The recently proposed index modulation-based up-link orthogonal frequency division multiple access (IM-OFDMA) scheme has outperformed the conventional schemes in terms of spectral efficiency and error performance. However, the induced computational complexity at the receiver forms a bottleneck in real-time implementation due to the joint detection of all users. In this paper, based on deep learning principles, a convolutional neural network (CNN)-based signal detector is proposed for data detection in IM-OFDMA systems instead of the optimum Maximum Likelihood (ML) detector. A CNN-based detector is constructed with the created dataset of the IM-OFDMA transmission by offline training. Then, the convolutional neural network (CNN)-based detector is directly applied to the IM-OFMDA communication scheme to detect the transmitted signal by treating the received signal and channel state information (CSI) as inputs. The proposed CNN-based detector is able to reduce the order of the computational complexity from O(n2n) to O(n2) as compared to the ML detector with a slight impact on the error performance.

Physical Layer Network Coding (PLNC) allows the exchange of data packets in a two-way relay channel with higher spectral efficiency than traditional network coding (NC). This paper considers the effect that the choice of pulse shape has on the amount of time synchronism needed between the two users in a PLNC system. It is shown that the choice of pulse shape effects both the performance and the complexity of the receiver at the relay node. Furthermore, it is seen that regardless of what pulse shape is used, there must be some level of time synchronization between the two users to avoid substantial performance loss.

This article investigates the usefulness of using two sensors and a blind-separation algorithm in reducing the effect of surface noise in downhole communication systems. The acoustic channel provides a challenging environment for the acoustic waves that propagate from the downhole to the surface of an oil or gas well. As a result, acoustic waves experience a noticeable attenuation before reaching the surface of the well. Consequently, surface noise, which is generated by the surface tools, dominates the performance of acoustic downhole communication systems that have the receiver unit close to the well surface. The application of a two-receiver noise cancellation algorithm is investigated. The article also describes a testbed that was designed to study the effectiveness of the proposed algorithm in reducing the impact of the surface noise. The communication system was built using two speakers, five connected segments of 7 inch production pipes, and two microphones. One of the speakers was used to transmit a noise-like signal in order to simulate the surface noise. The noise cancellation algorithm was applied to the outputs of the two microphones, and the quality of the acoustic signals is investigated after applying the noise cancellation solution. Results of this work emphasize the usefulness of the proposed solution in enhancing the performance of the acoustic downhole communication systems.

Symmetrical multilevel diversity coding (SMDC) is a network compression problem introduced by Roche (1992) and Yeung (1995). In this setting, a simple separate encoding strategy known as superposition coding was shown to be optimal in terms of achieving the minimum sum rate (Roche-Yeung-Hau 1997) and the entire admissible rate region (Yeung-Zhang 1999) of the general problem. This paper considers a natural generalization of SMDC to the secure communication setting with an additional eavesdropper. It is required that all sources need to be kept perfectly secret from the eavesdropper as long as the number of encoder outputs available at the eavesdropper is no more than a given threshold. First, the problem of encoding individual sources is studied. A precise characterization of the entire admissible rate region is established via a connection to the problem of ramp-type secret sharing (Yamamoto 1985 and Blakley-Meadows 1985) and utilizing some basic polyhedral structure of the admissible rate region. Building on this result, it is then shown that superposition coding remains optimal in terms of achieving the minimum sum rate for the general secure SMDC problem.

In this paper, we present an algorithm for joint decoding of the modulo-2 sum of the bits transmitted from two unsynchronized transmitters using Physical Layer Network Coding (PLNC). We address the problems that arise when the boundaries of the signals do not align with each other and when the channel parameters are slowly varying and are not known to the receiver at the relay node. Our approach first estimates jointly the timing and fading gains of both the signals, and uses a state-based Viterbi decoding scheme that takes into account the timing offsets between the interfering signals. We also track the amplitude and phase of the channel which may be slowly varying. Simulation results demonstrate the sensitivity of the detection performance at the relay node to the relative offset of the timings of the two user's signals as well as the advantage of our algorithm over previously published algorithms.

A wireless communication system where a common base station (BS) is scheduled to transmit information to mobile users on a time division (TD) basis, under quality of service (QoS) constraints is considered. The quality of service requirement is specified in terms of the asymptotic decay rate of the buffer occupancy. The effective capacity of this system is characterized and the optimal scheduling scheme which achieves the boundary of the effective capacity region is found. It is further shown that the optimal scheduling scheme without quality of service constraints and the opportunistic scheduling schemes fall out as special cases of a more generalized class of scheduling schemes derived in this paper.

A wireless communication system where a common base station is scheduled to transmit information to multiple mobile users on a time division (TD) basis is considered. The capacity region of this system is found for the two user case and the optimal scheduling scheme is proposed which achieves the boundary of the capacity region. Furthermore, the optimal scheduling scheme is found for the case when the remote mobile users can perform downlink cooperation and the achievable rate region of the downlink cooperative system is characterized. Finally, a simple iterative algorithm is proposed for finding the resource allocation parameters and the scheduling scheme for the cooperative system.

In this paper, a multiuser communication system in which wireless users cooperate to transmit information to a base station is considered. The proposed scheme can significantly enlarge the achievable rate region, provided that the wireless connections between pairs of cooperating users are stronger than the connection from every user to the base station. The gains in transmission rate remain substantial even when the channel phase information is only available at the receivers, not at the transmitters. In the proposed scheme, a transmission period is divided into two time intervals. During the first time interval, wireless users send data to the base station and to the neighboring users simultaneously using a broadcast channel paradigm. During the second time interval, the users cooperate to transmit information to the base station. The achievable rate region corresponding to this paradigm is characterized under a random phase channel model for a two-user system. Results are then generalized to a multiple-user scenario. For fixed system parameters, the achievable rate region is strictly larger than that of the traditional multiple-access channel, thereby allowing a fair distribution of the wireless resources among users. Numerical analysis suggests that cooperating with a single partner is enough to achieve most of the benefits associated with cooperation.

Wireless network analysis and simulations rely on simple yet accurate physical layer models. In this paper, a new four-state Markov model is proposed for the flat Rayleigh fading channel by approximating the good/bad frame run length distributions with a mixture of geometric distributions. The transition probabilities in the four-state model can be analytically established without having to run extensive physical layer simulations. This greatly simplifies the process of model establishment. Physical layer good/bad frame run length distributions are compared and it is shown that the four-state Markov model reasonably approximates the run length distributions. Ns2 simulations are performed and comparisons are made with the popular two-state Markov model and also the finite state Markov chain model.

This paper presents an improved upper bound on the performance of convolutionally coded systems over quasi-static fading channels (QSFC). The bound uses a combination of a classical union bound when the fading channel is in a high signal-to-noise ratio (SNR) state together with a new upper bound for the low SNR state. This new bounding approach is applied to both BPSK convolutional and turbo codes, as well as serially concatenated BPSK convolutional/turbo and space-time block codes. The new analytical technique produces bounds which are usually about 1 dB tighter than existing bounds. Finally, based on the proposed bound, we introduce an improved design criterion for convolutionally coded systems in slow flat fading channels. Simulation results are included to confirm the improved ability of the proposed criterion to search for convolutional codes with good performance over a QSFC.

In this paper, we propose a frequency domain approach to design a set of spreading sequences for a DS-CDMA system in the presence of frequency selective fading. We show that the number of parameters involved in the design of optimal spreading sequences can be made smaller when they are analyzed in the frequency domain, than when they are analyzed in the time domain. Further, it is shown that for multipath channels, designing a set of spreading sequences in the frequency domain reduces to assigning a spreading sequence for each user with a spectrum whose spectral lines are located at the strongest spectral components of the user's channel's spectrum while keeping the cross correlation at a very low value for spreading sequences with overlapping spectra. We show that by designing spreading sequences in the frequency domain, the performance of some existing algorithms for the design of spreading sequences can be improved and/or the computational complexities can be reduced. We further show that the proposed idea can be extended to the design of spreading sequences where the chips of the spreading sequences are derived from a constant amplitude alphabet

Although there are attractive sequence design algorithms for code division multiple access (CDMA) systems in frequency selective channels, unfortunately in most of these algorithms, the designing has been done with the assumption that the chips of the sequences have unconstrained amplitudes and phases. This increases the required feedback bandwidth and the peak to average power ratio (PAPR) at the transmitter. In this paper, we observe that the spectrum of the spreading sequences designed according to existing algorithms are narrowband and are located at the strongest spectral components of the users' channels. The narrowbandness of spreading sequences allows one to represent each user's sequence with fewer number of parameters and, hence, reduces the required number of feedback bits to represent those sequences. Further, the narrowbandness of the spectra allows one to break the global optimization of the users' cross correlation matrix into simple sub optimal local (grouped) optimizations. In this paper, we capitalize the aforementioned observations to design constrained amplitude sequences, for frequency selective channels

This paper proposes new upper hounds on the performance of space-time trellis codes (STTC) over quasi-static fading channels (QSFC). The proposed bounding approach combines a new low signal-to-noise ratio upper bound together with the traditional union bound. The derived upper bounds show much improved ability to predict the performance of STTC over QSFC, as can be seen from applying them to a published example of STTC.

In this paper, we investigate the development of channel models for wireless network simulations based upon distributions of a representative frame error process. Our approach utilizes the distributions of 1) good frame run lengths and 2) bad frame run lengths as a metric for developing low complexity channel models. Two unique, low complexity flat Rayleigh fading channel approximations are generated and compared against the traditional two-state Markov approximation. We quantify the performance of channel approximations through ns2 network simulations. This work furthers the understanding of channel model development by proposing a metric by which to design and evaluate channel approximations. In addition, this metric is evaluated for the case of flat Rayleigh fading.

This correspondence presents a general approach to upper bounding coded system performance over quasistatic fading channels (QSFC). This approach has the advantage of yielding a closed-form upper bound that converge for all signal-to-noise ratios (SNRs). The proposed approach is used to upper-bound the performance of space-time trellis codes (STTC) over QSFCs. The resulting upper bounds for STTCs are better adapted to the QSFC and present an improvement over worst case pairwise error probability (PEP) analysis used so far. In its second part, this correspondence investigates several ways to reduce the complexity of computing the distance spectrum of STTCs. The combined result obtained from using the new upper bounds and the computed distance spectra are shown to be close to simulated performance for all SNRs.

This paper presents an improved upper bound on the performance of convolutional codes over quasistatic fading channels (QSFC). The bound uses a combination of a classical union bound when the fading channel is in a high signal-to-noise ratio (SNR) state together with a new upper bound for the low SNR state. This new analytical technique produces bounds which are usually about 1 dB tighter than existing bounds. The new approach is applied to both convolutional and turbo codes as well as serially concatenated convolutional/turbo and space time block codes.

We propose a novel technique to design a set of spreading sequences for the uplink of a DS-CDMA system in the presence of frequency selective fading. The focus is on the case when channel state information (CSI) is available at the transmitter and when the chips in the spreading sequences are confined to be from a finite alphabet We consider two cases: (i) knowledge of the CSI of all the users is available centrally during the design process, such as in a base station; (ii) the design is at the mobile unit when each user only knows the CSI of his/her frequency selective channel. In both cases, we propose novel techniques to design spreading sequences whose spectra are well matched to that of the frequency selective channel. In the former case, a set of short spreading sequences is designed that is both well matched to the frequency selective channels and possesses low cross correlation. In the latter case, the use of long spreading sequences and error control coding is proposed. Results show that the proposed techniques provide significantly improved performance over the use of Gold codes and that they outperform single user performance over an additive white Gaussian noise channel. The designed short spreading sequences are also well suited to be used with low complexity multiuser detectors.

This paper considers the performance analysis of coded OFDM systems over frequency-selective fading channels. It is shown that both the random coding upper bounds and the strong converse lower bounds of a coded OFDM system converge to the channel outage probability for large OFDM block lengths. Thus primary attention is given to the analysis of the outage probability, which is taken as the optimal system performance. Instead of evaluating the outage probability numerically, we provide in this paper a simple analytical close form approximation of the outage probability for a coded OFDM system over frequency-selective quasistatic fading channels. Simulation results of the turbocoded OFDM systems further confirm the approximation of the outage probability.

In this paper, we investigate the performance of partial interference cancellers in forward link DS-CDMA over time-invariant frequency selective fading channels. An analytical tool to find the bit error probability is presented. We also compare the analysis with computer simulation. The results show that the presented approximation can predict the performance at low to moderate signal to noise ratios.

In this paper, two code acquisition schemes are studied for use in conjunction with transmit diversity direct-sequence code-division multiple access (DS-CDMA). One is a training-based single-user maximum-likelihood (SUML) estimator, which can achieve code acquisition very fast at low computational complexity. The other is a blind estimator based on the multiple signal classification (MUSIC) algorithm. Two recently proposed transmit diversity schemes known as orthogonal transmit diversity simulcast (OTD-S) and space-time selective spreading transmit diversity (STSTD) are considered. While the advantages of transmit diversity from the detection standpoint are well known, less is known about how code acquisition performance is affected by the use of transmit diversity. Through the analysis in this paper, it is proven that the SUMI. estimator should give the same performance in both the OTD-S and STSTD schemes in a single-user environment. In a multiple-user environment, simulation results show that the STSTD system offers slightly better code acquisition performance. It is also seen that the SUML estimators provide significantly better code acquisition performance than the MUSIC estimators in either transmit diversity system. From the standpoint of robustness to carrier frequency offset, it Is found that the training-based SUML estimator is very sensitive to frequency offset, while the MUSIC estimator is quite robust. A simple frequency offset estimator to be used in conjunction with the SUML estimator is also proposed and is shown to make the timing estimator quite tolerant of substantial frequency offsets.

This paper proposes effective ways to reduce the complexity of computing the distance spectrum of space-time trellis codes. First the notion of Voronoi neighborhood is extended to space-time trellis codes. Next, only simple error events are used, then a stack algorithm is used to further reduce complexity. The obtained reduced spectra are then used in conjunction with some previously proposed upper bounds to predict the performance of some published codes over quasi-static fading channels. Of interest are published space-time codes for which only tentative performance characterization has been proposed so far. The numerical results show that indeed the performance of these codes only depends on their distance spectrum and that remarkably tight bounds can be obtained.

This paper investigates several upper bounds and their validity for a particular space-time concatenated coding scheme. The system under study is a serial concatenation of a trellis outer code to provide coding gain followed by a space-time block code to insure diversity. Of interest is the performance of the coding scheme over a quasi-static fading channel (QSFC). It is shown in this paper how the classical union bound is divergent for the QSFC and this makes it unsuitable for studying codes and building design criteria. A different upper bound based on Berlekamp's tangential bound (TB) is derived and its suitability for the purpose of analyzing this concatenation scheme is explored. Finally a new upper bound is derived which results in a more stable alternative. The result is a remarkably simple yet very reliable upper bound for space-time coded systems. Simulation results from an example scheme are provided and show that the new bound does indeed give a useful upper bound on the true system performance.

Two code acquisition schemes are studied for use in conjunction with transmit diversity direct sequence code division multiple access (DS-CDMA). One is a training based single user maximum likelihood (SUML) estimator, which can achieve code acquisition very fast at low computation complexity. The other is a blind estimator based on the multiple signal classification (MUSIC) algorithm. Two previously proposed transmit diversity schemes known as orthogonal transmit diversity simulcast (OTD-S) and space time selective transmit diversity (STSTD) are considered. While the advantages of transmit diversity from the detection standpoint are well known; less is known about how code acquisition performance is affected by the use of transmit diversity. Through the analysis, it is proven that the SUML estimator should give the same performance in both the OTD-S and STSTD schemes in a single user environment. In a multiple user environment, simulation results show that the STSTD system offers slightly better code acquisition performance. It is also seen that the SUML estimators provide significantly better code acquisition performance than the MUSIC estimators in either transmit diversity system.

Adaptive linear interference suppression for direct-sequence (DS) code-division multiple access (CDMA) is studied in the presence of time- and frequency-selective fading. Interference suppression is achieved with an adaptive digital filter which spans a single symbol interval. Both decision-directed and blind adaptive algorithms, which do not require a training sequence, are considered. Modifications to least squares adaptive algorithms are presented which are compatible with differential coding and detection. For frequency-selective fading, adaptive algorithms are presented based upon different assumptions concerning knowledge of the desired user's channel. Specifically, the cases considered are as follows: (1) perfect knowledge of the desired channel; (2) knowledge of only the relative path delays; and (3) knowledge of only one delay corresponding to the strongest path. Computer simulation results are presented which compare the performance of these algorithms with the analogous RAKE receivers. These results show that for case (3), even slow fading can cause a significant degradation in performance. Effective use of channel parameters in the adaptive algorithm reduces the sensitivity to fade rate, although moderate to fast fading can significantly compromise the associated performance gain relative to the RAKE receiver.

We present herein a decoupled multiuser acquisition (DEMA) algorithm for code-timing estimation in asynchronous code-division multiple-access (CDMA) communication systems. The DEMA estimator is an asymptotic (for large data samples) maximum-likelihood method that models the channel parameters as deterministic unknowns. By evoking the mild assumption that the transmitted data bits for all users are independently and identically distributed, we show that the multiuser timing estimation problem that usually requires a search over a multidimensional parameter space decouples into a set of noniterative one-dimensional problems. Hence, the proposed algorithm is computationally efficient. DEMA has the desired property that, in the absence of noise, it obtains the exact parameter estimates even with a finite number of data samples which can be heavily correlated. Another important feature of DEMA is that it exploits the structure of the receiver vectors and, therefore, is near-far resistant. Numerical examples are included to demonstrate and compare the performances of DEMA and a few other standard code-timing estimators.

Code division multiplexing (CDM) is proposed as a modulation format for single user communication over highly dispersive fading channels. This modulation format uses well-known spread spectrum technology together with newly developed multiuser detection techniques to produce a modulation and demodulation technique for non-spread data transmission. It is shown that for frequency selective fading channels with large delay spreads, the CDM modulation and demodulation formats developed in this paper represent a superior option to both single carrier modulation formats such as phase shift keying (PSK) as well as multicarrier formats such as orthogonal frequency division multiplexing (OFDM). Hence, CDM is an attractive modulation format for future wireless systems requiring high data rates such as wireless Internet applications and wireless video transmission.

Iterative schemes for demodulating M-ary orthogonal signaling formats in direct-sequence code-division multiple access (DS-CDMA) systems are proposed and compared with the standard matched filter receiver and the standard parallel interference cancellation (PIC) receiver. The considered system is similar to the uplink (reverse link) of an IS-95 system. The channel is assumed to be a time-varying flat Rayleigh-fading channel, and a channel estimation scheme is proposed. The proposed receivers are shown to provide large capacity gains compared with the conventional receiver.

Minimum mean-squared error (MMSE) detection of multicarrier code-division multiple-access (CDMA) signals is investigated. The theoretical performance of two different design strategies for MMSE detection are compared. In one case, the MMSE filters are designed separately for each carrier, while in the other case the optimization of the filters is done jointly. Naturally, the joint optimization produces a better receiver, but the difference in performance is shown to be substantial. The multicarrier CDMA performance is then compared to that of a single-carrier CDMA system on a frequency-selective fading channel. A mechanism is then developed to track the channel fading parameters for all the users' signals so that joint optimization of the receiver filters is possible in a time-varying channel. Simulation results show that the performance of this receiver is close to ideal theoretical results for moderate vehicle speeds. The performance begins to degrade when the normalized Doppler rate is higher than about 1%.

The performance of the minimum mean-squared error (MMSE) receiver for the detection of direct sequence code division multiple access is considered in various fading channel models. Several modifications to the basic MMSE receiver structure which have been previously proposed for use on nonselective fading channels are reviewed and shown to represent different approximations to a single common form. The performance of this general structure is analyzed as well as various extensions suitable for frequency-selective fading channels. Particular attention is given to the performance advantage gained through knowledge of the fading parameters of the various transmission paths of each user's signal. It is shown that having this knowledge is not particularly useful on a flat fading channel unless the loading is very heavy and even then the difference in performance is only minimal. On the other hand, having this knowledge is crucial in a multipath fading channel and the inability to learn the fading channel parameters will lead to substantial degradation in capacity. A heuristic explanation to support this result based on a dimensionality argument is also presented.

In this paper, the overlay of a code-division multiple-access (CDMA) cellular system in a frequency band which is already providing service to a narrow-band cellular system is investigated. The motivation for this study is to demonstrate the potential of CDMA overlay as an efficient method for making a long-term transition from narrow-band cellular to CDMA cellular. For the single cell case, it was shown by Rainbolt and Miller (see IEEE J. Select. Areas Commun., vol.16, p.1756-64, 1998) that interference caused to the narrow-band system by even a lightly-loaded CDMA system is significant, a problem which can be greatly alleviated through the use of notch filtering in the CDMA transmitters to avoid active narrow-band users. Similar conclusions will be reached in this paper for the cellular case. The effects of notching on CDMA performance will be looked at in detail, and it will be shown that the notching is less of a problem than is the narrow-band interference inherent in overlay systems.

A code-division multiple access (CDMA) cellular overlay system is investigated, employing the idea of multicarrier CDMA, which has previously received significant attention as an alternative to traditional single-carrier CDMA. Overlay is pursued here as a means of long-term transition from narrowband cellular to CDMA cellular. A major result of this paper is the demonstration that the use of multicarrier CDMA in a fading channel is particularly beneficial to the narrowband system, as the CDMA users can reduce their transmitted powers as a result of diversity. Another significant conclusion is that the use of transmitter notching in the CDMA system in order to avoid active narrowband users outperforms a strategy in which a narrowband user is avoided by simply dropping the particular carrier which overlays it. Finally, results on the use of the minimum mean-squared error (MMSE) receiver in a fading channel are extended for use in the overlay scenario.

By using continuous-phase frequency shift keying (CPFSK) in a direct-sequence code-division multiple-access (DS-CDMA) system, and by partitioning the spreading sequence so that differing patterns of plus and minus chips are multiplied onto successive data bits, the net phase change will differ from bit to bit, giving an apparent variation in modulation index similar to multi-h CPFSK. This gives similar coding gain effects as multi-h CPFSK when maximum-likelihood sequence detection is used. Improved probability of error performance in terms of increased minimum distance has been demonstrated when the spreading sequence of a CDMA system is properly overlaid on outgoing data bits in the CPFSK modulation scheme. At the best modulation index observed, h=5/7, a computer search shows /spl sim/1.4-dB advantage over binary phase shift keying, and a /spl sim/0.6-dB advantage over the best fixed-h CPFSK.

This paper presents a single-user code timing estimation algorithm for direct-sequence code-division multiple access that is based on processing the weight vector of an adaptive filter. The filter weight vector can be shown to adapt in the mean to a scaled time-shifted version of the spreading code of the desired user. Therefore, our algorithm requires very little side information in order to form its estimate. The acquisition performance of the algorithm is investigated when the filter is adapted using the least mean square (LMS) or the recursive least square (RLS) algorithm. The proposed algorithm is shown through experimental results to be resistant to the near-far problem when the RLS adaptation algorithm is used, but not when the LMS algorithm is used. However, the performance of this code-acquisition technique is still substantially better than the traditional correlator-based approach, even when the computationally simple LMS algorithm is used. As an extension to the basic timing estimator algorithm, we consider the effect of frequency synchronization error on the performance of the timing estimate. As expected, frequency-offset error degrades the performance of the timing estimate. However, a modified version of the adaptive filter is presented to combat this effect.

The single-bit single-user minimum mean-squared error (MMSE) receiver for direct-sequence code-division multiple access systems is studied in great detail. It is shown that the MMSE receiver is near-far resistant and asymptotically optimum as the interference or noise disappear (i.e., in interference or noise dominated scenarios). It is experimentally shown that the popular Gaussian approximation of the error probability is reasonable for the MMSE receiver over a broad range of operating conditions-i.e., signal-to-noise ratios and interference situations-even though this may not be the case for the standard matched-filter receiver. The concept of dimension reduction is introduced to reduce the complexity of the adaptive part of the receiver and the optimum dimension reduction transform is derived. Unfortunately, the optimum transform is dependent on unknown side-information and is, therefore, of limited practical use. However, several practical schemes are proposed (one which is asymptotically optimum). The schemes are shown to outperform previously known methods.

We propose a receiver diversity based code-timing estimator for DS-CDMA systems. The systems are assumed to work in a flat fading and near-far environment, where an arbitrary antenna array is used at the receiver of the system to achieve the spatial diversity. The algorithm is derived by modeling the known training sequence as the desired signal and all other signals including the multiuser interfering signals and the additive noise as unknown colored Gaussian noise so that a knowledge of the number of active users is not required. We show that by utilizing the information collected via multiple antenna sensors, the length of the training sequences can be greatly reduced. We also show that the algorithm is an asymptotic maximum Likelihood estimator. As a result, the mean-squared error of the code-timing estimates obtained by the algorithm approaches the Cramer-Rao lower bound (CRB) as the length of the training sequence increases. Moreover, the algorithm does not require the search over a parameter space and the code-timing is obtained by rooting a second-order polynomial, which is computationally very efficient. Simulation results show that the algorithm is quite robust against the near-far problem and requires a much shorter training sequence than the existing estimators.

This paper studies the problem of noncoherent detection of Gaussian minimum shift keying (GMSK). Particular attention is focussed on the popular case of BT=0.25, but the concepts presented apply to other cases as well in addition to the general case of continuous phase frequency shift keying (CPFSK). The optimum noncoherent detector is shown to consist of a bank of matched filters followed by a sequence estimator. The metric which must be optimized by the sequence estimator is explicitly given. A simple approximation is introduced which allows the number of matched filters to be reduced from 16 to 3 for GMSK (BT=0.25). A suboptimal metric is introduced which is interpreted as an N-symbol differential detection metric. Based on this metric, a receiver structure is then presented which uses the Viterbi algorithm with a 2/sup N/ state trellis to perform the sequence estimation. Simulation results indicate that it is possible to obtain performance within 1 dB of optimum coherent detection on either a Gaussian noise or a slowly Rayleigh fading channel with the N=4 symbol differential detector.

In order to take advantage of the unused bandwidth available in sparsely populated narrowband systems, the concept of code-division multiple-access (CDMA) overlay has been examined (Milstein et al., 1992). It was shown in Rainbolt and Miller (1997) that even a CDMA system that is lightly loaded can cause significant degradation to the performance of the narrowband system, perhaps precluding the possibility of overlay. However, by notching the CDMA signal in those frequencies at which a narrowband user is operating, the effect of the CDMA interference on the performance of a narrowband user was shown to be greatly alleviated. This paper examines the effects of notch filtering on the performance of the CDMA system in an overlay scenario, and it is shown that the minimum mean-squared error (MMSE) receiver (Miller 1995; Rapajic and Vucetic 1994; Madhow and Honig 1994; Pateros and Saulnier 1996) can be used successfully, and offers a substantial improvement over the conventional matched-filter receiver. The results of this paper, along with those of Rainbolt and Miller, give strong support to both the feasibility and usefulness of CDMA overlay.

The concept of code division multiple access (CDMA) overlay has been proposed as a way to take advantage of the unused bandwidth available in sparsely populated narrowband systems. Most research up to this point has focused on the effects of the narrowband system on the CDMA system. However, because the CDMA system is likely to be overlaid on a narrowband system that is already established, it is of equal or greater importance to examine how the CDMA overlay would affect the narrowband system's performance. It is shown that even a lightly loaded CDMA system can have a detrimental effect on a narrowband system to the point where the possibility of overlay may be precluded. To solve this problem, the idea of notch filtering the CDMA signals in an effort to avoid those bands already occupied by narrowband users is examined. For certain filtering methods the effects of CDMA interference on the narrowband system are shown to be greatly reduced, thus making overlay feasible from the perspective of the narrowband system. For the CDMA system, the performance of the minimum mean-squared error (MMSE) receiver was evaluated and it was found to work quite well with CDMA signals that have been notched. With CDMA transmitter filtering, the prospects for overlay become very encouraging.

This paper examines the behavior of the minimum mean-squared error (MMSE) receiver in frequency-nonselective-fading channels. It is noted that the MMSE receiver will often lose phase lock on the desired signal when the desired signal dips into a deep fade. A modification to the MMSE receiver is presented which is demonstrated to function quite nicely in flat-fading channels. Analytical results for the modified MMSE receiver are presented and found to agree very well with simulation results. These analytical results are then compared to the theoretical performance of the conventional (i.e., correlator) receiver in terms of both bit-error rate (BER) and capacity. As expected, the modified MMSE receiver was found to offer a substantial improvement in both BER and capacity. Finally, a simple empirically derived formula is given which will give a good approximation to the BER of the modified MMSE receiver in a Rayleigh-fading environment. This formula can also be used to determine the number of users a given system can support. It is noted that as E/sub b//N/sub 0/ grows, it is quite feasible to approach 100% channel utilization with the MMSE receiver, whereas a conventional receiver is typically limited to a utilization of 10%-20%.

We propose an efficient algorithm for estimating the code timing of direct-sequence code-division multiple-access (DS-CDMA) systems that consist of an arbitrary antenna array at the receiver and work in a flat-fading and near-far environment. The algorithm is an asymptotic (for large number of data samples) maximum-likelihood (ML) estimator that is derived by modeling the known training sequence as the desired signal and all other signals including the interfering signals and the additive noise as unknown colored Gaussian noise. The algorithm does not require the search over a parameter space and the code timing is obtained by rooting a second-order polynomial, which is computationally very efficient. Simulation results show that the algorithm is quite robust against the near-far problem and channel fading. It requires a shorter training sequence than the single-antenna-based estimators.

In portable wireless communication systems, power consumption is of major concern. Traditional modulation and coding schemes have been designed from the standpoint of minimizing average power. However, for linear power amplifiers needed for spectrally efficient modulation formats, amplifier efficiency and hence power consumption are determined by the peak power of the transmitted signal. This paper looks into modulation formats which minimize peak power and retain high spectral efficiency. Peak power is broken into a sum (in terms of decibels) of average power and a peak-to-average power ratio, and a variety of modulation formats are evaluated in terms of peak power efficiency in both a Gaussian noise and Rayleigh fading channel. A generalized phase shift keying (PSK) modulation format is developed and shown to offer superior peak power efficiency relative to that of commonly used linear modulation formats. Two schemes are presented for reducing the peak-to-average power ratio of various modulation formats. First, data translation codes are used to avoid data sequences which cause large peaks in the transmitted signal. This approach was found to be most productive in quadrature amplitude modulation (QAM) formats. Finally, an adaptive peak suppression algorithm is presented which further reduces the peak-to-average power ratios of the PSK and generalized PSK formats. The peak suppression algorithm is also applicable to /spl pi//4-QPSK and was found to improve peak power efficiency of that format by about 1.25 dB over a Rayleigh fading channel.

In this paper, we present an efficient algorithm for estimating the propagation delay of a known training sequence in an asynchronous direct-sequence code division multiple access (DS-CDMA) system. The algorithm is a large sample maximum likelihood (LSML) estimator that is derived by modeling the known training sequence as the desired signal and all other signals including the interfering signals and thermal noise as unknown colored Gaussian noise that are uncorrelated with the desired signal. LSML is asymptotically statistically efficient as the length of the training sequence goes to infinity. We shall show that LSML is robust against the near-far problem. The performance of the LSML estimator is compared with that of the MUSIC estimator via numerical examples and is shown to be better than that of the latter method in the sense of having lower root-mean-squared errors, lower computational complexity, and tolerating more users.

The problem of estimating propagation delays of the transmitted signals in a direct-sequence code-division multiple-access (DS-CDMA) system operating over fading channels is considered. Even though this study is limited to the case when the propagation delays are fixed during the observation interval, the channel gain and phase are allowed to vary in time. Special attention is given to the near-far problem which is catastrophic for the standard acquisition algorithm. An estimator based on subspace identification techniques is proposed, and the Cramer-Rao bound, which serves as an optimality criterion, is derived. The Cramer-Rao bound is shown to be independent of the near-far problem, which implies that there is no fundamental reason for propagation delay estimators to be near-far limited. Furthermore, the proposed algorithm is experimentally shown to be robust against the near-far problem.

There are some applications where the information source is nonsymmetric. Here, it is shown that differential pulse code modulation (DPCM) of video signals results in a nonsymmetric source. Previous works on 2-D signal design have been focused around equally likely sources. In this paper, an iterative algorithm for minimum error signal design subject to average power (or peak power) constraint is presented. Then, design of minimum cost signal sets is addressed, and application to a DPCM system for picture transmission is developed. Comparisons are made between the nonsymmetric and symmetric signal design in one case and between minimum error signal design and minimum cost signal selection in the other case, for the transmission of DPCM-coded signals. It is seen that nonsymmetric signal design gains significant improvements over the equally likely signal selection when the channel is noisy. Minimum cost signal design also demonstrated some improvement relative to minimum error signal selection over noisy channels.

This paper studies the transient behavior of an adaptive near-far resistant receiver for direct-sequence (DS) code-division multiple-access (CDMA) known as the minimum mean-squared error (MMSE) receiver. This receiver structure is known to be near-far resistant and yet does not require the large amounts of side information that are typically required for other near-far resistant receivers. In fact, this receiver only requires code timing on the one desired signal. The MMSE receiver uses an adaptive filter which is operated in a manner similar to adaptive equalizers. Initially there is a training period where the filter locks onto the signal that is sending a known training sequence. After training, the system can then switch to a decision-directed mode and send actual data. This work examines the length of the training period needed as a function of the number of interfering users and the severity of the near-far problem. A standard least mean-square (LMS) algorithm is used to adapt the filter and so the trade-off between convergence and excess mean-squared error is studied. It is found that in almost all cases a step size near 1.0/(total input power) gives the best speed of convergence with a reasonable excess mean-squared error. Also, it is shown that the MMSE receiver can tolerate a 30-40 dB near-far problem without excessively long convergence time.

In an asynchronous direct-sequence code-division multiple access (DS-CDMA) communication system, the parameter estimation problem, i.e., estimating the propagation delay, attenuation and phase shift of each user's transmitted signal, may be complicated by the so-called near-far problem. The near-far problem occurs when the amplitudes of the users received signals are very dissimilar, as the case might be in many important applications. In particular, the standard method for estimating the propagation delays will fail in a near-far situation. Several new estimators, the maximum likelihood, an approximative maximum likelihood and a subspace-based estimator, are therefore proposed and are shown to be robust against the near-far problem. No knowledge of the transmitted bits is assumed, and the proposed estimators can thus be used for both acquisition and tracking. In addition, the Cramer-Rao bound is derived for the parameter estimation problem.

This paper studies trellis codes to be used with orthogonal signal constellations. The intended application is in direct-sequence code-division multiple-access communication systems. The problem of designing good codes is addressed showing how to construct trellis codes from block codes with good burst-error detecting capabilities. Several examples are given resulting in codes equivalent to those found previously by computer search (Ryan and Wilson, 1991). The procedure presented can be done by hand and is applicable for designing codes with long constraint lengths and large signal sets where a computer search becomes prohibitive. Performance of such coded systems can be determined through transfer functions of the specific codes used. A general transfer function is given which provides a reasonable approximation to the actual transfer function of good codes. Thus, a first-order approximation to the performance of a given system can be determined without actually going through the process of designing and analyzing a specific code.>

An adaptive receiver is considered for use in combating the near-far problem in direct-sequence code-division multiple-access (CDMA) communication networks. The focus of the paper is on the multiuser interference rejecting capability of the receiver. The receiver uses a chip matched filter followed by an adaptive equalizer structure to perform the despreading operation. This adaptive structure allows the receiver to adjust to the prevailing interference and noise environment. An analysis of the receiver giving the form of the equalizer coefficients and the mean squared error in steady state is given. From these results the capacity of a CDMA network using this receiver structure is obtained and compared with a system using conventional receivers. The receiver is shown to be immune to the near-far problem in the sense that the performance without any power control is nearly identical to the performance with perfect power control. The receiver is also shown to offer a two-fold increase in capacity relative to a conventional receiver with perfect power control.>

A model based on some results from queueing theory is developed which takes into account the possibility of overlapping bursts in a meteor-burst (MB) communication system. Application of this model to a simple broadcast MB system shows that for meteor arrival rates that correspond to times of slowest meteor activity the assumption of non-overlapping bursts is justified, while for arrival rates corresponding to times of highest meteor activity the problem of overlapping trails may cause significant performance degradation.>

The problem of DPCM picture transmission over noisy channels is considered. It is well known that DPCM systems are very sensitive to channel errors. The goal in this work is to build robustness against channel errors. Three methods are proposed in this paper and are obtained by modeling the encoded signal as a Markov sequence. First, an optimum method for decoding correlated sequences is derived, and it is shown to require Viterbi decoding. Then, a modified MAP method (MMAP) for Markov sequences is described. A maximal signal-to-noise (MSNR) receiver for DPCM systems is also developed that minimizes the distortion power due to channel errors. The appropriate cost matrix for this receiver is computed. These methods are applied to DPCM picture transmission over noisy channels and are compared with a another method. The SNR graphs, as well as subjective examination of the received pictures, demonstrate that the proposed procedures are quite effective and superior to that method. Among the proposed methods, the MSNR receiver was found to be more effective than the others for a given order of the Markov model. It is observed that the proposed methods are most beneficial for low detail pictures.

The near-far problem can be the capacity limiting factor for direct-sequence code division multiple access (DS-CDMA) communication systems. The "classical" near-far resistant receivers are multi-user detectors with rather high complexity requirements. In order to reduce the complexity, a class of minimum mean square error (MMSE) receivers was proposed. These receivers typically use an adaptive filter to suppress the multi-user interference. The length of the filter is equal to the number of chips per bit. To further reduce the complexity, several methods that shorten the length of the adaptive filter have been developed. We present a common mathematical framework in which these receivers can be evaluated in terms of their probability of bit error. The form of the optimum complexity reduction is derived. Two new structures are proposed and are shown to outperform previously proposed receivers.>

We consider an asynchronous direct-sequence code division multiple access (DS-CDMA) communication system operating over an AWGN channel. In certain applications, the near-far problem can be the limiting factor for the capacity of a DS-CDMA system. Several near-far resistant receivers have therefore been proposed (e.g., the decorrelating receiver). These receivers assume perfect knowledge of the propagation delay from all users to the receiver. In practice, the delays need to be estimated and will thus be subject to estimation errors. We evaluate the performance penalty these errors will impose on previously proposed near-far resistant receivers.>

This letter examines the correlation parameters of interest in a DS-CDMA communication network using M-ary FSK modulation. It is shown that the key multi-user interference parameters can be written in terms of cross-correlation functions with respect to a sinusoidal weighting function. Mean-squared interference is also studied and can be expressed in terms of the auto-correlation of the spreading sequences used. This is a generalization of a well-known result for binary DS-CDMA systems.>

The time-varying signal-to-noise ratio (SNR) model for the meteor burst (MB) channel is reviewed. Bounds on the capacity of the channel are derived for both a constant SNR model and a time-varying SNR model. These bounds show that there is a significant throughput improvement to be gained by using forward error correction. Two methods are given for determining the performance of an MB system when packets of information are encoded with an (n,k) linear block code. Numerical results are generated using high-rate BCH codes, and it is found that about 25% improvement over uncoded systems can be obtained by choosing the code rate properly. In addition, some suggestions for techniques that provide further improvement are given.>

The performance of a meteor-burst (MB) communication system using three different protocols is examined. The protocols were chosen to study the relative advantages of using a simple stop-and-wait automatic repeat request (ARQ) scheme as well as the ability to sense the presence of a channel. It was found that for long messages, the use of the ARQ scheme gave significant performance improvement; however, when the message length was on the order of the average burst length, the use of the ARQ scheme was not helpful. For packet sizes on the order of the average burst length, it was found that the ability to sense the presence of the channel gave about a 20% reduction in the time necessary to complete a message, but for much shorter packet sizes, the channel probing was not needed. The model used to analyze these systems takes into account the time-varying signal-to-noise ratio (SNR) present in a MB channel. The performance predicted by this model is compared to a less sophisticated constant SNR (signal-to-noise) model and its is found that the author's model predicted significantly more optimistic performance.>