Split Learning (SL) is a promising Distributed Learning approach in electromyography (EMG) based prosthetic control, due to its applicability within resource-constrained environments. Other learning approaches, such as Deep Learning and Federated Learning (FL), provide suboptimal solutions, since prosthetic devices are extremely limited in terms of processing power and battery life. The viability of implementing SL in such scenarios is caused by its inherent model partitioning, with clients executing the smaller model segment. However, selecting an inadequate cut layer hinders the training process in SL systems. This paper presents an algorithm for optimal cut layer selection in terms of maximizing the convergence rate of the model. The performance evaluation demonstrates that the proposed algorithm substantially accelerates the convergence in an EMG pattern recognition task for improving prosthetic device control.

In this paper, we propose to investigate a single RF chain multi-antenna full-duplex (FD) relay built with$b$-bit analog phase shifters and passive self-interference cancellation. Next, assuming only passive self-interference cancellation at the FD relay, we derive the achievable data rate of a system comprised of a source, the proposed FD relay, and a destination. We then compare the achievable data rate of the proposed FD relaying system with the achievable data rate of the same system but with the FD relay replaced by an ideal passive RIS. Our results show that the proposed relaying system with 2-bit quantized analog phase shifters significantly outperforms the RIS-assisted system. In fact, the performance gains are so large, at least for small to intermediate numbers of antenna elements, that we believe it makes this result of interest to the wireless community. The proposed FD relay can also be built with reconfigurable holographic surfaces, one surface for the transmit-side and one for the receive-side. For such a scenario, we derive the energy efficiency of the relay-assisted system and compare it with the RIS-assisted system. Our numerical results show that the energy efficiency of the relay-assisted system built with reconfigurable holographic surfaces is significantly higher than the energy efficiency of the RIS-assisted system. Intuitively, the RIS system is at a disadvantage since there the total transmit power$P_{T}$is used entirely by the source, whereas in the FD relaying system the total transmit power$P_{T}$is shared by the source and the FD relay in addition to the noise-cleansing process performed by the decode-and-forwarding at the FD relay.

In this paper, we investigate the secrecy capacity of a point-to-point, full-duplex (FD) wirelesly powered communication system in the presence of a passive eavesdropper (EVE). The considered system is comprised of an energy transmitter (ET), an energy harvesting user (EHU), and a passive EVE. The ET transmits radio-frequency energy, which is used for powering the EHU as well as for generating interference at the EVE. The EHU uses the energy harvested from the ET to transmit confidential messages back to the ET. As a consequence of the FD mode of operation, both the EHU and the ET are subjected to self-interference, which has different effects at the two nodes. In particular, the self-interference impairs the decoding of the received message at the ET, whilst it serves as an additional energy source at the EHU. For this system model, we derive an upper and a lower bound on the secrecy capacity. For the lower bound, we propose a simple achievability scheme. Our numerical results show significant improvements in terms of achievable secrecy rate when the proposed communication scheme is employed against its half-duplex counterpart, even for practical self-interference values at the ET.

In this paper, we investigate the capacity of a point-to-point, full-duplex (FD) wirelessly powered communication system impaired by self-interference. This system is comprised of an energy transmitter (ET) and an energy harvesting user (EHU), which operates in the FD mode. The ET transmits energy toward the EHU. The EHU harvests this energy and uses it to transmit information back to the ET. As a result of the FD mode, both nodes are affected by self-interference. The self-interference has different effects at the two nodes. In particular, the self-interference impairs the decoding of the received information signal at the ET, whereas it serves as an additional source of energy at the EHU. In this paper, we derive the capacity of the adopted system model assuming a processing cost at the EHU and an additive white Gaussian noise channel with block fading. Thereby, we show that the capacity achieving scheme is relatively simple and therefore applicable to devices with limited resources. Moreover, our numerical results show significant improvements in terms of data rate when the capacity achieving strategy is employed compared to half-duplex transmission, even for very high self-interference at the ET. Moreover, we show the positive effects of the self-interference at the EHU, as well as the crippling effect of the processing cost.

In this paper, we consider a point-to-point multi-input multi-output wireless-powered communication system, where the source S is powered by a dedicated power beacon (PB) with multiple antennas. Employing the time splitting protocol, the energy constrained source S first harvests energy through the radio-frequency signals sent by the PB and then uses this energy to transmit information to the destination D. Unlike several prior works, we assume that the energy transfer link is subjected to Rician fading, which is a real fading environment, due to relatively short range power transfer distance and the existence of a strong line of sight path. We present a comprehensive analysis of the achievable ergodic capacity in two scenarios, depending on the availability of channel state information (CSI) at PB, namely, the absence of CSI and partial CSI. For the former case, equal power allocation is used, while for the later one, energy beamforming is used to enhance the energy transfer efficiency. For both the cases, closed-form expressions for the upper and lower bounds of the ergodic capacity are derived. Furthermore, the optimal time split is discussed, and the capacity in the low and high signal-to-noise ratio regimes is studied through simple closed-form expressions. Numerical results and simulations are provided to validate the theoretical analysis. The results show that the Rician factor K has a significant impact on the ergodic capacity performance, and this impact strongly depends on the availability of the CSI at the PB.

Energy harvesting has proved to be a viable solution for energy bottlenecks of wireless networks. This paper considers a single tier cellular network, in which the mobile nodes harvest energy from the aggregate downlink inter-cell interference and latter use it as uplink transmit power. A successful uplink transmission is assumed only when a user harvested enough energy in its battery and the received signal-to-interference ratio (SIR) is above a predefined threshold. We investigate the performance of the proposed system model in both downlink and uplink using stochastic geometry and provide theoretical analysis and numerical evaluation. Denser macro tiers show higher uplink successful transmission probability due to the increased number of available power extraction sources i.e macro base stations.

In this paper, we apply the notion of opportunistic scheduling in wireless-powered communication networks (WPCNs). The considered WPCN model consists of a base station (BS) and multiple energy harvesting users (EHUs), where the BS broadcasts radio frequency energy to the EHUs over the downlink and receives information from the EHUs over the uplink. We differentiate the WPCNs based upon the battery management policy at the EHUs, i.e., whether an EHU spends the total amount of energy harvested in its battery for each IT (WPCN type 1), or spends only a part of it for the current IT and saves the other part for future ITs (WPCN type 2). We propose two opportunistic scheduling policies, referred to as the harvest-then-select and harvest-or-select protocols, employed at the WPCN type 1 and WPCN type 2, respectively. These protocols have significant practical advantages over the state-of-the-art schemes proposed for maximizing the WPCN sum-rate, because they introduce fairness in the resource utilization by the EHUs, and require much lower amount of channel state information. Both protocols achieve these benefits at the expense of a minor rate degradation relative to the rates achieved by their counterpart protocols employing multiple access, denoted as the harvest-then-transmit and harvest-or-concurrently-transmit protocols.

In this letter, we consider wireless powered communication networks which could operate perpetually, as the base station (BS) broadcasts energy to the multiple energy harvesting (EH) information transmitters. These employ “harvest then transmit” mechanism, as they spend all of their energy harvested during the previous BS energy broadcast to transmit the information towards the BS. Assuming time division multiple access (TDMA), we propose a novel transmission scheme for jointly optimal allocation of the BS broadcasting power and time sharing among the wireless nodes, which maximizes the overall network throughput, under the constraint of average transmit power and maximum transmit power at the BS. The proposed scheme significantly outperforms “state of the art” schemes that employ only the optimal time allocation. If a single EH transmitter is considered, we generalize the optimal solutions for the case of fixed circuit power consumption, which refers to a much more practical scenario.

Due to the fundamental tradeoffs, achieving spectrum efficiency and energy efficiency are two contending design challenges for the future wireless networks. However, applying radio-frequency (RF) energy harvesting (EH) in a cognitive radio system could potentially circumvent this tradeoff, resulting in a secondary system with limitless power supply and meaningful achievable information rates. This paper proposes an online solution for the optimal time allocation (time sharing) between the EH phase and the information transmission (IT) phase in an underlay cognitive radio system, which harvests the RF energy originating from the primary system. The proposed online solution maximizes the average achievable rate of the cognitive radio system, subject to the ε-percentile protection criteria for the primary system. The optimal time sharing achieves significant gains compared to equal time allocation between the EH and IT phases.

Wireless systems that carry delay-sensitive information (such as speech and/or video signals) typically transmit with fixed data rates, but may occasionally suffer from transmission outages caused by the random nature of the fading channels. If the transmitter has instantaneous channel state information (CSI) available, it can compensate for a significant portion of these outages by utilizing power allocation. In a conventional dual-hop bidirectional amplify-and-forward (AF) relaying system, the relay already has instantaneous CSI of both links available, as this is required for relay gain adjustment. We therefore develop an optimal power allocation strategy for the relay, which adjusts its instantaneous output power to the minimum level required to avoid outages, but only if the required output power is below some cutoff level; otherwise, the relay is silent in order to conserve power and prolong its lifetime. The proposed scheme is proven to minimize the system outage probability, subject to an average power constraint at the relay and fixed output powers at the end nodes.

For a point-to-point communication system generating its power via energy harvesting (EH), we derive the asymptotically optimal power allocation which optimizes a general utility function when the number of transmitted codewords N and the battery capacity Bmaxsatisfy N → ∞ and Bmax→ ∞. The considered family of utility functions is general enough to include the most important performance measurements in communication theory such as ergodic rate, outage probability, average signal-to-noise ratio, etc. The discovered solution is very simple. Namely, the optimal power allocation for the EH system is identical to the optimal power allocation of an equivalent non-EH communication system with infinite available energy, under the constraint that both systems use identical average transmit powers. Although the proposed solution is asymptotic, it is applicable to EH systems transmitting a large but finite number of codewords and having a battery capacity much larger than the average harvested power and/or the maximum average transmit power.

For the conventional three-node amplify-and-forward (AF) relaying setup, we investigate the effect of imperfect channel state information (CSI) at the relay on the overall performance. In particular, we consider variable gain (a.k.a, CSI-assisted) AF relaying and derive expressions for the outage and the error probability for the case where the relay gain is adjusted based on outdated estimates of the source-relay channel, when operating over Nakagami-m fading. For the case of Rayleigh fading in the source-relay link, we show that the results can be extended to the versatile case of imperfect CSI, where the estimation error is caused either by additive white Gaussian noise or by quantization noise. The obtained expressions are functions of the correlation coefficient between the actual source-relay channel and its corresponding estimate. We also optimize the power allocation for minimization of the outage probability under a total transmit power constraint. Numerical results reveal a considerable degradation of the overall performance, when CSI acquisition is not perfect. Moreover, it is shown that the average relay transmit power is affected when the CSI is outdated, a fact which impacts the design of variable gain relaying in practice. Since outdated CSI leads to fluctuations of the relay transmit power, we derive expressions for the complementary cumulative distribution function (CCDF) of the peak-to-average-power ratio (PAPR) at the relay. By comparing the error probability and the CCDF of the PAPR of variable gain relaying with those of fixed gain relaying, we shed some light onto the following question: How reliable has the instantaneous CSI at the relay to be for variable gain relaying to be preferable over fixed gain relaying, which requires only statistical CSI?

With reference to the typical three-node amplify-and-forward (AF) relaying setup, where a single relay assists the communication between a source terminal and a destination terminal, we investigate the effect of outdated channel state information (CSI) at the relay on the overall performance. In particular, we consider CSI-assisted (i.e., variable-gain) AF relaying and derive expressions for the outage and the error probability for the case where the relay gain is adapted based on outdated estimates of the source-relay channel, when operating over Nakagami-m fading. The obtained expressions are functions of the scale and shape fading parameters, as well as the correlation coefficient between the actual source-relay channel and its corresponding estimate. Numerical results reveal a considerable degradation of the overall performance, when CSI acquisition is not perfect.

A complete evaluation of the benefits of cooperative diversity schemes should not only include the outage and error rate performance but also the second-order statistics of the achievable information-theoretic capacity. In a non-ergodic fading channel, the system is said to be in outage when the destination cannot decode the fixed-rate transmitted signal with negligible error probability. Because of the Doppler effect, which is induced by the mobility of the wireless nodes, these outage capacity events are correlated. In this paper, we derive the average outage rate (AOR) and the average outage duration (AOD) of two well-known cooperative diversity protocols, decode-and-forward relaying and selection decode-and-forward relaying, operating in slow Rayleigh fading channels. We also analyze the asymptotic behavior of these statistical parameters for high SNRs, where it is shown that the AOR exhibits a similar behavior as the outage probability.

The outage probability is an important performance measure for cooperative diversity schemes. However, in mobile environments, the outage probability does not completely describe the behavior of cooperative diversity schemes since the mobility of the involved nodes introduces variations in the channel gains. As a result, the capacity outage events are correlated in time and second-order statistical parameters of the achievable information-theoretic capacity such as the average capacity outage rate (AOR) and the average capacity outage duration (AOD) are required to obtain a more complete description of the properties of cooperative diversity protocols. In this paper, assuming slow Rayleigh fading, we derive exact expressions for the AOR and the AOD of three well-known cooperative diversity protocols: variable-gain amplify-and-forward, decode-and-forward, and selection decode-and-forward relaying. Furthermore, we develop asymptotically tight high signal-to-noise ratio (SNR) approximations, which offer important insights into the influence of various system and channel parameters on the AOR and the AOD. In particular, we show that on a double-logarithmic scale, similar to the outage probability, the AOR asymptotically decays with the SNR with a slope that depends on the diversity gain of the cooperative protocol, whereas the AOD asymptotically decays with a slope of -1/2 independent of the diversity gain.

The WiMAX channel access is controlled by one of the several available Medium Access Control (MAC) procedures. This paper studies the contention-based bandwidth request procedure based on our original analytical model, facilitated by numerical analysis and simulation. We optimize the key parameters of the contention procedure in order to minimize the average bandwidth access delay, thus ensuring the highest possible quality of service to the WiMAX users. We also propose procedures for efficient and accurate estimation of all input parameters to carry out the procedure for bandwidth request optimization. We show that the analytical model and the estimation methods are highly accurate. The practical implementation of the proposed optimization is simple and results in gains of the delay and bandwidth utilization compared to the non-optimized case.

There are several cases in wireless communications theory where the statistics of the sum of independent or correlated Nakagami-m random variables (RVs) is necessary to be known. However, a closed-form solution to the distribution of this sum does not exist when the number of constituent RVs exceeds two, even for the special case of Rayleigh fading. In this paper, we present an efficient closed-form approximation for the distribution of the sum of arbitrary correlated Nakagami-m envelopes with identical and integer fading parameters. The distribution becomes exact for maximal correlation, while the tightness of the proposed approximation is validated statistically by using the Chi-square and the Kolmogorov-Smirnov goodness-of-fit tests. As an application, the approximation is used to study the performance of equal-gain combining (EGC) systems operating over arbitrary correlated Nakagami-m fading channels, by utilizing the available analytical results for the error-rate performance of an equivalent maximal-ratio combining (MRC) system.

We present a novel and accurate approximation for the distribution of the sum of equally correlated Nakagami-m variates. Ascertaining on this result we study the performance of Equal Gain Combining (EGC) receivers, operating over equally correlating fading channels. Numerical results and simulations show the accuracy of the proposed approximation and the validity of the mathematical analysis.

Second order statistics provides a dynamic representation of a fading channel and plays an important role in the evaluation and design of the wireless communication systems. In this paper, we present a novel analytical framework for the evaluation of important second order statistical parameters, as the level crossing rate (LCR) and the average fade duration (AFD) of the amplify-and-forward multihop Rayleigh fading channel. More specifically, motivated by the fact that this channel is a cascaded one and can be modeled as the product of N fading amplitudes, we derive novel analytical expressions for the average LCR and the AFD of the product of N Rayleigh fading envelopes (or of the recently so-called NRayleigh channel). Furthermore, we derive simple and efficient closed-form approximations to the aforementioned parameters, using the multivariate Laplace approximation theorem. It is shown that our general results reduce to the corresponding ones of the specific dual-hop case, previously published. Numerical and computer simulation examples verify the accuracy of the presented mathematical analysis and show the tightness of the proposed approximations.

We present a novel analytical framework for the evaluation of important second order statistical parameters, as the level crossing rate (LCR) and the average fade duration (AFD) of the amplify-and-forward multihop Rayleigh fading channel. More specifically, motivated by the fact that this channel is a cascaded one, which can be modelled as the product of N fading amplitudes, we derive novel analytical expressions for the average LCR and AFD of the product of N Rayleigh fading envelopes, or of the recently so-called N*Rayleigh channel. Furthermore, we derive simple and efficient closed-form approximations to the aforementioned parameters, using the multivariate Laplace approximation theorem. It is shown that our general results reduce to the specific dual-hop case, previously published. Numerical and computer simulation examples verify the accuracy of the presented mathematical analysis and show the tightness of the proposed approximations.

This paper provides exact closed-form expressions for the outage probability (OP), the average level crossing rate (LCR) and the average fade duration (AFD) of the selection combining (SC) systems exposed to the cochannel interference (CCI) in Nakagami and/or Rician fading channels. The SC is of an arbitrary diversity order and employs the signal-to- interference power (SIR) algorithm for branch selection, while all input signals are assumed independent. The analytical solution for the average LCR is quite general, applicable over arbitrary combinations of input signal statistics, and specializes to known expressions in absence of CCI and for non-diversity systems in Rayleigh fading. We also establish an important and yet simple relationship between the AFDs of SC systems and non-diversity systems.

Both the first-order signal statistics (e.g., the outage probability) and the second-order signal statistics [e.g., the average level crossing rate (LCR) and the average fade duration (AFD)] are important design criteria and performance measures for the wireless communication systems, including the equal gain combining (EGC) systems in presence of the cochannel interference (CCI). Although the analytical expressions for the outage probability of the coherent EGC systems exposed to CCI and various fading channels are already known, the respective ones for the average LCR and the AFD are not available in the literature. This paper presents such analytical expressions for the Rayleigh fading channel that are obtained by utilizing a novel analytical approach that does not require the explicit expression for the joint probability density function of the instantaneous output signal-to-interference ratio and its time derivative. Applying the characteristic function method and the Beaulieu series, we determined the average LCR and the AFD at the output of an interference-limited EGC system with an arbitrary diversity order and an arbitrary number of cochannel interferers in forms of an infinite integral and an infinite series. For the dual diversity case, the respective expressions are derived in closed forms in terms of the gamma and the beta functions.

Both the first-order signal statistics (e.g., the outage probability) and the second-order signal statistics [e.g., the average level crossing rate (LCR) and the average fade duration (AFD)] are important design criteria and performance measures for wireless communication systems, including the equal gain combining (EGC) systems in the presence of the cochannel interference (CCI). Although the analytical expressions for the outage probability of the coherent EGC systems are exposed to CCI and various fading channels are already known, the respective expressions for the average LCR and the AFD are not available in the literature. This paper presents such analytical expressions for the Rayleigh fading channel, which are obtained by utilizing a novel analytical approach that does not require the explicit expression for the joint probability density function (pdf) of the instantaneous output signal-to-interference ratio and its time derivative. Applying the characteristic function method and the Beaulieu series, we determined the average LCR and the AFD at the output of an interference-limited EGC system with an arbitrary diversity order and an arbitrary number of cochannel interferers in the form of an infinite integral and infinite series. For the dual diversity case, the respective expressions are derived in closed forms in terms of the gamma and beta functions.

This letter provides closed-form expressions for the outage probability, the average level crossing rate (LCR) and the average fade duration (AFD) of a dual diversity selection combining (SC) system exposed to the combined influence of the cochannel interference (CCI) and the thermal noise (AWGN) in Nakagami fading channel. The branch selection is based on the desired signal power SC algorithm with all input signals assumed to be independent, while the powers of the desired signals in all diversity branches are mutually equal but distinct from the power of the interference signals. The analytical results reduce to known solutions in the cases of an interference-limited system in Rayleigh fading and an AWGN-limited system in Nakagami fading. The average LCR is determined by an original approach that does not require explicit knowledge of the joint PDF of the envelope and its time derivative, which also paves the way for similar analysis of other diversity systems.

In this paper, we provide an extended model for analytical analysis of an IEEE 802.11 WLAN that accommodates the non-saturated operation of the network, the finite number of (re)transmission attempts for each station and the erroneous conditions in the wireless channel. This model builds on top of an existing Markov chain-based models for the station's backoff counter and proves to be more realistic that the original ones. We derive a generalized expression for the station's transmission probability, which is essential for determination of all system's performance measures. We also present a set of previously unpublished performance results for the throughput, the discard probability and the collision probability in function of the retry limit and the packet error rate

We introduce both the power and delay capture effect in BER performance analysis of the conventional RAKE receiver for DPSK modulated signals. Depending on the inter-arrival times between the useful and interfering frames from simultaneously transmitting stations, the interference signal noise can be highly correlated or uncorrelated with the DSSS signal from the captured frame.

We analytically analyzed the impact of an error-prone channel over all performance measures in a traffic-saturated IEEE 802.11 WLAN. We calculated station's transmission probability by using the modified Markov chain model of the backoff window size that considers the frame-error rates and maximal allowable number of retransmission attempts. The frame error rate has a significant impact over theoretical throughput, mean frame delay, and discard probability. The peak throughput of a WLAN is insensitive of the maximal number of retransmissions. Discard probabilities are insensitive to the station access method, basic or RTS/CTS.

In this paper, we examined the influence of capture effect with L-fold antenna diversity at the access point over IEEE 802.11b DCF at 11 Mbit/s. We are able to obtain an exact closed-form solution for the conditional capture probability in case of L-fold ideal selection diversity, and an approximate closed-form solution for the conditional capture probability in case of L-fold maximum selection diversity in Rayleigh-faded channel. By using numerical integration, we calculated the total frame capture probability vs. the number of contending stations and capture ratio, and then we analytically evaluated the network saturation throughput of a single IEEE 802.11 BSS operating in Basic or RTS/CTS access methods.

In this paper, we use a discrete Markov chain model to analyze the throughput-delay performance of the CSMA/CA protocol in WLANs under the influence of the capture effect. The stations are power-controlled, so we consider the possibility of frame capture of only a single station which is exposed to Rayleigh fading. By sampling the backoff deferral interval with fixed probability p (estimated in saturation conditions), we were able to determine the upper theoretical limits of channel utilization and mean delay for different traffic loads and different capture ratios.

We analytically analyzed the influence of capture effect over theoretical throughput and delay of a traffic-saturated IEEE 802.11b BSS in infrastructure and ad-hoc configurations. The capture probabilities are derived in presence of the near/far effect and/or Rayleigh fading, i.e. with or without power control mechanisms. The peak aggregate throughput of the BSS operating in Basic access mode is sensitive to the capture, while its impact in RTS/CTS access mode can be disregarded. The frame delay is also insensitive to the capture effect.

Considering imperfect physical and virtual carrier sensing caused by Rayleigh fading, we analyze the probability of a hidden station occurrence, P/sub h/, as a function of the cell radius. We also derive analytical expressions for the probability of frame capture, P/sub capture/, and saturation throughput, assuming that two discrete mean-power levels of transmitted frames at the receiver occur with equal probability. The simulations prove the validity of such a capture model assuming 20 dB difference between the two power levels. As expected, throughput improvement under capture is observed only when the given RTS/CTS "handshake" transmission scheme is not used.

We have compared different performance measures of the two standardized WLANs, IEEE 802.11 and ETSI HIPERLAN type 1. Two essential performance-limitation factors have been considered: the BER of the fading air medium and the number of stations in the ad-hoc network. Regarding the HIPERLAN 1, most of the performance measures are insensitive to the number of stations due to the more sophisticated but complex EY-NPMA MAC protocol. However, its high overhead redundancy causes great throughput degradation as the BER is increased. The relatively simple and low overhead CSMA/CD protocol in the IEEE 802.11 suffers from many collisions at higher loads, producing high sensitiveness to the number of stations.