FAS-enabled cooperative non-orthogonal multiple access (FAS-CoNOMA) systems capture the potential of fluid antenna systems in enhancing network performance. In this system, a base station (BS) transmits a superposition signal to a central user (CU) and a cell-edge user (EU), both equipped with FAS. Specifically, the CU decodes the signal intended for the EU and cooperatively relays it to improve the EU’s communication performance. The EU employs selective combining (SC) or maximum ratio combining (MRC) to receive signals from both the BS and CU. By leveraging the dynamic properties of FAS to improve user differentiation, the CoNOMA system effectively enhances network performance compared to traditional NOMA, OMA, and fixed position antenna (FPA) systems. To address the challenging spatial correlation properties in FAS, we utilize the block-diagonal matrix approximation (BDMA) model to calculate the outage probabilities for both the CU and EU. We then derive upper bound, lower bound, and asymptotic approximation of the outage probabilities to gain deeper insights. Furthermore, we optimize the EU’s outage probability under the CU’s outage constraint and total transmit power limits by adjusting the power allocation coefficient for the CU and the transmit powers for both the BS and CU. To simplify the optimization process, we reduce the number of variables and apply the alternating optimization (AO) algorithm to break down the problem into two sub-problems. Each sub-problem is solved using the bisection search method and gradient descent algorithm (GDA). Simulation results demonstrate that FAS significantly improves outage performance, especially for the EU, and that CoNOMA notably captures the potential of FAS beyond NOMA and OMA, offering a promising solution for future wireless networks.

This paper investigates a design framework for sparse fluid antenna systems (FAS) enabling high-performance direction-of-arrival (DOA) estimation, particularly in challenging millimeter-wave (mmWave) environments. By ingeniously harnessing the mobility of fluid antenna (FA) elements, the proposed architectures achieve an extended range of spatial degrees of freedom (DoFs) compared to conventional fixed-position antenna (FPA) arrays. This innovation not only facilitates the seamless application of super-resolution DOA estimators but also enables robust DOA estimation, accurately localizing more sources than the number of physical antenna elements. We introduce two bespoke FA array structures and mobility strategies tailored to scenarios with aligned and misaligned received signals, respectively, demonstrating a hardware-driven approach to overcoming complexities typically addressed by intricate algorithms. A key contribution is a light-of-sight (LoS)-centric, closed-form DOA estimator, which first employs an eigenvalue-ratio test for precise LoS path number detection, followed by a polynomial root-finding procedure. This method distinctly showcases the unique advantages of FAS by simplifying the estimation process while enhancing accuracy. Numerical results compellingly verify that the proposed FA array designs and estimation techniques yield an extended DoFs range, deliver superior DOA accuracy, and maintain robustness across diverse signal conditions.

Fluid antenna systems (FAS) enable dynamic antenna positioning, offering new opportunities to enhance integrated sensing and communication (ISAC) performance. However, existing studies primarily focus on communication enhancement or single-target sensing, leaving multi-target scenarios underexplored. Additionally, the joint optimization of beamforming and antenna positions poses a highly non-convex problem, with traditional methods becoming impractical as the number of fluid antennas increases. To address these challenges, this letter proposes a block coordinate descent (BCD) framework integrated with a deep reinforcement learning (DRL)-based approach for intelligent antenna positioning. By leveraging the deep deterministic policy gradient (DDPG) algorithm, the proposed framework efficiently balances sensing and communication performance. Simulation results demonstrate the scalability and effectiveness of the proposed approach. Unlike traditional optimization approaches that suffer from exponential complexity growth, our DRL-based method achieves real-time decision-making with superior scalability for complex multi-target scenarios while maintaining computational efficiency.

In recent years, periocular recognition has been developed as a valuable biometric identification approach, especially in wild environments (for example, masked faces due to COVID-19 pandemic) where facial recognition may not be applicable. This paper presents a new deep periocular recognition framework called attribute-based deep periocular recognition (ADPR), which predicts soft biometrics and incorporates the prediction into a periocular recognition algorithm to determine identity from periocular images with high accuracy. We propose an end-to-end framework, which uses several shared convolutional neural network (CNN) layers (a common network) whose output feeds two separate dedicated branches (modality dedicated layers); the first branch classifies periocular images while the second branch predicts soft biometrics. Next, the features from these two branches are fused together for a final periocular recognition. The proposed method is different from existing methods as it not only uses a shared CNN feature space to train these two tasks jointly, but it also fuses predicted soft biometric features with the periocular features in the training step to improve the overall periocular recognition performance. Our proposed model is extensively evaluated using four different publicly available datasets. Experimental results indicate that our soft biometric based periocular recognition approach outperforms other state-of-the-art methods for periocular recognition in wild environments.

We consider the problem of relay selection in mmWave networks with wireless energy transfer, assuming a two-hop decode-and-forward relaying. We incorporate the energy stored at each node in relay selection decision. We propose a centralized algorithm that maximizes an objective function with linear complexity. The results show that incorporating the energy at nodes in relay selection decision increases the number of packets successfully received over network lifetime.

When compared to unimodal systems, multimodal biometric systems have several advantages, including lower error rate, higher accuracy, and larger population coverage. However, multimodal systems have an increased demand for integrity and privacy because they must store multiple biometric traits associated with each user. In this paper, we present a deep learning framework for feature-level fusion that generates a secure multimodal template from each user's face and iris biometrics. We integrate a deep hashing (binarization) technique into the fusion architecture to generate a robust binary multimodal shared latent representation. Further, we employ a hybrid secure architecture by combining cancelable biometrics with secure sketch techniques and integrate it with a deep hashing framework, which makes it computationally prohibitive to forge a combination of multiple biometrics that passes the authentication. The efficacy of the proposed approach is shown using a multimodal database of face and iris and it is observed that the matching performance is improved due to the fusion of multiple biometrics. Furthermore, the proposed approach also provides cancelability and unlinkability of the templates along with improved privacy of the biometric data. Additionally, we also test the proposed hashing function for an image retrieval application using a benchmark dataset. The main goal of this paper is to develop a method for integrating multimodal fusion, deep hashing, and biometric security, with an emphasis on structural data from modalities like face and iris. The proposed approach is in no way a general biometrics security framework that can be applied to all biometrics modalities, as further research is needed to extend the proposed framework to other unconstrained biometric modalities.

In recent years, due to the emergence of deep learning, face recognition has achieved exceptional success. However, many of these deep face recognition models perform relatively poorly in handling profile faces compared to frontal faces. The major reason for this poor performance is that it is inherently difficult to learn large pose invariant deep representations that are useful for profile face recognition. In this paper, we hypothesize that the profile face domain possesses a gradual connection with the frontal face domain in the deep feature space. We look to exploit this connection by projecting the profile faces and frontal faces into a common latent space and perform verification or retrieval in the latent domain. We leverage a coupled generative adversarial network (cpGAN) structure to find the hidden relationship between the profile and frontal images in a latent common embedding subspace. Specifically, the cp-GAN framework consists of two GAN-based sub-networks, one dedicated to the frontal domain and the other dedicated to the profile domain. Each sub-network tends to find a projection that maximizes the pair-wise correlation between two feature domains in a common embedding feature subspace. The efficacy of our approach compared with the state-of-the-art is demonstrated using the CFp, CMU Multi-PIE, IJB-A, and IJB-C datasets.

In this paper, we propose a novel multimodal deep hashing neural decoder (MDHND) architecture, which integrates a deep hashing framework with a neural network decoder (NND) to create an effective multibiometric authentication system. The MDHND consists of two separate modules: a multimodal deep hashing (MDH) module, which is used for feature-level fusion and binarization of multiple biometrics, and a neural network decoder (NND) module, which is used to refine the intermediate binary codes generated by the MDH and compensate for the difference between enrollment and probe biometrics (variations in pose, illumination, etc.). Use of NND helps to improve the performance of the overall multimodal authentication system. The MDHND framework is trained in 3 steps using joint optimization of the two modules. In Step 1, the MDH parameters are trained and learned to generate a shared multimodal latent code; in Step 2, the latent codes from Step 1 are passed through a conventional error-correcting code (ECC) decoder to generate the ground truth to train a neural network decoder (NND); in Step 3, the NND decoder is trained using the ground truth from Step 2 and the MDH and NND are jointly optimized. Experimental results on a standard multimodal dataset demonstrate the superiority of our method relative to other current multimodal authentication systems.

Future wireless networks are expected to adopt many different network technologies and architectures that promise to greatly enhance data rate and provide ubiquitous coverage for end users, all while enabling higher spectral efficiency. These benefits come with an increased risk of co-channel interference and possible correlation in the aggregated interference, which impacts the communication performance. This paper introduces a mathematical framework to quantify the spatial correlation of the interference in finite and clustered wireless networks with signals subject to Rayleigh fading. Expressions are derived for the correlation coefficient of the outage probability when the interferers are located according to some of the most common point processes used in the literature to model the spatial distribution of the devices: binomial point process (i.e., relay networks with fixed users), Poisson point process (i.e., heterogeneous cellular networks), and Thomas point process (i.e., device-to-device networks).

A noncoherent two-way relaying system is developed using physical-layer network coding for improved throughput over conventional relaying in a fading channel. Energy-efficient noncoherent operation is achieved using multitone frequency shift keying (FSK). A novel soft-output demodulator is developed for the relay, and corresponding achievable exchange rates are found for Rayleigh fading and AWGN channels. Bit-error rate performance approaching the achievable rate is realized using a capacity-approaching channel code and a receiver architecture, which iterates between demodulation and channel decoding. Iterative decoding is performed feeding information back from the channel decoder to the demodulator. In addition, the error-rate performance is made to approach the achievable rate more closely by optimizing LDPC codes for this system. The energy efficiency improvement obtained by increasing the modulation order is more dramatic for the proposed physical-layer network coding scheme than it is for a conventional point-to-point system. Using optimized LDPC codes, the bit-error rate performance is improved by as much as 1.1 dB over a widely known standardized LDPC code, and comes to within 0.7 dB of the limit corresponding to the achievable rate. Throughout this paper, the performance for physical-layer network coding is compared with conventional network coding. When noncoherent FSK is used, physical-layer network coding enables higher achievable rates, and conventional network coding exhibits better energy efficiency at low rates.

Millimeter-wave systems are characterized by the use of highly directional antennas and the presence of blockages, which significantly alter the path-loss and small-scale fading parameters. The received power of each interferer depends on the direction it points and whether it is line-of- sight (LOS), non-LOS (i.e., partially blocked), or completely blocked. While interferers that are sufficiently far away will almost certainly be completely blocked, a finite number of interferers in close proximity will be subject to random partial blockages. Previous attempts to characterize mmWave networks have made the simplifying assumption that all interferers within some radius, called the LOS ball, are unblocked, while interferers beyond that radius are non-LOS. However, compared to simulation results, the LOS ball assumption tends to overestimate outage. In this paper, we present an accurate yet tractable analysis of finite mmWave networks that dispenses with the LOS ball assumption. In the analysis, each interferer has a distribution that is selected randomly from several possibilities, each representing different blockage and directivity states. First, the exact outage probability is found for a finite network with interferers in fixed locations. Then, the spatially averaged outage probability is found by averaging over the interferer locations. While the focus is on device-to-device networks, the analysis is general enough to find applications outside of the present mmWave framework.

Analog network coding (ANC) is a throughput increasing technique for the two-way relay channel (TWRC) whereby two end nodes transmit simultaneously to a relay at the same time and band, followed by the relay broadcasting the received sum of signals to the end nodes. Coherent reception under ANC is challenging due to requiring oscillator synchronization for all nodes, a problem further exacerbated by Doppler shift. This work develops a noncoherent M-ary frequency-shift keyed (FSK) demodulator implementing ANC. The demodulator produces soft outputs suitable for use with capacity-approaching channel codes and supports information feedback from the channel decoder. A unique aspect of the formulation is the presence of an infinite summation in the received symbol probability density function. Detection and channel decoding succeed when the truncated summation contains a sufficient number of terms. Bit error rate performance is investigated by Monte Carlo simulation, considering modulation orders two, four and eight, channel coded and uncoded operation, and with and without information feedback from decoder to demodulator. The channel code considered for simulation is the LDPC code defined by the DVB-S2 standard. To our knowledge this work is the first to develop a noncoherent soft-output demodulator for ANC.

Fifth-generation (5G) cellular networks are expected to exhibit at least three primary physical-layer differences relative to fourth-generation (4G) ones: millimeter-wave propagation, massive antenna arrays, and densification of base stations. As in 4G systems, such as LTE, 5G systems are likely to continue to use single-carrier frequency-division multiple-access on the uplink due to its advantageous peak-to-average power ratio. Moreover, 5G systems are likely to use frequency hopping on the uplink to help randomize interference and provide diversity against frequency-selective fading. In this paper, the implications of these and other physical-layer features on uplink performance are assessed using a novel millimeter-wave propagation model featuring distance-dependent parameters that characterize the path-loss, shadowing, and fading. The analysis proceeds by first fixing the location of the mobile devices and finding the performance conditioned on the topology. The spatially averaged performance is then found by averaging with respect to the location of the mobile devices. The analysis allows for the use of actual base-station topologies and the propagation model can leverage empirical millimeter-wave measurements. The benefits of base-station densification, highly directional sectorization, frequency hopping, a large available bandwidth, and a high code rate are illustrated. The minor importance of fractional power control is shown.

Emerging applications involving device-to-device communication among wearable electronics require gigabits per second throughput, which can be achieved by utilizing millimeter-wave (mmWave) frequency bands. When many such communicating devices are indoors in close proximity, such as in a train, car, or airplane cabin, interference can be a serious impairment. This paper uses stochastic geometry to analyze the performance of mmWave networks with a finite number of interferers in a finite network region. Prior work considered either lower carrier frequencies with different antenna and channel assumptions, or a network with an infinite spatial extent. In this paper, human users not only carry potentially interfering devices, but also act to block interfering signals. Using a sequence of simplifying assumptions, accurate expressions for coverage and rate are developed that capture the effects of key antenna characteristics, such as directivity and gain, and are a function of the finite area and number of users. The assumptions are validated through a combination of analysis and simulation. The main conclusions are that mmWave frequencies can provide gigabits per second throughput even with omni-directional transceiver antennas, and larger, more directive antenna arrays give better system performance.

In a centralized or cloud radio access network, certain portions of the digital baseband processing of a group of several radio access points are processed at a central data center. Centralization improves the flexibility, scalability, and utilization of computational assets. However, the performance depends critically on how the limited data processing resources are allocated to serve the needs of the different wireless devices. As the processing load imposed by each device depends on its allocated transmission rate and channel quality, the rate- allocation aspect of the scheduling should take into account the available computing resources. In this paper, two computationally aware schedulers are proposed that have the objective of maximizing the system sum-rate while satisfying a constraint on the offered computational load. The first scheduler optimally allocates resources and is implemented according to a water-filling algorithm. The second scheduler is suboptimal, but uses a simpler and intuitive complexity-cut-off approach. The performance of both schedulers is evaluated using an LTE- compliant system level simulator. It is found that both schedulers avoid outages that are caused by an overflow of computational load (i.e., computational outages) at the cost of a slight loss of sum-rate.

This work considers optimizing LDPC codes in the physical-layer network coded two-way relay channel using noncoherent FSK modulation. The error-rate performance of channel decoding at the relay node during the multiple-access phase was improved through EXIT-based optimization of Tanner graph variable node degree distributions. Codes drawn from the DVB-S2 and WiMAX standards were used as a basis for design and performance comparison. The computational complexity characteristics of the standard codes were preserved in the optimized codes by maintaining the extended irregular repeat-accumulate (eIRA). The relay receiver performance was optimized considering two modulation orders M = {4; 8} using iterative decoding in which the decoder and demodulator refine channel estimates by exchanging information. The code optimization procedure yielded unique optimized codes for each case of modulation order and available channel state information. Performance of the standard and optimized codes were measured using Monte Carlo simulation in the flat Rayleigh fading channel, and error rate improvements up to 1:2 dB are demonstrated depending on system parameters.

Geographic routing protocols greatly reduce the requirements of topology storage and provide flexibility in the accommodation of the dynamic behavior of mobile ad hoc networks. This paper presents performance evaluations and comparisons of two geographic routing protocols and the popular AODV protocol. The tradeoffs among the average path reliabilities, average conditional delays, average conditional numbers of hops, and area spectral efficiencies and the effects of various parameters are illustrated for finite ad hoc networks with randomly placed mobiles. This paper uses a dual method of closed-form analysis and simple simulation that is applicable to most routing protocols and provides a much more realistic performance evaluation than has previously been possible. Some features included in the new analysis are shadowing, exclusion and guard zones, distance-dependent fading, and interference correlation.

In a centralized radio access network (RAN), the signals from multiple radio access points (RAPs) are centrally processed in a data center. A centralized RAN enables advanced interference coordination strategies while leveraging the elastic provisioning of data processing resources. It is particularly well suited for dense deployments, such as within a large building where the RAPs are connected via fiber and where many cells are underutilized. This paper considers the computational requirements of a centralized RAN with the goal of illuminating the benefits of pooling computational resources. A new analytical framework is proposed for quantifying the computational load associated with the centralized processing of uplink signals in the presence of block Rayleigh fading, a distance-dependent path loss, and fractional power control. Several new performance metrics are defined, including the computational outage probability, the outage complexity, the computational gain, the computational diversity, and the complexity-rate tradeoff. The validity of the analytical framework is confirmed by numerically comparing it with a simulator compliant with the 3GPP LTE standard. Using the developed metrics, it is shown that centralizing computing resources provides a higher net throughput per computational resource as compared with local processing.

Centralized radio access network architectures consolidate the baseband operation towards a cloud-based platform, thereby allowing for efficient utilization of computing assets, effective inter-cell coordination, and exploitation of global channel state information. This paper considers the interplay between computational efficiency and data throughput that is fundamental to centralized RAN. It introduces the concept of computational outage in mobile networks, and applies it to the analysis of complexity constrained dense centralized RAN networks. The framework is applied to single-cell and multi-cell scenarios using parameters drawn from the LTE standard. It is found that in computationally limited networks, the effective throughput can be improved by using a computationally aware policy for selecting the modulation and coding scheme, which sacrifices spectral efficiency in order to reduce the computational outage probability. When signals of multiple base stations are processed centrally, a computational diversity benefit emerges, and the benefit grows with increasing user density.

This paper investigates the problem of linear spatial collaboration for distributed estimation in wireless sensor networks. In this context, the sensors share their local noisy (and potentially spatially correlated) observations with each other through error-free, low cost links based on a pattern defined by an adjacency matrix. Each sensor connected to a central entity, known as the fusion center (FC), forms a linear combination of the observations to which it has access and sends the resulting signal to the FC through an orthogonal fading channel. The FC combines these received signals to find the best linear unbiased estimator of the vector of unknown signals observed by individual sensors. The main novelty of this paper is the derivation of an optimal power-allocation scheme in which the coefficients used to form linear combinations of noisy observations at the sensors connected to the FC are optimized. Through this optimization, the total estimation distortion at the FC is minimized, given a constraint on the maximum cumulative transmit power in the entire network. Numerical results show that even with a moderate connectivity across the network, spatial collaboration among sensors significantly reduces the estimation distortion at the FC.

A new accurate analysis is presented for an OFDM-based multicast-broadcast single-frequency network (MBSFN). The topology of the network is modeled by a constrained random spatial model involving a fixed number of base stations placed over a finite area with a minimum separation. The analysis is driven by a new closed-form expression for the conditional outage probability at each location of the network, where the conditioning is with respect to the network realization. The analysis accounts for the diversity combining of signals transmitted by different base stations of a given MBSFN area, and also accounts for the interference caused by the base stations of other MBSFN areas. The analysis features a flexible channel model, accounting for path loss, Nakagami fading, and correlated shadowing. The analysis is used to investigate the influence of the minimum base-station separation and provides insight regarding the optimal size of the MBSFN areas. In order to highlight the percentage of the network that will fail to successfully receive the broadcast, the area below an outage threshold (ABOT) is here used and defined as the fraction of the network that provides an outage probability (averaged over the fading) that meets a threshold.

Multi-cell cooperation (MCC) mitigates intercell interference and improves throughput at the cell edge. This paper considers a cooperative downlink, whereby cell-edge mobiles are served by multiple cooperative base stations. The cooperating base stations transmit identical signals over paths with non-identical path losses, and the receiving mobile performs diversity combining. The analysis in this paper is driven by a new expression for the conditional outage probability when signals arriving over different paths are combined in the presence of noise and interference, where the conditioning is with respect to the network topology and shadowing. The channel model accounts for path loss, shadowing, and Nakagami fading, and the Nakagami fading parameters do not need to be identical for all paths. To study performance over a wide class of network topologies, a random spatial model is adopted, and performance is found by statistically characterizing the rates provided on the downlinks. To model realistic networks, the model requires a minimum separation among base stations. Having adopted a realistic model and an accurate analysis, the paper proceeds to determine performance under several resource-allocation policies and provides insight regarding how the cell edge should be defined.

Digital network coding improves the throughput of the two-way relay channel by allowing multiple sources to transmit simultaneously to the relay. This work considers the development of a relay receiver applying a specific modulation and channel coding technique - turbo-coded noncoherent orthogonal FSK in the two-way relay channel operated with digital network coding. The relay receiver supports any modulation order which is a power of two, and iterative channel decoding with information feedback from decoder to demodulator, using bit interleaved coded modulation with iterative decoding (BICM-ID). The performance of the receiver is investigated in fading channels through error-rate simulations and capacity analysis, and results show an energy efficiency improvement of 0.5-0.9 dB over similar systems which do not utilize BICM-ID.

A new analysis is presented for the direct-sequence code-division multiple access (DS-CDMA) cellular uplink. For a given network topology, closed-form expressions are found for the outage probability and rate of each uplink in the presence of path-dependent Nakagami fading and log-normal shadowing. The topology may be arbitrary or modeled by a random spatial distribution for a fixed number of base stations and mobiles placed over a finite area with the separations among them constrained to exceed a minimum distance. The analysis is more detailed and accurate than existing ones and facilitates the resolution of network design issues, including the influence of the minimum base-station separation, the role of the spreading factor, and the impact of various power-control and rate-control policies. It is shown that once power control is established, the rate can be allocated according to a fixed-rate or variable-rate policy with the objective of either meeting an outage constraint or maximizing throughput. An advantage of the variable-rate policy is that it allows an outage constraint to be enforced on every uplink, whereas the fixed-rate policy can only meet an average outage constraint.

This paper considers ad hoc networks that use the combination of coded continuous-phase frequency-shift keying (CPFSK) and frequency-hopping multiple access. Although CPFSK has a compact spectrum, some of the signal power inevitably splatters into adjacent frequency channels, thereby causing adjacent-channel interference (ACI). The amount of ACI is controlled by setting the fractional in-band power; i.e., the fraction of the signal power that lies within the band of each frequency channel. While this quantity is often selected arbitrarily, a tradeoff is involved in the choice. This paper presents a new analysis of frequency-hopping ad hoc networks that carefully incorporates the effect of ACI. The analysis accounts for the shadowing, Nakagami fading, CPFSK modulation index, code rate, number of frequency channels, fractional in-band power, and spatial distribution of the interfering mobiles. Expressions are presented for both outage probability and transmission capacity. With the objective of maximizing the transmission capacity, the optimal fractional in-band power that should be contained in each frequency channel is identified.

The central issue in direct-sequence code-division multiple-access (DS-CDMA) ad hoc networks is the prevention of a near-far problem. This paper considers two types of guard zones that may be used to control the near-far problem: a fundamental exclusion zone and an additional CSMA guard zone that may be established by the carrier-sense multiple-access (CSMA) protocol. In the exclusion zone, no mobiles are physically present, modeling the minimum physical separation among mobiles that is always present in actual networks. Potentially interfering mobiles beyond a transmitting mobile's exclusion zone, but within its CSMA guard zone, are deactivated by the protocol. This paper provides an analysis of DS-CSMA networks with either or both types of guard zones. A network of finite extent with a finite number of mobiles and uniform clustering as the spatial distribution is modeled. The analysis applies a closed-form expression for the outage probability in the presence of Nakagami fading, conditioned on the network geometry. The tradeoffs between exclusion zones and CSMA guard zones are explored for DS-CDMA and unspread networks. The spreading factor and the guard-zone radius provide design flexibility in achieving specified levels of average outage probability and transmission capacity. The advantage of an exclusion zone over a CSMA guard zone is that since the network is not thinned, the number of active mobiles remains constant, and higher transmission capacities can be achieved.

A new analysis is presented for the direct-sequence code-division multiple access (DS-CDMA) cellular uplink. For a given network topology, closed-form expressions are found for the outage probability and rate of each uplink in the presence of path-dependent Nakagami fading and shadowing. The topology may be arbitrary or modeled by a random spatial distribution with a fixed number of base stations and mobiles placed over a finite area. The analysis is more detailed and accurate than existing ones and facilitates the resolution of network design issues including the influence of the minimum base-station separation, the role of the spreading factor, and the impact of various power-control and rate-control policies. It is shown that once power control is established, the rate can be allocated according to a fixed-rate or variable-rate policy with the objective of either meeting an outage constraint or maximizing throughput. An advantage of variable-rate power control is that it allows an outage constraint to be enforced on every uplink, which is impossible when a fixed rate is used throughout the network.

Physical-layer network coding is a protocol capable of increasing throughput over conventional relaying in the two- way relay channel, but is sensitive to phase and frequency offsets among transmitted signals. Modulation techniques which require no phase synchronization such as noncoherent FSK can compensate for phase offset, however, the relay receiver must still compensate for frequency offset. In this work, a soft- output noncoherent detector for the relay is derived, under the assumption that the source oscillators generating FSK tones lack frequency synchronization. The derived detector is shown through simulation to improve error rate performance over a conventional detector which does not model offset, for offset values on the order of a few hundredths of a fraction of FSK tone spacing.

An ad hoc network with a finite spatial extent and number of nodes or mobiles is analyzed. The mobile locations may be drawn from any spatial distribution, and interference-avoidance protocols or protection against physical collisions among the mobiles may be modeled by placing an exclusion zone around each radio. The channel model accounts for the path loss, Nakagami fading, and shadowing of each received signal. The Nakagami m-parameter can vary among the mobiles, taking any positive value for each of the interference signals and any positive integer value for the desired signal. The analysis is governed by a new exact expression for the outage probability, defined to be the probability that the signal-to-interference-and-noise ratio (SINR) drops below a threshold, and is conditioned on the network geometry and shadowing factors, which have dynamics over much slower timescales than the fading. By averaging over many network and shadowing realizations, the average outage probability and transmission capacity are computed. Using the analysis, many aspects of the network performance are illuminated. For example, one can determine the influence of the choice of spreading factors, the effect of the receiver location within the finite network region, and the impact of both the fading parameters and the attenuation power laws.

An energy-efficient approach is presented for shaping a bit-interleaved low-density parity-check (LDPC) coded amplitude phase-shift keying (APSK) system. A subset of the interleaved bits output by a binary LDPC encoder are passed through a nonlinear shaping encoder whose output is more likely to be a zero than a one. The "shaping" bits are used to select from among a plurality of subconstellations, while the unshaped bits are used to select the symbol within the subconstellation. Because the shaping bits are biased, symbols from lower-energy subconstellations are selected more frequently than those from higher-energy subconstellations. An iterative decoder shares information among the LDPC decoder, APSK demapper, and shaping decoder. Information rates are computed for a discrete set of APSK ring radii and shaping bit probabilities, and the optimal combination of these parameters is identified for the additive white Gaussian noise (AWGN) channel. With the assistance of extrinsic-information transfer (EXIT) charts, the degree distributions of the LDPC code are optimized for use with the shaped APSK constellation. Simulation results show that the combination of shaping, degree-distribution optimization, and iterative decoding can achieve a gain in excess of 1 dB in AWGN at a rate of 3 bits/symbol compared with a system that does not use shaping, uses an unoptimized code from the DVB-S2 standard, and does not iterate between decoder and demodulator.

In order to address the long runtimes required to fully simulate hybrid automatic repeat-request (ARQ) systems, such as high-speed downlink packet access (HSDPA), we propose an information-theoretic approach to accelerate the simulation while still maintaining its fidelity. The approach is based on the concept of information-outage probability, and involves the computation of the mutual information between the code bits transmitted over the channel and the bit log-likelihood ratio coming out of the demodulator. Because the mutual information is measured at the demodulator output, the metric naturally accounts for the parameters of the channel and modulation as well as the length of the codeword. An information outage is declared whenever the measured mutual information is below a threshold (the code rate), and the frequency of simulated information outages provides an accurate prediction of the rate that the turbo decoder fails. Because the prediction is sufficiently accurate, it is possible to run the simulation without actually decoding the turbo code. The result is a 10-30 fold speedup in simulation runtime, while the predicted frame error rate and throughput results are still within 1-2 dB of the values obtained with a more accurate (though lengthier) simulation that decodes each turbo codeword.

This paper considers a technique for shaping a turbo-coded amplitude-phase shift keying (APSK) constellation. After bit-interleaving, a subset of the bits output by a binary turbo encoder are passed through a nonlinear shaping encoder. The bits at the output of the shaping encoder are more likely to be a zero than a one. These "shaping'' bits are interleaved and used to select from among a plurality of subconstellations, while the unshaped bits are used to select the symbol within the subconstellation. Symbols from lower-energy subconstellations are selected more frequently than those from higher-energy subconstellations. Information rates are computed for a variety of APSK and shaping code parameters, in an effort to optimize these parameters. It is found that, in theory, shaping gains of slightly over 0.3 dB may be achieved with 16-APSK and 32-APSK in AWGN, while the gains in ergodic Rayleigh fading are slightly lower. Simulation results show the bit error performance of the constellation-shaping strategy with an actual turbo code. The BER results suggest that gains significantly beyond 0.3 dB may be possible in AWGN when an iterative decoder is used, presumably due to the additional coding gain of the shaping code.

A channel-coded physical-layer network coding strategy is refined for practical operation. The system uses frequency-shift keying (FSK) modulation and operates noncoherently, providing advantages over coherent operation: there are no requirements for perfect power control, phase synchronism, or estimates of carrier-phase offset. In contrast with analog network coding, which relays received analog signals plus noise, the system relays digital network codewords, obtained by digital demodulation and channel decoding at the relay. The emphasis of this paper is on the relay receiver formulation. Closed-form expressions are derived that provide bitwise log-likelihood ratios, which may be passed through a standard error-correction decoder. The role of fading-amplitude estimates is investigated, and an effective fading-amplitude estimator is developed. Simulation results are presented for a Rayleigh block-fading channel, and the influence of block length is explored. An example realization of the proposed system demonstrates a 32.4% throughput improvement compared to a similar system that performs network coding at the link layer. By properly selecting the rates of the channel codes, this benefit may be achieved without requiring an increase in transmit power.

Space-time block codes with orthogonal structures typically provide full-diversity reception and simple receiver processing. However, rate-1 orthogonal codes for complex constellations have not been found for more than two transmit antennas. By using a genetic algorithm, rate-1 space-time block codes that accommodate very simple receiver processing at the cost of reduced diversity are designed in this paper for more than two transmit antennas. Simulation results show that evolved codes combined with efficient outer codes provide better performance over fading channels than minimum-decoding-complexity quasiorthogonal codes at typical operating signal-to-noise ratios. When the fading is more severe than Rayleigh fading, the spectral efficiency is specified, and an efficient outer code is used, evolved codes outperform orthogonal space-time block codes.

A system involving the multisymbol noncoherent reception of coded continuous-phase frequency-shift keying is developed, optimized, and analyzed. Unlike coherent systems, the modulation index of the waveform does not need to be rational with a small denominator, and the oscillator only needs to be stable for the duration of a small block of symbols. The achievable performance over AWGN and Rayleigh block-fading channels is determined by computing the average mutual information, which is the capacity of a channel using the given modulation format and receiver architecture under the constraint of uniformly distributed input symbols. The code rate and modulation index are jointly optimized with respect to average mutual information under a bandwidth constraint. For binary and quaternary signaling, the information-theoretic results are corroborated by bit-error-rate curves generated using a standardized turbo code in conjunction with iterative demodulation and decoding. It is shown that, while more robust than a system with coherent reception, the proposed system offers superior energy efficiency compared with conventional single-symbol noncoherent reception.

The symmetric information rate of a modulation-constrained transmission system is the information-theoretic limit on performance under the assumption that the inputs are independent and uniformly distributed. The symmetric information rate for continuous-phase frequency-shift keying (CPFSK) over an AWGN channel may be estimated by considering the system to be a finite-state Markov channel and executing a BCJR-like algorithm. In this paper, the estimated symmetric information rate is used along with the exact expression for the 99% power bandwidth to determine the information-theoretic tradeoff between energy and spectral efficiency for CPFSK modulation. Using this tradeoff, the code rate and modulation index are jointly optimized for a particular spectral efficiency and alphabet size. Codes are then designed for the optimized system. The codes are comprised of variable nodes (which represent irregular repetition codes), check nodes (which represent single parity-check codes), and an interleaver connecting the variable and check nodes. The degree distributions of the code are optimized from the system's EXIT chart by using linear programming. Additional details of the code design, including labeling and interleaver design, are also discussed. Simulation results show that the optimized coded systems achieve bit error rates within 0.4 dB of the information-theoretic limits at BER = 10-5.

A robust frequency-hopping system with noncoherent detection, iterative turbo decoding and demodulation, and channel estimation is presented. The data modulation is the spectrally compact nonorthogonal continuous-phase frequency-shift keying, which strengthens the frequency-hopping system against multiple-access interference and multitone jamming. An analysis based on information theory provides the optimal values of the modulation index when there is a bandwidth constraint. The channel estimator, which is derived by applying the expectation- maximization algorithm, accommodates both frequency-selective fading and interference. Simulation experiments demonstrate the excellent system performance against both partial-band and multiple-access interference.

A constellation labeling map is the assignment of a bit pattern to each symbol in a signal-set constellation. In a system with iterative decoding and demodulation, the error floor of the bit error rate is highly dependent on the labeling map. A simple class of labeling maps that significantly lower the error floors is presented. Examples show the applications of the proposed mapping to multiple phase-shift keying (MPSK), quadrature amplitude modulation (QAM), and continuous-phase frequency-shift keying (CPFSK). Simulation results indicate that the proposed labeling maps are comparable to or better than other labeling maps in providing a low error floor. A major advantage of the proposed labeling maps is that they are easily generated even when the alphabet size is large.

This paper investigates the impact of inter-user non-orthogonality and asynchronous communication on the information-outage probability performance of multi-user decode-and-forward (DF) cooperative diversity in a code-division multiple-access (CDMA) uplink. Each user in the proposed system transmits its own data towards the base station and also serves as a relay for other users. We assume full-duplex communication so that each user can transmit and receive simultaneously at the same frequency. Each user attempts to decode the messages of a plurality of other users and forwards the superposition of multiple re-encoded and re-spread messages. Our cooperative scheme employs a sub-optimum decorrelating receiver to suppress the multi-user interference at both the base station and the relay-side. We evaluate the information-outage probability performance of the proposed scheme in an underloaded, fully-loaded and overloaded CDMA uplink. We consider combining schemes at the base station where the source information is code combined with the relayed information, while the information from multiple relays is either code combined or diversity combined. Under the system parameters contemplated in this paper, diversity combining of the relayed information is nearly as good as code combining because of the associated probabilities of decoding at the relays. We then examine the effect of using practical modulation formats on the information-outage probability performance of the proposed DF multi-user sharing scheme under diversity combining. We see that the performance loss due to modulation constraints and the use of diversity combining instead of code combining is relatively small.

Cooperative diversity is a form of distributed space- time macrodiversity capable of mitigating the detrimental effects of multipath fading in a wireless network. Prior work on cooperative diversity has focused on the case that all terminals have a single antenna. However, in many practical situations, it is feasible for one or more terminals to be equipped with an array of multiple antennas. This paper investigates the impact of the presence of a single antenna array in a three terminal orthogonal relaying network. The array may be at the source, relay, or destination. The information outage probability under diversity combining is derived, and closed form expressions given wherever possible. The numerical results suggest that it is best to place the array at the destination, and that it is better to place the array at the relay than at the source.

A robust frequency-hopping system with noncoherent detection, iterative turbo decoding and demodulation, and channel estimation is presented. The data modulation is the spectrally compact nonorthogonal continuous-phase frequency-shift keying, which strengthens the frequency-hopping system against multiple- access interference and multitone jamming. An analysis based on information theory provides the optimal values of the modulation index when there is a bandwidth constraint. The channel estimator, which is derived by applying the expectation-maximization algorithm, accommodates both frequency-selective fading and interference. Simulation experiments demonstrate the excellent system performance against partial-band interference.

This paper presents a methodology for determining the capacity of coherently detected continuous- phase frequency shift keying (CPFSK) modulation under the constraints of binary coding and an ergodic channel. Building upon the capacity results, the coded CPFSK parameters of code rate, alphabet size, and modulation index are jointly optimized by using capacity as a cost function. From this optimization, it is possible to determine the minimum epsivb/No required for bit-interleaved coded CPFSK to achieve an arbitrarily low error rate as a function of spectral efficiency. Results showing this minimum epsivb/No are presented for a range of spectral efficiencies and several alphabet sizes in an AWGN channel.

An iterative noncoherent receiver is developed for bit-interleaved coded orthogonal multiple frequency-shift keying (FSK) over block fading channels. The receiver uses the Expectation Maximization (EM) algorithm to jointly estimate the received amplitude and noise spectral density of each block. The received symbols and their corresponding channel estimates are passed through a soft-output demapper, deinterleaved, and decoded. Soft-outputs from the decoder are passed back to the channel estimator and demapper to refine estimates of the channel and bit likelihoods, respectively. Several techniques for reducing the estimator's complexity are discussed, and the performance is assessed through simulation.

A union bound on the performance of bit interleaved coded orthogonal modulation (BICOM) is presented. The bound is derived by representing the system as an equivalent serially concatenated convolutional code (SCCC). Differential precoding prior to orthogonal modulation is shown to provide an interleaver gain, and termination of the differential precoder with a tail bit is taken into account in the bound. The bounds are evaluated for both coherent and noncoherent detection in AWGN and fully interleaved Rayleigh fading and shown to be tight by comparison against simulation results

This paper investigates a capacity-based approach to parameter optimization for energy and bandwidth efficient communication systems. In particular, non-coherently detected, bit-interleaved coded, M-ary Gaussian frequency shift keying (GFSK) is considered. Non-coherent detection is accomplished using a sequential, soft-output (SO), soft-decision differential phase detector (SDDPD). The capacity of the proposed system under modulation, channel and detector design constraints is calculated. For a wide range of spectral efficiencies, the most energy efficient combination of GFSK parameters and code rates is identified using information theoretic bounds on reliable signaling. The information outage probabilities under modulation and detector design constraints are calculated for the block fading channel. Select results reveal that the capacity-based approach also helps in identifying the combination of modulation parameters and code rates with the lowest outage probabilities in block fading

This paper investigates an energy efficient, noncoherent system design for Bluetooth. We propose using a soft decision differential phase detector (SDDPD) along with the (15, 10) shortened Hamming channel (SHC) code. Unlike previous work, where SDDPD with ML detection gives hard estimates of the modulated bits, we develop a sequential, soft-in/soft-out (SISO) SDDPD which exchanges soft information with the channel decoder. The proposed soft decision based receiver gives significant improvements in energy efficiency and throughput over the benchmark Bluetooth receiver using limiter discriminator integrator (LDI) detection with hard decision channel decoding. It also outperforms SDDPD with ML detection followed by hard decision channel decoding. Additional gains are obtained by extending our proposed system to perform bit-interleaved coded modulation (BICM). The effect of bit-interleaving on the performance of LDI detection and SDDPD with hard decision decoding is studied. The capacity of our system under BICM is evaluated and extrinsic information transfer (EXIT) chart analysis is used to analyze convergence behavior of the proposed receiver

Cooperative diversity, which employs multiple nodes for the simultaneous relaying of a given packet in wireless ad hoc networks, has been shown to be an effective means of improving diversity, and, hence, mitigating the detrimental effects of multipath fading. However, in previously proposed cooperative diversity schemes, it has been assumed that coordination among the relays allows for accurate symbol-level timing synchronization at the destination and orthogonal channel allocation, which can be quite costly in terms of signaling overhead in mobile ad hoc networks, which are often defined by their lack of a fixed infrastructure and the difficulty of centralized control. In this paper, cooperative diversity schemes are considered that do not require symbol-level timing synchronization or orthogonal channelization between the relays employed. In the process, a novel minimum mean-squared error (MMSE) receiver is designed for combining disparate inputs in the multiple-relay channel. Outage probability calculations and simulation results demonstrate the not unexpected significant performance gains of the proposed schemes over single-hop transmission, and, more importantly, demonstrate performance comparable to schemes requiring accurate symbol-level synchronization and orthogonal channelization.

The paper proposes and analyzes a packet-level macrodiversity combining scheme for improving the performance of the uplink of IEEE 802.11a networks operating in infrastructure mode. By receiving the mobile station's signal at three equidistant access points, a gain of between 6 dB and 8 dB is observed at a packet error rate of 10/sup -2/ when the receiver is able to estimate the channel perfectly. A similar gain is found when the receiver operates without channel state information and must estimate the channel. Results are presented showing the improvement in packet error rate and throughput for several different mobile station positions. It is found that the gain due to macrodiversity combining diminishes as the mobile station moves closer to a single access point.

This paper considers bit-interleaved coded modulation (BICM) with a turbo channel code and M-ary orthogonal modulation. The BICM signal is iteratively demodulated and decoded in a noncoherent fashion. A soft demodulator suitable for noncoherent orthogonal modulation is presented, and the convergence of the iterative receiver is analyzed through extrinsic information transfer charts. The demodulator can work either with or without fading amplitude estimates. Extensive simulation results are presented for the well-known cdma-2000 turbo code, and the results are compared with the corresponding channel capacities, which are computed using a Monte Carlo technique. The results indicate gains of up to 1 dB relative to noniterative BICM can be achieved with the iterative receiver.

Wireless networks contain an inherent distributed spatial diversity that can be exploited by the use of relaying. Relay networks take advantage of the broadcast-oriented nature of radio and require node-based, rather than link-based protocols. Prior work on relay networks has studied performance limits either with unrealistic assumptions, complicated protocols, or only a single relay. In this paper, a practical approach to networks comprising multiple relays operating over orthogonal time slots is proposed based on a generalization of hybrid-automatic repeat request (ARQ). In contrast with conventional hybrid-ARQ, retransmitted packets do not need to come from the original source radio but could instead be sent by relays that overhear the transmission. An information theoretic framework is exposed that establishes the performance limits of such systems in a block fading environment, and numerical results are presented for some representative topologies and protocols. The results indicate a significant improvement in the energy-latency tradeoff when compared with conventional multihop protocols implemented as a cascade of point-to-point links.

Turbo codes are sensitive to both (timing) synchronization errors and signal-to-noise ratio (SNR) mismatch. Since turbo codes are intended to work in environments with very low SNR, conventional synchronization methods often fail. This paper investigates blind symbol-timing synchronization and SNR estimation based on oversampled data frames. The technique is particularly suitable for low-rate turbo codes operating in additive white Gaussian noise at low SNR and modest data-transfer rates, as in deep space, satellite, fixed wireless, or wireline communications. In accordance with the turbo principle, intermediate decoding results are fed back to the estimator, thereby facilitating decision-directed estimation. The analytical and simulated results show that with three or more samples per symbol and raised cosine-rolloff pulse shaping, performance approaches that of systems with perfect timing and SNR knowledge at the receiver.

This paper proposes and analyzes a new cross-layer protocol for ad hoc and sensor networks that unifies the concepts of geographic random forwarding (GeRaF) and hybrid-ARQ. The protocol is given the descriptive name hybrid ARQ-based intra-cluster geographically informed relaying (HARBINGER). Like GeRaF, HARBINGER assumes that each node knows its own position and that messages are addressed by location. As is common for sensor networks, the nodes cycle on-and-off according to a sleep schedule. Unlike GeRaF, which returns to the initial transmission state If no active node is within range, the nodes in HARBINGER combine transmissions thereby achieving an additional time-diversity benefit. With HARBINGER, a lower density of active nodes achieves almost the same delay and energy efficiency as GeRaF, implying that a lower duty cycle sleep schedule could be used to prolong the useful lifetime of the network. The paper gives detailed analysis of a version of the protocol (Fast-HARBINGER) whereby the sleep states of the network are synchronized with the data packet transmission rate of the network.

Turbo codes are sensitive to both (timing) synchronization errors and signal-to-noise ratio (SNR) mismatch. Since turbo codes are intended to be deployed in environments with very low SNR, conventional synchronization methods often fail. This paper introduces a solution for jointly estimating the SNR and achieving timing synchronization based on the statistics of the received signal. Simulation results show only a small loss in coding gain relative to perfect timing and SNR estimation while requiring only slightly more complexity and latency.

In this paper, we suggest a postdetection strategy for combining observations made at multiple single-antenna access points in a wireless LAN, using Bluetooth as a running example. The result is a diversity gain that is analogous to a conventional antenna array, only the array is distributed throughout the network. In quasi-static Rayleigh fading and for a target packet error probability of 10/sup -2/, the proposed technique allows a significant (18 dB) reduction in mobile transmit power when as many as six equidistant access points are used. This gain is reduced as the mobile moves closer to any one of the access points or fewer access points are used. In addition, the role of ARQ and the impact of this reception technique on throughput is investigated.

Embedded networks of sensors and actuators must operate at extremely low power and use inexpensive single-antenna transceivers. The economics of such systems preclude the use of complex signal processing or antenna arrays at any one device. However, the same economics allows the coverage area to be blanketed with a high density of devices, which results in a rich spatial diversity. This spatial diversity can be exploited by forming a virtual antenna array, which combines observations made at multiple receivers. This paper illustrates the potential transmit energy savings that are possible by using such macrodiversity-combining approaches by the analysis and simulation of an idealized system. It concludes by discussing practical issues that must be considered before integrating macrodiversity-combining strategies into actual embedded networks.

This paper addresses some critical implementation issues involved in the development of a turbo decoder, using the UMTS specification as a concrete example. The assumption is that the decoder is to be implemented in software rather than hardware, and thus a variable number of decoder iterations is not only possible, but desirable. Three twists on the decoding algorithm am proposed: (1) a linear approximation of the correction function used by the max* operator which reduces complexity with only a negligible loss in BER performance; (2) a method for normalizing the backward recursion which yields a 12.5% savings in memory usage; and (3) a simple method for halting the decoder iterations based only on the log-likelihood ratios.

Analytical expressions for the throughput (in kbit/s) as a function of channel symbol signal-to-noise ratio (E/sub s//N/sub 0/) are derived for the six Bluetooth ACL packets that use automatic repeat request (ARQ). The analysis is exact under the assumptions that the outer CRC code provides perfect error detection and that the channel remains stationary for the duration of each packet. Using an expression for noncoherent correlated (h<0.5) full response FSK signals, numerical results are provided for AWGN and quasi-static Rayleigh fading channels. These curves are an appropriate benchmark against which practical demodulators and custom error control techniques may be compared.

A soft-input soft-output (SISO) multiuser detector (MUD) suitable for inclusion in iterative processing architectures is presented and applied to the detection of the coded time division multiple access (TDMA) cellular uplink. A SISO-MUD processor is located at each base station in the network, and adjacent base stations share information concerning the mobiles they serve. Because the MUD outputs are soft, they are suitable for postdetection macrodiversity combining. The combined signals are then passed to a SISO forward error correction (FEC) decoder, and the soft outputs are fed back to the multiuser detectors. Processing continues in an iterative fashion in accordance with the turbo principle. Simulation results are presented that indicate that use of such a scheme enables cellular systems to be overloaded with more than just one cochannel user per cell at the price of a minimal loss in signal-to-noise ratio (SNR). The proposed implementation assumes the availability of both perfect channel state information and a high capacity backhaul.

A method for coherently detecting and decoding turbo-coded binary phase shift keying (BPSK) signals transmitted over frequency-flat fading channels is discussed. Estimates of the complex channel gain and variance of the additive noise are derived first from known pilot symbols and an estimation filter. After each iteration of turbo decoding, the channel estimates are refined using information fed back from the decoder. Both hard-decision and soft-decision feedback are considered and compared with three baseline turbo-coded systems: (1) a BPSK system that has perfect channel estimates; (2) a system that uses differential phase shift keying and hence needs no estimates; and (3) a system that performs channel estimation using pilot symbols but has no feedback path from decoder to estimator. Performance can be further improved by borrowing channel estimates from the previously decoded frame. Simulation results show the influence of pilot symbol spacing, estimation filter size and type, and fade rate. Performance within 0.49 and 1.16 dB of turbo-coded BPSK with perfect coherent detection is observed at a bit-error rate of 10/sup -4/ for normalized fade rates of f/sub d/T/sub s/=0.005 and f/sub d/T/sub s/=0.02, respectively.

We propose a new type-II code-combining hybrid automatic repeat request (ARQ) strategy that uses both parallel and serial concatenated convolutional codes (PCCCs and SCCCs). The ARQ technique is based on the observation that a PCCC can be represented by an equivalent punctured SCCC. The initial transmission is the PCCC. When a retransmission is requested, the system only transmits the additional parity bits that, together with the previously transmitted PCCC bits, constitute the related SCCC. Simulation results show remarkable power efficiency and throughput efficiency when using the proposed system.

A soft-input, soft-output multiuser detection algorithm based on the log-MAP algorithm is presented for asynchronous multiple access networks. The algorithm is applied to a convolutionally coded direct sequence code division multiple access (DS-CDMA) system. The proposed receiver architecture consists of a log-MAP multiuser detector followed by a log-MAP channel decoder. Information is fed back from the channel decoder to the multiuser detector, and processing proceeds in an iterative fashion analogous to the decoding of turbo codes. Simulation results show that the performance of a heavily loaded power controlled DS-CDMA system approaches the single-user bound for both additive white Gaussian noise (AWGN) and fully-interleaved Rayleigh flat-fading channels.