of CCTA datasets without compromising image sharpness or accurate stenosis measurements, with the most prominent benefits at the highest strength level.

Physical Modeling and Compensation for Systematic Negative Errors in GNSS-R Snow Depth Retrieval

Previous studies have reported that signal penetration will introduce an underestimation of snow depth, referred to as the snow depth difference. So far, however, there have been few detailed investigations into the relationship between snow depth difference and signal-to-noise ratio (SNR) metrics. In this study, we briefly describe the snow depth difference and provide a physical explanation of the systematic negative error. The baseline- and short-term variations of snow depth difference and SNR metrics were identified, and their relationships during various snow periods were investigated. The results indicated that the systematic negative errors and SNR metrics during the stable and melting periods are dominated by the layered structures and liquid water content of snowfall, respectively. Meanwhile, compared with the baseline terms, the short-term variations of snow depth difference and SNR metrics were more sensitive to fresh, low-density snowfall over the old snow surface. Additionally, an improved method is proposed to compensate for systematic differences using 2- and 5-parameter multiple linear regression (MLR) models with SNR metrics as independent variables. The results showed that the compensation values conformed with the measured values with correlation coefficients exceeding 0.85. In terms of accuracy, once the MLR models were applied, the root mean squared errors (RMSEs) decreased from 22.05 cm to 3.89 cm and 3.40 cm, respectively. Moreover, the corrected estimates agree well with the meteorological records, with regression slope deviations of less than 2% and correlation coefficients of over 0.97, suggesting no systematic errors between the estimates and reference.

Atrial fibrillation cardiac radioablation target visibility on magnetic resonance imaging

Magnetic resonance imaging (MRI) guided cardiac radioablation (CR) for atrial fibrillation (AF) is a promising treatment concept. However, the visibility of AF CR targets on MRI acquisitions requires further exploration and MRI sequence and parameter optimization has not yet been performed for this application. This pilot study explores the feasibility of MRI-guided tracking of AF CR targets by evaluating AF CR target visualization on human participants using a selection of 3D and 2D MRI sequences.MRI datasets were acquired in healthy and AF participants using a range of MRI sequences and parameters. MRI acquisition categories included 3D free-breathing acquisitions (3Dacq), 2D breath-hold ECG-gated acquisitions (2DECG-gated), stacks of 2D breath-hold ECG-gated acquisitions which were retrospectively interpolated to 3D datasets (3Dinterp), and 2D breath-hold ungated acquisitions (2Dreal-time). The ease of target delineation and the presence of artifacts were qualitatively analyzed. Image quality was quantitatively analyzed using signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR) and non-uniformity. Confident 3D target delineation was achievable on all 3Dinterp datasets but was not possible on any of the 3Dacq datasets. Fewer artifacts and significantly better SNR, CNR and non-uniformity metrics were observed with 3Dinterp compared to 3Dacq. 2Dreal-time datasets had slightly lower SNR and CNR than 2DECG-gated and 3Dinterp n datasets. AF CR target visualization on MRI was qualitatively and quantitatively evaluated. The study findings indicate that AF CR target visualization is achievable despite the imaging challenges associated with these targets, warranting further investigation into MRI-guided AF CR treatments.

Retrieving the SNR metrics with different antenna configurations for GNSS-IR

Multipath, which is a major source of error for precise positioning, is the effect that occurs when Global Navigation Satellite Systems (GNSS) signals reach the receiver by reflecting from one or more surfaces. Reflected signals affect the signal-to-noise ratio (SNR) data provided by the receiver, indicating the signal strength. The structure of the antenna of the receiver and the direction in which it is oriented also change the strength of the received signal. In this study, the effect of antenna orientation and polarization on SNR data was demonstrated by using the method called GNSS-Interferometric Reflectometry (GNSS-IR), in terms of reflector height estimates. A geodetic GNSS receiver (CHC i50) and two different smartphones (Xiaomi Mi8 and Xiaomi Mi8 Lite) were used in the four-day experiments. The geodetic receiver was established as zenith-looking (ZL) in the first two days and as horizon-looking (HL) in the last two days. Smartphones were placed on the same mast with the HL receiver in the last two days. It was seen that it is more appropriate to use a 0°-60° satellite elevation angle range in the common use of all receivers’ data. In the 30°-60° range where the ZL installation receives the multipath signals weakly, it has been found that the HL receiver and smartphones have reflector height estimation accuracies with values ranging from 1.9 cm to 2.5 cm. In short, for different elevation angle ranges, accuracies below 2 cm could be obtained with each receiver. Thus, different antenna configurations may be used in GNSS-IR studies, depending on the characteristics of the study area and the surface feature to be determined.

Quantification of DAS VSP Quality: SNR vs. Log-Based Metrics.

The initial quantification of data quality is an important step in seismic data acquisition design, including the choice of sensing strategy. The signal-to-noise ratio (SNR) often drives the choice of distributed acoustic sensing (DAS) parameters in vertical seismic profiling (VSP). We compare this established approach for data quality assessment with metrics comparing DAS data products to available well logs. First, we create kinematic and dynamic data products derived from original seismic data, such as the interval velocity and amplitude of P-wave arrivals. Next, we quantify the quality of derived data products using well log data by calculating various statistical metrics. Using a large dataset of 220 different VSP experiments with a fixed source location and various DAS acquisition parameters, such as gauge length (GL), conveyance type, and lead-in length, we analyzed the statistical distribution of various metrics. The results indicate the decoupling between seismic-based and log-based metrics as well as between the quality of dynamic and kinematic data-products for the same record. Therefore, we propose using fit-for-purpose metrics to optimize the acquisition cost. In particular, for ray-based tomographic processing, it is sufficient to use traveltime-based metrics. On the other hand, for advanced dynamic analysis, amplitude-based metrics define the quality of final processing products. Hence, it is crucial to use fit-for-purpose metrics to optimize DAS VSP acquisition.

Acceptable Noise Levels and Preferred Signal-to-noise Ratios for Speech and Music.

The objective of this study was to evaluate the influence of music versus speech on a listener's signal-to-noise ratio (SNR) preferences when listening in a noise background. SNR preferences were quantified using acceptable noise level (ANL) and preferred SNR metrics. The measurement paradigm for ANL allows the listener to adjust the level of background noise while listening to the target at their most comfortable loudness level. A higher ANL indicates less tolerance for noise and a lower ANL indicates high tolerance for noise. The preferred SNR is simply the SNR the listener prefers when attending to a target in a fixed-amount (level) of background noise. In contrast to the ANL, the listener does not have control over the noise. Rather, they are only able to manipulate the target level. The first aim of the study was to determine if listeners' tolerances for noise, quantified using the ANL, when listening to music is different from that when listening to speech. The second aim of the study was to determine if listeners' tolerances for noise, quantified using their preferred SNR, when listening to music is different from that when listening to speech. The third aim of the study was to quantify the relationship between ANL and preferred SNR. Ninety-nine normal-hearing, native-English speakers participated in this study. The ANL and preferred SNR were measured for speech and music targets. Music targets included two variations (with lyrics and without lyrics) of the song "Rocky Top." Measurements were made in the sound field at 0° azimuth, 1.5 m from a loud speaker. For both ANL and preferred SNR, targets were presented in 12-talker babble noise. The level of the noise was adjusted by the listener during ANL measurement but was fixed in level during the preferred SNR measurement (75 dB A). Repeated-measures analysis of variance was performed to identify any significant effect of target on the ANL and preferred SNR. Correlation analysis was performed to evaluate the relationship between ANL and preferred SNR. Findings demonstrate a significant effect of target on ANL and preferred SNR. ANLs were highest for speech (mean = 7.2 dB), followed by music with lyrics (6.1 dB), and music without lyrics (2.5 dB). Preferred SNRs were highest for music with lyrics (mean = 2.3 dB), followed by speech (1.2 dB), and music without lyrics (-0.1 dB). A listener's ANL for a given target was strongly correlated with their ANL for a different target (the same was true for preferred SNR); however, ANL for a given target was not a statistically significant predictor of preferred SNR for the same target. When listening in a background of noise, the listener's tolerance for noise depends on the target to which they are attending, whether music or speech. This dependence is especially evident for ANL measures, and less so for preferred SNR measures. Despite differences in ANL and preferred SNR across targets, a listener's ANL and preferred SNR for one target predicts their ANL and preferred SNR, respectively, for a different target. The lack of correlation between ANL and preferred SNR suggests different mechanisms underly these listener-preference metrics.

Spatial resolution as a trade-space for low-light imaging of sensitive cultural heritage documents

Abstract A 14th-century manuscript was imaged at high image quality with an HSI detector, using a low-illumination dose of 57 lx h, significantly less than one day’s worth of museum display (≈ 400 lx h). Initial work digitally downsampled spatial resolution from 333 pixels per inch (ppi) to 161 and 80 ppi, and increased the signal-to-noise ratio (SNR) by a factor of 2.00 and 2.79. Despite downsampling, pan-sharpened images were up to 1.5 times spatially sharper than the reference HSI image. Relatively high spectral accuracy was maintained, with spectral angle mapper (SAM) measurements of 0.0527–0.0963. The process was then repeated with physical adjustments to imaging height replacing the digital downsampling process, decreasing resolution from 403 ppi to 111 ppi, increasing SNR from 27.1 to 95.4, and sharpness by 60% using pan-sharpening, while maintaining SAM values of 0.110. An overall image quality metric is proposed, composed of fundamental spatial, spectral, and SNR metrics. This metric highlights efficient lighting and processing methods for increased image quality while limiting total illumination dose on the target, and can be tailored based on one’s imaging constraints.

Robust three-dimensional registration on optical coherence tomography angiography for speckle reduction and visualization.

In the clinical applications of optical coherence tomography angiography (OCTA), the repeated scanning and averaging method can provide better contrast with reduced speckle noises in the final results, which are useful for visualizing and quantifying vascular components with high accuracy, reproducibility, and reliability. However, the inevitable patient motion presents a challenge to this method. The objective of this study is to meet this challenge by introducing a 3D registration method to register optical coherence tomography (OCT)/OCTA scans for precise volume averaging of multiple scans to improve the signal-to-noise ratio (SNR) and increase quantification accuracy. The proposed method utilized both rigid affine transformation and non-rigid B-spline transformation in which their parameters were optimized and calculated by the average stochastic gradient descent on OCT structural images. In addition, we also introduced a multi-level resolution approach to further improve the robustness and computational speed of our proposed method. The imaging performance was tested on in vivo imaging of human skin and eye and assessed by SNR, peak signal-to-noise ratio (PSNR) and normalized correlation coefficient (NCC). Five subjects were enrolled in this study for obtaining in vivo images of skin and retina. The proposed registration and averaging method provided substantial improvements of the imaging performance in terms of vessel connectivity and signal to noise ratio. The increase of repeated volume numbers in the averaging improves all the metrics assessed, i.e., SNR, PSNR and NCC. An improvement of the SNR from 10 to 40 dB after 10 repeated volumetric averaging was achieved. The proposed 3D registration and averaging method is effective in reducing speckle noises and suppressing motion artifacts, thereby improving SNR, PSNR and NCC metrics for final averaged images. It is expected that the proposed algorithm would be practically useful in better visualization and more reliable quantification of in vivo OCT and OCTA data, which would be beneficial to OCT clinical applications.

Multi-Hop Relay Selection Based on Fade Durations

In cooperative relaying, the selection of relays could be based on different parameters. The most well-known and frequently used metric is the signal-to-noise ratio (SNR). In this method of relay selection, the rapid fluctuation of the signal (i.e., fading) is not taken into account in the selection criteria. Such rapid signal change may cause significant loss of information, degrade signal quality for voice or video connections, or could make the channel coding fail. An alternative method of relay selection in a cooperative relay network is by considering fading. Such methods include average fade duration (AFD) and fade duration outage probability (FDOP), which are based on time correlation statistics. Both the AFD and the FDOP are computed in reference to a threshold value for signal quality. This work derives new formulas for two hop and three hop relay paths, with three hop paths given a penalty cost. Then optimization algorithms for each type of relay selection method are derived, including total path and link-by-link optimization. Simulation results provide optimal AFD and FDOP paths for various random network topologies. These paths are then compared to paths that would be found if SNR metrics were used instead. It is shown that SNR optimization results in much different performance. For cases of four sources and four relays, SNR based optimization frequently chose different relay paths, as low as only 63% of the same relay paths as FDOP or AFD optimizations. Because fade duration methods more accurately control the fading nature and true quality of the signals, the results here provide significant improvements in relay performance and allow two and three hop relay paths to be implemented effectively.

Using BDS SNR Observations to Measure Near-Surface Soil Moisture Fluctuations: Results From Low Vegetated Surface

The feasibility of using China’s Beidou Navigation Satellite System (BDS) signal-to-noise ratio (SNR) data for estimating soil moisture is demonstrated in this letter. Previous studies for the Global Navigation Satellite System (GNSS) multipath for estimating soil moisture have concentrated on the Global Positioning System (GPS) SNR data. The SNR data recorded by BDS receivers can also be strongly impacted by the relative permittivity, which however, has not been comprehensively investigated. This letter used a commercially available geodetic-quality BDS/GPS receiver to collect BDS SNR data on both B1 and B2 frequencies. Two methods were investigated to demonstrate the relationship between BDS SNR metrics and soil moisture. One was the phase method proposed by Larson et al. , and the other was an interference model proposed by Peng et al. The results show that both B1 and B2 frequencies perform well to reflect the moisture fluctuations, particularly for reflecting big precipitation events. Comparisons between BDS B1 and B2 with GPS L2C and L5 demonstrate the results from these two constellations are comparable. BDS could be a new data source for producing global high-temporal soil moisture products using GNSS-based approaches.

Parametric Comparison of Modeled and Measured Heat-Assisted Magnetic Recording Using a Common Signal-to-Noise Metric

The auto-correlation signal-to-noise (ACSN) method is employed on micromagnetically modeled heat-assisted magnetic recording (HAMR) readback waveforms and compared with experimental ACSN data. The breakdown between transition signal-to-noise ratio (SNR) and remanence SNR is provided for both modeled and measured data. Good agreement is observed between both simulated and experimental SNRs as a function of laser power and as a function of linear density. An efficiency factor that relates maximum temperature observed in the media to the applied laser current is obtained by combining the modeled and experimental data. Usage of this methodology provides SNR metrics that allow a reliable comparison of micromagnetic modeling of HAMR recording systems to experimental measurements.

SNR Maximization Hashing

We propose a novel robust hashing algorithm based on signal-to-noise ratio (SNR) maximization to learn compact binary codes, where the SNR metric is used to select a set of projection directions, and one hash bit is extracted from each projection direction. We first motivate this approach under a Gaussian model for the underlying signals, in which case maximizing SNR is equivalent to minimizing the robust hashing error probability. A globally optimal solution can be obtained by solving a generalized eigenvalue problem. We also develop a multibit per projection algorithm to learn longer hash codes when the number of high-SNR projections is limited. The proposed algorithms are tested on both synthetic and real data sets, showing significant performance gains over existing hashing algorithms.

Vegetation Sensing Using GPS-Interferometric Reflectometry: Theoretical Effects of Canopy Parameters on Signal-to-Noise Ratio Data

The potential to use GPS signal-to-noise ratio (SNR) data to estimate changes in vegetation surrounding a ground-based antenna is evaluated. A 1-D plane-stratified model that simulates the response of GPS SNR data to changes in both soil moisture and vegetation is presented. The model is validated against observations of SNR data from four field sites with varying vegetation cover. Validation shows that the average correlation between modeled and observed SNR data is higher than the average correlation between concurrent SNR observations from different satellite tracks at a site. The model also reproduces variations in the SNR metrics amplitude, phase, and effective reflector height over a range of vegetation wet weights from 0 to 4 $\hbox{kg}\cdot\hbox{m}^{-2}$ , with $r^{2}$ values of 0.79, 0.84, and 0.62, respectively. Model simulations indicate that the amplitude of SNR oscillations may be used to estimate vegetation amount when vegetation wet weight is below 1.5 $\hbox{kg}\cdot\hbox{m}^{-2}$ . When vegetation wet weight exceeds 1.5 $\hbox{kg}\cdot\hbox{m}^{-2}$ , the sensitivity of amplitude to changes in vegetation amount decreases. Phase of SNR oscillations also varies consistently with vegetation up to 1.5 $\hbox{kg}\cdot\hbox{m}^{-2}$ . However, phase is also very sensitive to soil moisture variations, thus limiting its utility for estimating vegetation. Effective reflector height is not a consistent indicator of vegetation change. Beyond 1.5 $\hbox{kg}\cdot\hbox{m}^{-2}$ , the constant frequency assumption used to characterize SNR fluctuations does not adequately describe observed data. A more complex approach than the standard SNR metrics used here is required to extend GPS-Interferometric Reflectometry sensing to thicker canopies.

Particle image velocimetry correlation signal-to-noise ratio metrics and measurement uncertainty quantification

In particle image velocimetry (PIV) the measurement signal is contained in the recorded intensity of the particle image pattern superimposed on a variety of noise sources. The signal-to-noise-ratio (SNR) strength governs the resulting PIV cross correlation and ultimately the accuracy and uncertainty of the resulting PIV measurement. Hence we posit that correlation SNR metrics calculated from the correlation plane can be used to quantify the quality of the correlation and the resulting uncertainty of an individual measurement. In this paper we extend the original work by Charonko and Vlachos and present a framework for evaluating the correlation SNR using a set of different metrics, which in turn are used to develop models for uncertainty estimation. Several corrections have been applied in this work. The SNR metrics and corresponding models presented herein are expanded to be applicable to both standard and filtered correlations by applying a subtraction of the minimum correlation value to remove the effect of the background image noise. In addition, the notion of a ‘valid’ measurement is redefined with respect to the correlation peak width in order to be consistent with uncertainty quantification principles and distinct from an ‘outlier’ measurement. Finally the type and significance of the error distribution function is investigated. These advancements lead to more robust and reliable uncertainty estimation models compared with the original work by Charonko and Vlachos. The models are tested against both synthetic benchmark data as well as experimental measurements. In this work, uncertainties are estimated at the 68.5% confidence level while uncertainties are estimated at 95% confidence level. For all cases the resulting calculated coverage factors approximate the expected theoretical confidence intervals, thus demonstrating the applicability of these new models for estimation of uncertainty for individual PIV measurements.

A computational model of the stellate cell microcircuit in the auditory brainstem

The cochlear nucleus (CN) is the first processing center of auditory signals from the cochlea. Several distinct neuronal circuits have been identified that form different parallel representations of auditory information from auditory nerve fibers (ANFs). One cell circuit of interest is the stellate cell microcircuit of the ventral cochlear nucleus (VCN), which is centered on two populations of stellate cells, known as T and D stellate cells [1]. The output of the T stellate (TS) cell population has been shown experimentally to accurately encode important frequency components over a wide range of input sound pressure levels (SPL) and in the presence of significant signal noise [2]. It is thought that one purpose of the stellate cell circuit is to increase the dynamic range and the signal-to-noise ratio (SNR) of the incoming signal at all levels in order to more accurately highlight the most significant frequencies. A clear application of this in speech processing is the identification of vowels, which requires detection of the first two or three formant frequencies. A neural network model of the stellate microcircuit was created using leaky integrate-and-fire neurons for each cell in the network. The key mechanisms incorporated to achieve SNR and dynamic range improvement were wide-band lateral inhibition and selective processing of excitatory and inhibitory inputs [3]. The number and type of cell inputs, the network connectivity, and the time constants associated with cellular dynamics were made, where possible, to agree with experimental observations. In addition, the performance of the model in response to a variety of stimuli was adjusted to achieve the desired signal processing functions regarding noise reduction, peak identification and dynamic range. To quantify the enhancement achieved by the network, two signal-to-noise metrics were used. SNR metric 1 measures the level of noise reduction and is calculated as the ratio of the average output rate of cells within 1.5 critical bands of the known stimulus frequency to the average background output rate. SNR metric 2 measures the level of peak enhancement and is calculated as the ratio of the maximum output rate of the cells within 1.5 critical bands of the known stimulus frequency to the maximum background output rate. The metrics were calculated for the input signal, the input ANF population, and the TS cell population. Each metric was calculated for the first three formants of the vowel /o:/ and for a range of input SNR values. The SNR metric 1 was significantly higher than for the ANF population for the majority of input SNR levels. Similarly, SNR metric 2 was also higher for the TS cell population than for the ANF population. This demonstrates that the TS cells improved the ability to decode the spectral information in noise and the model illustrates the mechanisms by which this is achieved.

Slice accelerated gradient‐echo spin‐echo dynamic susceptibility contrast imaging with blipped CAIPI for increased slice coverage

To improve slice coverage of gradient echo spin echo (GESE) sequences for dynamic susceptibility contrast (DSC) MRI using a simultaneous-multiple-slice (SMS) method. Data were acquired on 3 Tesla (T) MR scanners with a 32-channel head coil. To evaluate use of SMS for DSC, an SMS GESE sequence with two-fold slice coverage and same temporal sampling was compared with a standard GESE sequence, both with 2× in-plane acceleration. A signal to noise ratio (SNR) comparison was performed on one healthy subject. Additionally, data with Gadolinium injection were collected on three patients with glioblastoma using both sequences, and perfusion analysis was performed on healthy tissues as well as on tumor. Retained SNR of SMS DSC is 90% for a gradient echo (GE) and 99% for a spin echo (SE) acquisition, compared with a standard acquisition without slice acceleration. Comparing cerebral blood volume maps, it was observed that the results of standard and SMS acquisitions are comparable for both GE and SE images. Two-fold slice accelerated DSC MRI achieves similar SNR and perfusion metrics as a standard acquisition, while allowing a significant increase in slice coverage of the brain. The results also point to a possibility to improve temporal sampling rate, while retaining the same slice coverage.

Energy Efficiency of Power-Adaptive Spatial Diversity Methods

The energy efficiency of power-adaptive multipleinput multiple-output (MIMO) diversity methods is studied in this paper. The considered diversity methods are antenna selection (AS), maximum ratio transmission (MRT), and equal gain transmission (EGT) at the transmitter and maximum ratio combining (MRC) at the receiver. The transmitter energy efficiency is evaluated using the average power amplifier (PA) efficiency and the transceiver energy efficiency is evaluated using the bit error rate (BER) as a function of average PA input signal-to-noise ratio (SNR), which is a new metric. With the new metric, the effect of the PA efficiency is taken into account in the performance evaluation. The analytical results are verified by Monte Carlo simulations. It is shown that larger diversity in the spatial or frequency domain improves the average PA efficiency in a system with channel inversion. The BER results show that the performance improvement from channel inversion diminishes due to the nonideal PA efficiency. Even though MRT is the received SNR maximizing transmitter diversity method, EGT requires less PA input energy per bit when a PA with nonideal efficiency is used. These conclusions could not have been reached using the traditional SNR metrics that do not measure the PA input energy.

Spectral Models for Bitrate Measurement from Packet Sampled Traffic

In network measurement systems, packet sampling techniques are usually adopted to reduce the overall amount of data to collect and process. Being based on a subset of packets, they introduce estimation errors that have to be properly counteracted by using a fine tuning of the sampling strategy and sophisticated inversion methods. This problem has been deeply investigated in the literature with particular attention to the statistical properties of packet sampling and to the recovery of the original network measurements. Herein, we propose a novel approach to predict the energy of the sampling error in the real time estimation of traffic bitrate, based on spectral analysis in the frequency domain. We start by demonstrating that the error introduced by packet sampling can be modeled as an aliasing effect in the frequency domain. Then, we derive closed-form expressions for the Signal-to-Noise Ratio (SNR) to predict the distortion of traffic bitrate estimates over time. The accuracy of the proposed SNR metric is validated by means of real packet traces. Furthermore, a comparison with respect to an analogous SNR expression derived using classic stochastic tools is proposed, showing that the frequency domain approach grants for a higher accuracy when traffic rate measurements are carried out at fine time granularity.

Error prediction for adaptive modulation and coding in multiple-antenna OFDM systems

In this paper, the problem of packet error rate (PER) prediction is addressed in the multiple-antenna broadband OFDM context, and its impact on adaptive modulation and coding (AMC) is quantified. The analysis is based on a physical layer comprising various modulation and coding schemes, ranging from robust space–time block coding (STBC) modes to high bit rate spatial division multiplexing (SDM) modes, and also hybrid SDM–STBC schemes. For each mode the expression of several link quality metrics (LQM) enabling PER prediction in the broadband OFDM channel, such as instantaneous signal-to-noise ratio (SNR), capacity, or exponential effective SNR metrics are provided. Their advantages and limitations are investigated. Finally, their performance is benchmarked in the IEEE 802.11a/g/n context. It is shown that the choice of the LQM has a significant impact on the throughput performance of the AMC algorithm.

Some results and insights on the performance gains of MIMO systems

Multiple-input-multiple-output (MIMO) systems generally fall into two categories, depending on the kind of gain they provide: spatial multiplexing (SM) methods yield capacity gain and diversity methods yield link quality gain [measured here by the post-processing signal-to-noise ratio (SNR)]. We consider a set of systems from each category, quantify their gains analytically or via simulations, and show how these gains vary with the receiver input SNR and the numbers of antennas. The contribution of this work resides in both the closed-form analytical results and the numerical comparisons. We both highlight the benefits of using additional transmit antennas and provide comparisons among diversity-based and SM-based MIMO schemes. The analytical results are for a diversity-based scheme that combines selection diversity at the transmit side with maximum ratio combining (MRC) at the receive side, which we show to upper bound the SNR performance of other diversity-based schemes. We find that, for practical system parameters, the relevant SNR metric is 6-12 dB higher for the diversity-based schemes than for SM-based schemes. At the same time, SM-based schemes yield capacity metrics which range from 30% higher to double that of diversity-based schemes.