Conventional deep learning (DL) model compression methods affect all input samples equally. However, as samples vary in difficulty, a dynamic model that adapts computation based on sample complexity offers a novel perspective for compression and scaling. Despite this potential, existing dynamic techniques are typically monolithic and have model-specific implementations, limiting their generalizability as broad compression and scaling methods. Additionally, most deployed DL systems are fixed, and unable to adjust once deployed. This paper introduces DyCE, a dynamically configurable system that can adjust the performance-complexity trade-off of a DL model at runtime without needing re-initialization or re-deployment. DyCE achieves this by adding exit networks to intermediate layers, thus allowing early termination if results are acceptable. DyCE also decouples the design of exit networks from the base model itself, enabling its easy adaptation to new base models. We also propose methods for generating optimized configurations and determining exit network types and positions for dynamic trade-offs. By enabling simple configuration switching, DyCE enables fine-grained performance-complexity tuning in real-time. We demonstrate the effectiveness of DyCE through image classification tasks using deep convolutional neural networks (CNNs). DyCE significantly reduces computational complexity by 26.2% for ResNet 152 , 26.6% for ConvNextv2 tiny and 32.0% for DaViT base on ImageNet validation set, with accuracy reductions of less than 0.5%. • Effectively compress the computational complexity of deep learning models. • Dynamically Scale any AI model and select a complexity-performance tradeoff point for the model in run-time. • Enable early exiting on any existing deep learning models by attaching tiny exits. • Generates the best early-exit configuration in a multi-exit system for various trade-off preferences.

This paper introduces TinySaver, an early-exit-like dynamic model compression approach which employs tiny models to substitute large models adaptively. Distinct from traditional compression techniques, dynamic methods like TinySaver can leverage the difficulty differences to allow certain inputs to complete their inference processes early, thereby conserving computational resources. Most existing early exit designs are implemented by attaching additional network branches to the model’s backbone. Our study, however, reveals that completely independent tiny models can replace a substantial portion of the larger models’ job with minimal impact on performance. Employing them as the first exit can remarkably enhance computational efficiency. By searching and employing the most appropriate tiny model as the computational saver for a given large model, the proposed approaches work as a novel and generic method to model compression. This finding will help the research community in exploring new compression methods to address the escalating computational demands posed by rapidly evolving AI models. Our evaluation of this approach in ImageNet-1k classification demonstrates its potential to reduce the number of compute operations by up to 90%, with only negligible losses in performance, across various modern vision models.

Time-interleaved analog-to-digital converters (TIADCs) are widely used in communication systems due to their exceptionally high sampling rates; however, in real-world applications, the offset, gain, and time-skew mismatches in TIADCs are a significant challenge for the circuit system. This article proposes a forward error correction (FEC)-aided decision feedback blind mismatch calibration for TIADCs in the time-varying channels environment specific to the orthogonal frequency-division multiplexing (OFDM) system. In our proposed approach, we use an FEC decision feedback technique to generate a ground truth reference signal for the purpose of calibration. There are two stages. In the first stage, the offset and gain mismatches are estimated and corrected using standard techniques. In the second stage, an adaptive filter bank corrects the time-skew mismatch directly without the need for any additional calibration hardware. The coefficients of this adaptive filter are continuously adjusted in the background based on an error signal derived from the decision feedback ground truth signal. This calibration algorithm significantly reduces the bit error rate (BER) and improves the system performance. The efficacy of these approaches is validated through comprehensive simulations to attain a performance assessment, quantified by the BER, using a realistic wireless time-varying channel system configuration.

Electrocardiogram (ECG) analysis is widely used in the diagnosis of cardiovascular diseases. This paper proposes an explainable rule-mining strategy for prioritizing abnormal class detection in ECG data. The proposed method utilizes a biased-trained Artificial Neural Network (ANN) with input features derived from an ECG beat sequence and formulates a set of rules at each node of an on-demand tree-like search algorithm. The rule base at each node is derived from a linear combination of the most impactful features identified using gradient analysis in an ANN. The final derived model is an explainable rule-based system that detects abnormal heartbeats based on statistical and morphological features from ECG. The model achieves the target sensitivity, and accuracy with a low run-time complexity through a comprehensive offline rule mining process and is trained using the MIT-BIH Arrhythmia Database. The system achieves an accuracy of 93% with only nine nodes and a test sensitivity of 90% and 80% respectively for VEB and SVEB beat types, when tested on previously unseen ECG data from the INCART database. The model performance and complexity can be easily adjusted based on the real-time resource constraints of a wearable sensor. The model was deployed on an ARM Cortex M4-based embedded device and is shown to achieve a >50% reduction in sensor power consumption when only abnormal beats are wirelessly transmitted. i.e RF transmission is gated using the model output and transmission is disabled when the subject’s ECG is normal. The proposed technique is highly suited for healthcare applications because of its explainability, lower complexity, and real-time flexibility when deployed in the Internet of Things (IoT) enabled wearable edge sensors.

This paper proposes a foreground digital calibration algorithm that estimates and corrects the offset, gain, and time-skew mismatches for time-interleaved analog-to-digital converters (TIADCs) furthermore our algorithm is designed to correct for harmonic distortion introduced by the presence of a nonlinear front-end. We propose a novel simplified non-linear model in place of the more complex conventional Volterra series based structure. The mismatch estimation technique based on the Fast Fourier Transform (FFT) is proposed to estimate the various time-interleaving mismatches simultaneously. A Taylor-based technique is applied to compensate for these mismatches. We also consider the choice of an appropriate time reference for the time-skew correction algorithm by theoretical analysis. The nonlinear distortion correction technique is based on estimating and inverting an assumed$3^{\text {rd}}$order nonlinearity with a fractional delay. To do this, we design a customized filter in an offline process. Our algorithms are designed to operate in any Nyquist zone. The proposed techniques are verified by a Xilinx Zynq UltraScale+ RFSoC ZCU111 evaluation kit containing a 12-bit, 4.096 GHz TI-ADC with 8 sub-ADCs operating in the$2^{\text {nd}}$Nyquist zone. Accordingly, we observed an improvement in SFDR of 14 dB for mismatch calibration alone and up to another 12 dB with nonlinear correction enabled.

The accurate detection of heartbeats is of paramount importance in the current healthcare scenario as they act as an indicator for various underlying cardiac conditions and provides an indication of cardiorespiratory fitness. The article presents a novel multimodal data fusion technique using the discrete wavelet transform (DWT) and an application for fusing electrocardiogram (ECG), and photoplethysmogram (PPG) signals to improve beat detection accuracy in ambulatory monitoring using Internet of Things (IoT) sensors. The characteristics of interest from the input signals are first isolated in the wavelet domain and then combined to form a fused feature signal using a weighted average. The weights used are derived from a signal quality index calculation algorithm, suitable for periodic/quasiperiodic signals of different wave morphologies. The peak detection process to identify the heartbeat locations is carried out on the final fused signal. The research evaluates the algorithm performance when different types of noises at varying amplitudes corrupt the ECG and PPG signal inputs, affecting the signal-to-noise ratios (SNRs). The algorithm consistently exhibited a sensitivity of 99.69%, positive predictive value (PPV) of 99.64%, mean beat-to-beat interval relative error of 0.01, and an error spread (corresponding to 90th percentile of relative errors) of 0.02 in the −30 to 50 dB SNR range for all noise scenarios considered. The proposed algorithm exhibits improved detection sensitivities and PPVs under ambulatory conditions compared to state-of-the-art beat detection algorithms and can be used to accurately detect heartbeats where single-channel monitoring tends to fail in IoT devices.

Internet of Things (IoT) enabled wearable sensors for health monitoring are widely used to reduce the cost of personal healthcare and improve quality of life. The sleep apnea-hypopnea syndrome, characterized by the abnormal reduction or pause in breathing, greatly affects the quality of sleep of an individual. This paper introduces a novel method for apnea detection (pause in breathing) from electrocardiogram (ECG) signals obtained from wearable devices. The novelty stems from the high resolution of apnea detection on a second-by-second basis, and this is achieved using a 1-dimensional convolutional neural network for feature extraction and detection of sleep apnea events. The proposed method exhibits an accuracy of 99.56% and a sensitivity of 96.05%. This model outperforms several lower resolution state-of-the-art apnea detection methods. The complexity of the proposed model is analyzed. We also analyze the feasibility of model pruning and binarization to reduce the resource requirements on a wearable IoT device. The pruned model with 80% sparsity exhibited an accuracy of 97.34% and a sensitivity of 86.48%. The binarized model exhibited an accuracy of 75.59% and sensitivity of 63.23%. The performance of low complexity patient-specific models derived from the generic model is also studied to analyze the feasibility of retraining existing models to fit patient-specific requirements. The patient-specific models on average exhibited an accuracy of 97.79% and sensitivity of 92.23%. The source code for this work is made publicly available.

In this article, non-uniformly sampled electrocardiogram (ECG) signals obtained from level-crossing analog-to-digital converters (LC-ADCs) are analyzed for event-driven classification and compression performance. The signal compression results show that it is important to assess the distortion in eventdriven signals when simulating LC-ADC models, especially at lower resolutions and larger quantization steps. The effects of varying the LC-ADC parameters for the application of cardiac arrhythmia classifiers are also assessed using an artificial neural network (ANN) and the MIT-BIH Arrhythmia Database. In comparison with uniformly-sampled data, it is possible to achieve comparable classification accuracy at a much lower complexity with event-driven ECG signals. The results show the best eventdriven model achieves over 97% accuracy with 79% reduction in ANN complexity with signal-to-distortion ratio (S/D)>21dB. For S/D<; 21dB, the best event-driven model achieves 93% accuracy with a 96% reduction in ANN complexity. An open-source event-driven arrhythmia database is also presented.

Over the past several years, the electrocardiogram (ECG) has been investigated for its uniqueness and potential to discriminate between individuals. This paper discusses how this discriminatory information can help in continuous user authentication by a wearable chest strap which uses dry electrodes to obtain a single lead ECG signal. To the best of the authors' knowledge, this is the first such work which deals with continuous authentication using a genuine wearable device as most prior works have either used medical equipment employing gel electrodes to obtain an ECG signal or have obtained an ECG signal through electrode positions that would not be feasible using a wearable device. Prior works have also mainly dealt with using the ECG signal for identification rather than verification, or dealt with using the ECG signal for discrete authentication. This paper presents a novel algorithm which uses QRS detection, weighted averaging, Discrete Cosine Transform (DCT), and a Support Vector Machine (SVM) classifier to determine whether the wearer of the device should be positively verified or not. Zero intrusion attempts were successful when tested on a database consisting of 33 subjects.

Non-orthogonal multiple access (NOMA) has drawn tremendous attention, being a potential candidate for the spectrum access technology for the fifth-generation (5G) and beyond 5G(B5G) wireless communications standards. Most research related to NOMA focuses on the system performance from Shannon's capacity perspective, which, although a critical system design criterion, fails to quantity the effect of delay constraints imposed by future wireless applications. In this paper, we analyze the performance of a single-input multiple-output (SIMO) two-user downlink NOMA system, in terms of the link-layer achievable rate, known as effective capacity (EC), which captures the performance of the system under a delay-limited quality-of-service (QoS) constraint. For signal combining at the receiver side, we use generalized selection combining (GSC), which bridges the performance gap between the two conventional diversity combining schemes, namely selection combining (SC) and maximal-ratio combining (MRC). We also derive two approximate expressions for the EC of NOMA-GSC which are accurate at low-SNR and at high-SNR, respectively. The analysis reveals a tradeoff between the number of implemented receiver radio-frequency (RF) chains and the achieved performance, and can be used to determine the appropriate number of paths to combine in a practical receiver design.

IoT wearable devices are widely expected to reduce the cost and risk of personal healthcare. However, ambulatory data collected from such devices are often corrupted or contaminated with severe noises. Signal Quality Indicators (SQIs) can be used to assess the quality of data obtained from wearable devices, such that transmission/ storage of unusable data can be prevented. This article introduces a novel and generalized SQI which can be implemented on an edge device for detecting the quality of any quasi-periodic signal under observation, regardless of the type of noise present. The application of this SQI on Electrocardiogram (ECG) signals is investigated. From the analysis carried out, it was found that the proposed generalized SQI is suitable for quality assessment of ECG signals and exhibits a linear behavior in the medium to high SNR regions under all noise conditions considered. The proposed SQI was used for acceptability testing of ECG records in CinC Physionet 2011 challenge dataset and found to be accurate for 90.4% of the records while having minimal computational complexity.

In orthogonal time frequency space (OTFS) systems, channel estimation (CE) is often performed using a pilot based approach. This is usually done in two steps: first, valid channel paths are detected by comparing the magnitude of received symbols against a threshold, and then the associated channel coefficient is estimated. In an attempt to avoid channel path misdetection, existing approaches often deploy large guard bands surrounding a relatively high-power pilot symbol. Furthermore, it is generally assumed that this guard region is sufficiently large to ensure that the time/frequency spreading of the channel does not result in any interference between the pilot and data symbols, thus facilitating simple channel estimation schemes. In this paper, we propose a channel estimation scheme which works even when this assumption is no longer true, i.e., when the channel results in pilot-data interference as may occur in scenarios involving high mobility and/or large delay-spread. The ability of a receiver to operate in such scenarios allows system designers the freedom to use smaller guard bands based on typical (not worst-case) channel spread parameters, thereby yielding higher spectral efficiency. In this work, we derive a maximum likelihood channel path detection scheme followed by a minimum mean-square error channel estimator. The performance advantage of the proposed receiver is verified using extensive bit error rate simulations.

Non-orthogonal multiple access (NOMA) is widely recognized as a potential multiple access technology for efficient radio spectrum utilization in the fifth-generation (5G) wireless communications standard. In this paper, we study the average achievable rate and outage probability of a cooperative relaying system (CRS) based on NOMA (CRS-NOMA) over wireless links governed by the α-μ generalized fading model; here α and μ designate the nonlinearity and clustering parameters, respectively, of each link. The average achievable rate is represented in closed-form using Meijer's G-function and the extended generalized bivariate Fox's H-function (EGBFHF), and the outage probability is represented using the lower incomplete Gamma function. Our results confirm that the CRS-NOMA outperforms the CRS with conventional orthogonal multiple access (CRS-OMA) in terms of spectral efficiency at high transmit signal-to-noise ratio (SNR). It is also evident from our results that with an increase in the value of the nonlinearity/clustering parameter, the SNR at which the CRS-NOMA outperforms its OMA based counterpart becomes higher. Furthermore, the asymptotic analysis of the outage probability reveals the dependency of the diversity order of each symbol in the CRS-NOMA system on the α and μ parameters of the fading links.

In this paper, we present the error performance analysis of a multiple-input multiple-output (MIMO) physical-layer network coding (PNC) system with two different user-antenna selection (AS) schemes in asymmetric channel conditions. For the first antenna selection scheme (AS1), where the user antenna is selected in order to maximize the overall channel gain between the user and the relay, we give an explicit analytical proof that for binary modulations, the system achieves full diversity order of min(NA, NB) × NRin the multiple-access (MA) phase, where NA, NB, and NRdenote the number of antennas at user A, user B, and relay R, respectively. We present a detailed investigation of the diversity order for the MIMO-PNC system with AS1 in the MA phase for any modulation order. A tight closed-form upper bound on the average SER is also derived for the special case when NR= 1, which is valid for any modulation order. We show that in this case, the system fails to achieve transmit diversity in the MA phase, as the system diversity order drops to 1 irrespective of the number of transmit antennas at the user nodes. Additionally, we propose a Euclidean distance (ED) based user-antenna selection scheme (AS2) that outperforms the first scheme in terms of error performance. Moreover, by deriving upper and lower bounds on the diversity order for the MIMO-PNC system with AS2, we show that this system enjoys both transmit and receive diversity, achieving full diversity order of min(NA, NB) × NRin the MA phase for any modulation order. Monte Carlo simulations are provided which confirm the correctness of the derived analytical results.

Non-orthogonal multiple access (NOMA) and spectrum sharing (SS) are two emerging multiple access technologies for efficient spectrum utilization in future wireless communications standards. In this paper, we present the performance analysis of a NOMA-based cooperative relaying system (CRS) in an underlay spectrum sharing scenario, considering a peak interference constraint (PIC), where the peak interference inflicted by the secondary (unlicensed) network on the primary-user (licensed) receiver (PU-Rx) should be less than a predetermined threshold. In the proposed system the relay and the secondary-user receiver (SU-Rx) are equipped with multiple receive antennas and apply selection combining (SC), where the antenna with highest instantaneous signal-to-noise ratio (SNR) is selected, and maximal-ratio combining (MRC), for signal reception. Closed-form expressions are derived for the average achievable rate and outage probabilities for SS-based CRS-NOMA. These results show that for large values of peak interference power, the SS-based CRS-NOMA outperforms the CRS with conventional orthogonal multiple access (OMA) in terms of spectral efficiency. The effect of the interference channel on the system performance is also discussed, and in particular, it is shown that the interference channel between the secondary-user transmitter (SU-Tx) and the PU-Rx has a more severe effect on the average achievable rate as compared to that between the relay and the PU-Rx. A close agreement between the analytical and numerical results confirm the correctness of our rate and outage analysis.

This paper presents a generic foreground calibration algorithm which compensates for memoryless nonlinear impairments in pipeline, SAR or hybrid ADC architectures. Amplifier nonlinearity, comparator offsets, capacitance mismatch and settling time errors are considered. During the calibration process, each element of a look up table is computed by mapping each raw ADC output value to an estimate of the corresponding input, and the most likely input corresponding to each raw ADC output is computed and stored in the table; this table is then used during normal operation to map the raw values to the calibrated ADC outputs. Complexity reduction techniques are presented to facilitate an in-circuit hardware implementation in order to reduce foreground calibration time. The algorithm's performance is evaluated using a SAR ADC model suffering from various nonlinear impairments. Results are presented for settling time errors, capacitor mismatch scenarios, and a wide range of nonlinear amplifier parameters, demonstrating a significant performance improvement in all cases.

Automatic classification of sentiment data (e.g., reviews, blogs) has many applications in enterprise user management systems, and can help us understand people’s attitudes about products or services. However, it is difficult to train an accurate sentiment classifier for different domains. One of the major reasons is that people often use different words to express the same sentiment in different domains, and we cannot easily find a direct mapping relationship between them to reduce the differences between domains. So, the accuracy of the sentiment classifier will decline sharply when we apply a classifier trained in one domain to other domains. In this paper, we propose a novel approach called words alignment based on association rules (WAAR) for cross-domain sentiment classification, which can establish an indirect mapping relationship between domain-specific words in different domains by learning the strong association rules between domain-shared words and domain-specific words in the same domain. In this way, the differences between the source domain and target domain can be reduced to some extent, and a more accurate cross-domain classifier can be trained. Experimental results on Amazon® datasets show the effectiveness of our approach on improving the performance of cross-domain sentiment classification.

Physical-layer network coding (PNC) is now well- known as a potential candidate for delay-sensitive and spectrally efficient communication applications, especially in two-way relay channels (TWRCs). In this paper, we present the error performance analysis of a multiple-input single- output (MISO) fixed network coding (FNC) system with two different transmit antenna selection (TAS) schemes. For the first scheme, where the antenna selection is performed based on the strongest channel, we derive a tight closed-form upper bound on the average symbol error rate (SER) with M-ary modulation and show that the system achieves a diversity order of 1 for M > 2. Next, we propose a Euclidean distance (ED) based antenna selection scheme which outperforms the first scheme in terms of error performance and is shown to achieve a diversity order lower bounded by the minimum of the number of antennas at the two users.

This paper presents a technique to estimate the time skew in time-interleaved ADCs. The proposed method estimates all of the time skew parameters jointly based on observations from a bank of correlators. The proposed method works for an arbitrary number of sub-ADCs. For implementation of the correlator bank, we propose the use of Mitchell's logarithmic multiplier and a hardware reuse mechanism, thereby reducing the complexity and power consumption. Also, we explain why blind estimation techniques alone (including the proposed one) are not always sufficient for time skew estimation for certain classes of input signal; for the proposed approach, however, a simple modification to the analogue circuit (suitable for SAR ADCs) is shown to successfully deal with such problems, with only a minor penalty in power and area. The technique is verified by extensive simulations including a spectrally rich input signal in which an MTPR (multi-tone power ratio) improvement from 29dB to 62dB was achieved for a TIADC system having 16 sub-ADCs.

Efficient heterogeneous communication technologies are critical components to provide flawless connectivity in smart cities. The proliferation of wireless technologies, services and communication devices has created the need for green and spectrally efficient communication technologies. Physical-layer network coding (PNC) is now well-known as a potential candidate for delay-sensitive and spectrally efficient communication applications, especially in bidirectional relaying, and is therefore well-suited for smart city applications. In this paper, we provide a brief introduction to PNC and the associated distance shortening phenomenon which occurs at the relay. We discuss the issues with existing schemes that mitigate the deleterious effect of distance shortening, and we propose simple and effective solutions based on the use of multiple antenna systems. Simulation results confirm that full diversity order can be achieved in a PNC system by using antenna selection schemes based on the Euclidean distance metric.

A novel algorithm is presented for determining the complex excitations in antenna arrays that are required to have a standard or a non-standard power pattern and where the individual antennas can have arbitrary locations and arbitrary patterns. The technique is simple and overcomes many of the disadvantages of traditional approaches such as those based on stochastic optimisation or least squares formulations. Our approach is iterative and involves repeatedly finding and reducing the peak pattern error. The method is particularly suited to large arrays with severe requirements such as those involving multiple beams with a mix of wide-angle flat-top and narrow regions, multiple attenuation zones with different specifications as well as shaped transition regions. Examples are presented demonstrating the effectiveness of the method.