In this paper the downlink of a multi-user MIMO (MUMIMO) system with multi-mode transmission is considered. We propose a low-complexity algorithm for selecting users and the corresponding number of data streams to each user, denoted as user transmission mode (UTM). The selection is only based on the average received signal-to-noise ratio (SNR) from the base station (BS) for each user. This reduces the overall amount of feedback for scheduling, as opposed to techniques that assume perfect instantaneous channel state information (CSI) from all users. Analytical average throughput approximations are derived for each user at different UTMs. Simulation results demonstrate that the proposed algorithm provides performance close to dirty paper coding (DPC) with considerably reduced feedback.

For wireless systems with adaptive modulation and/or scheduling, feedback of channel quality information is often necessary. It has been questioned whether the increased system performance is worth the additional feedback rate and the increased algorithm complexity. In this paper, we study how the feedback rate can be minimized, without losing the gains due to adaptive modulation and multiuser diversity. We present an in-depth study of the literature in the area, and evaluate the performance of several state-of-the-art channel quality feedback schemes. By illustrating the compromise between system throughput and feedback channel rate for various schemes, we are able to give valuable insight in choice of method for feedback rate reduction. A major conclusion is that for multicarrier systems, a lossy compression scheme is the best choice, while for single-carrier systems, schemes limiting feedback to only high-SNR users show good performance. Another conclusion is that there are still many issues to study before the schemes can be used in practice.

High data rates, high spectral efficiency, flexibility, and low delays over the air interface will be important features in next-generation wireless systems. The overall challenge will be packet scheduling and adaptive radio transmission for multiple users, via multiple antennas and over frequency-selective wideband channels. This problem needs to be structured to obtain feasible solutions. The basic simplifying assumptions used here are clustering of antennas into cells, orthogonal transmission by use of cyclic-prefix orthogonal frequency-division multiplexing (OFDM) and a time-scale separation view of the total link adaptation, scheduling and intercell coordination problem. Based on these assumptions, we survey techniques that adapt the transmission to the temporal, frequency, and spatial channel properties. We provide a systematic overview of the design problems, such as the dimensioning of the allocated time-frequency resources, the influence of duplexing schemes, adaptation control issues for downlinks and uplinks, timing issues, and their relation to the required performance of channel predictors. Specific design choices are illustrated by recent research within the Swedish Wireless IP program and the EU IST-WINNER project. The presented results indicate that high-performance adaptive OFDM transmission systems are indeed feasible, also for challenging scenarios that involve vehicular velocities, high carrier frequencies, and high bandwidths.

The problem of finding the optimal labeling (bit-to-symbol mapping) of multilevel coherent phase shift keying (PSK), pulse amplitude modulation (PAM), and quadrature amplitude modulation (QAM) constellations with respect to minimizing the bit-error probability (BEP) over a Gaussian channel is addressed. We show that using the binary reflected Gray code (BRGC) to label the signal constellation results in the lowest possible BEP for high enough signal energy-to-noise ratios and analyze what is "high enough" in this sense. It turns out that the BRGC is optimal for PSK and PAM systems whenever the target BEP is at most a few percent, which covers most systems of practical interest. New and simple closed-form expressions are presented for the BEP of PSK, PAM, and QAM using the BRGC

This paper presents an extension for adaptive transmission of a multi-user scheduling algorithm to a cellular environment. In an effort to manage inter-cell interference, scheduling between neighboring base stations is coordinated. To this end, the amount of buffered data at each base station is exchanged within a small group of base stations. More precisely, the sectors that illuminate each cell form inter-sector scheduling groups. We divide the scheduling into two separate operations, inter and intrasector scheduling, that are performed separately. From a comparison against a scheme with a static frequency reuse of three we observe a 20-50% reduction in average packet delays, depending on the traffic model used.

Using an OFDM based uplink for future cellular networks has been a controversial issue due to difficulties in carrier frequency synchronization. In this paper, a model is provided to quantify the effects of Gaussian distributed carrier frequency offsets in a multiuser OFDM uplink. Both theoretical evaluations and simulations on spectral efficiency are performed in a slotted OFDM uplink, in which time-frequency bins are allocated adaptively to different users based on their channel conditions. The results show that the loss in spectral efficiency is small when the standard deviation of each user's carrier frequency offset is no more than 2% of the subcarrier spacing

This paper presents two new layer 4 fairness measures, the worst case TCP fairness index and the TCP fairness index. The purpose of the two indices is to measure the performance of algorithms scheduling elastic traffic in wireless networks. Numerical evaluation of round-robin, proportional-fair and max-rate scheduling shows that the new measures can accurately classify the fairness properties of the studied scheduling algorithms. This means that the new fairness measures are appropriate for measuring TCP fairness. The worst case TCP fairness index is the primary measure. The TCP fairness index can be used as a complement in certain cases. The new fairness measures are also compared with two layer 2 fairness measures. Numerical evaluation shows that the new layer 4 measures and the layer 2 measures in some aspects show similar results. However, when evaluating transport layer fairness, it is better to use the new layer 4 measures as they actually measure fairness on layer 4 and the results are easier to interpret

This paper presents a novel unified analysis for the bit-error rate in the downlink of convolutionally coded and single-user detected multicarrier code-division multiple access (CDMA), multicarrier direct-sequence CDMA, time and frequency-domain spreading CDMA, and orthogonal frequency-division multiple access. Performance is analyzed under the assumption of Hadamard spreading codes, equal-gain chip combining, and a highly correlated frequency-selective Rayleigh-fading channel. Closed-form expressions are presented for the cumulative distribution function, probability distribution function, and moment-generating function for the signal-to-noise plus interference ratio after despreading. The presented results assume error-free channel estimates, a perfectly synchronized receiver, and are found to reasonably well agree with simulation results.

This paper is concerned with the problem of selecting a binary labeling for the signal constellation in an M-PSK communication system. A good starting point is labelings having the Gray property, but this is not altogether enough, since the number of distinct Gray labelings that result in different bit error probability grows rapidly with increasing constellation size. By introducing a recursive Gray labeling construction method called expansion, the paper answers the natural question of what labeling, among all possible constellation labelings (not only Gray), that will give the lowest possible average probability of bit errors. Under certain assumptions on the channel, the answer is that the labeling originally proposed by Gray, the binary reflected Gray code, is the optimal labeling for M-PSK systems, which has, surprisingly, never been proved before.

This paper proposes that fairness in wireless networks should be measured using one of the following new measures: the deterministic unfairness bound called the wireless absolute fairness bound (WAFB) or the statistical unfairness bound called the 99-percentile wireless absolute fairness bound (WAFB/sub 99/). Compared with previous fairness definitions, the new fairness measures are better suited for measuring fairness of scheduling disciplines that exploit multiuser diversity. A new scheduling discipline called opportunistic proportional fair scheduling is defined. Numerical results show that the new scheduling discipline has slightly higher throughput and slightly better fairness than proportional fair scheduling.

This paper concerns the problem of selecting a binary labeling for the signal constellation in M-PSK, M-PAM, and M-QAM communication systems. Gray labelings are discussed and the original work by Frank Gray is analyzed. As is noted, the number of distinct Gray labelings that result in different bit-error probability grows rapidly with increasing constellation size. By introducing a recursive Gray labeling construction method called expansion, the paper answers the natural question of what labeling, among all possible constellation labelings, will give the lowest possible average probability of bit errors for the considered constellations. Under certain assumptions on the channel, the answer is that the labeling proposed by Gray, the binary reflected Gray code, is the optimal labeling for all three constellations, which has, surprisingly, never been proved before.

This paper explores the utilization of soft information in image decoding. Error free transmission is not always possible in today's nor future's communication networks without retransmissions. Retransmissions, however, introduce delay. To delay sensitive applications the concept of soft information, i.e. lower layer knowledge about the channel conditions, could reduce the number of retransmissions. It is not sufficient to treat the soft information and the image decoder separately. By combining soft information and the knowledge of the image structure in the decoding process image quality can be improved. A modified JPEG2000 image decoder that utilizes soft information has been developed. Experimental results with images transmitted over a simulated wireless channel show that iterative decoding with soft information can give high gains in image quality.

The problem of calculating the average bit-error probability (BEP) of coherent M-ary phase-shift keying (PSK) over a Gaussian channel has been studied previously in the literature. A solution to the problem for systems using a binary reflected Gray code (BRGC) to map bits to symbols was first presented by P.J. Lee (see ibid., vol.COM-34, p.488-91, 1986). We show that the results obtained by Lee are incorrect for M/spl ges/16. We show that the reason for this is an invalid assumption that the bit-error rate (BER) is independent of the transmitted symbols, an assumption which has also propagated to textbooks. We give a new expression for the BER of M-PSK systems using the BRGC and compare this with Lee's results.

To evaluate the performance of a convolutional code it is common to use a union bound on either the bit-error rate or the first-event error rate. Since these bounds are given as infinite sums they must be truncated to be evaluated. However, as the tail of the infinite sum is discarded, the upper bound turns into an approximation. Two methods to evaluate the union bound while taking the summation tail into account and thus maintaining the upper bound are described. The first method is based on a state matrix description of the convolutional encoder, while the second method is based on the transfer function. Results are given that compare the upper bounds and the truncated bound (approximation) to computer simulations. The standard rule-of-thumb that the union bound is tight as long as the error rates are below 10/sup -3/ is readily seen to be valid.

Recent publications on new improved convolutional codes claim different codes to be optimum under the optimum distance spectrum (ODS) criterion. This correspondence presents results from a study independent of the previous works, and seeks to clarify some inconsistencies. In addition, some new improved feed-forward convolutional encoders are presented.

Currently, a global third-generation cellular system based on code-division multiple-access (CDMA) is being developed with a wider bandwidth than existing second-generation systems. The wider bandwidth provides increased multipath resolution in a time-dispersive channel, leading to higher frequency-selectivity. A generalized RAKE receiver for interference suppression and multipath mitigation is proposed. The receiver exploits the fact that time dispersion significantly distorts the interference spectrum from each base station in the downlink of a wideband CDMA system. Compared to the conventional RAKE receiver, this generalized RAKE receiver may have more fingers and different combining weights. The weights are derived from a maximum likelihood formulation, modeling the intracell interference as colored Gaussian noise. This low-complexity detector is especially useful for systems with orthogonal downlink spreading codes, as orthogonality between own cell signals cannot be maintained in a frequency-selective channel. The performance of the proposed receiver is quantified via analysis and simulation for different dispersive channels, including Rayleigh fading channels. Gains on the order of 1-3.5 dB are achieved, depending on the dispersiveness of the channel, with only a modest increase in the number of fingers. For a wideband CDMA (WCDMA) system and a realistic mobile radio channel, this translates to capacity gains of the order of 100%.

In a CDMA cellular system, the process of the mobile station searching for a cell and achieving code and time synchronization to its downlink scrambling code is referred to as cell search. Cell search is performed in three scenarios: initial cell search when a mobile station is switched on, idle mode search when inactive, and active mode search during a call. The latter two are also called target cell search. This paper presents algorithms and results for both initial and target cell search scenarios for the wideband CDMA (W-CDMA) standard. In W-CDMA, the cell search itself is divided into five acquisition stages: slot synchronization, frame synchronization and scrambling code group identification, scrambling code identification, frequency acquisition, and cell identification. Initial cell search needs all five stages, while target cell search in general does not need the last two stages. A pipelined process of the first three stages that minimizes the average code and time acquisition time, while keeping the complexity at a reasonable level, is considered. The frequency error in initial cell search, which may be as large as 20 kHz, is taken care of by partial symbol despreading and noncoherent combining. Optimization of key system parameters such as the loading factors for primary synchronization channel, synchronization channel, and common pilot channel for achieving the smallest average code and time acquisition time is studied. After code and time synchronization (the first three stages), a maximum likelihood (ML)-based frequency acquisition method is used to bring down the frequency error to about 200 Hz. The gain of this method is more than 10 dB compared to an alternative scheme that obtains a frequency error estimate using differential detection.

In code division multiple-access (CDMA) systems, maximum total throughput assuming a matched filter receiver can be obtained by spreading with low-rate error control codes. Previously, orthogonal, bi-orthogonal and super-orthogonal codes have been proposed for this purpose. We present in this paper a family of rate-compatible low-rate convolutional codes with maximum free distance. The performance of these codes for spectrum spreading in a CDMA system is evaluated and shown to outperform that of orthogonal and super-orthogonal codes as well as conventionally coded and spread systems. We also show that the proposed low rate codes will give simple encoder and decoder implementations. With these codes, any rate 1/n, n/spl les/512, are obtained for constraint lengths up to 11, resulting in a more flexible and powerful scheme than those previously proposed.

We consider joint source-channel and multiuser decoding for frequency selective Rayleigh fading code-division multiple-access channels. The block source-channel encoder is defined by a vector quantizer. We investigate optimal (minimum mean-square error) decoding and "user-separated" decoding of lower complexity. The studied decoders are soft in the sense that they utilize all soft information available at the receiver. Simulations indicate significant performance gains of the introduced decoders compared with a tandem approach that uses maximum-likelihood multiuser detection plus table-lookup decoding.

The performance of an asynchronous code division multiple access (CDMA) system (uplink) employing very low-rate maximum free distance codes for combined coding and spreading is analyzed when successive or parallel interference cancellation is applied. An analytical approach to the evaluation of the bit error rate is presented and shown to give results close to simulations. Our results show that the code-spread system outperforms the conventionally coded and spread system. Without interference cancellation the single-user bound is never reached (except for one user). With two stages of parallel interference cancellation, a code-spread system with a load only slightly less than 1 bit/chip can obtain a bit error rate very close to that of a single-user system.

New rate-compatible convolutional (RCC) codes with high constraint lengths and a wide range of code rates are presented. These new codes originate from rate 1/4 optimum distance spectrum (ODS) convolutional parent encoders with constraint lengths 7-10. Low rate encoders (rates 115 down to 1/10) are found by a nested search, and high rate encoders (rates above 1/4) are found by rate-compatible puncturing. The new codes form rate-compatible code families more powerful and flexible than those previously presented. It is shown that these codes are almost as good as the existing optimum convolutional codes of the same fates. The effects of varying the design parameters of the rate-compatible punctured convolutional (RCPC) codes, i.e., the parent encoder rate, the puncturing period, and the constraint length, are also examined. The new codes are then applied to a multicode direct-sequence code-division multiple-access (DS-CDMA) system and are shown to provide good performance and rate-matching capabilities. The results, which are evaluated in terms of the efficiency for Gaussian and Rayleigh fading channels, show that the system efficiency increases with decreasing code rate.

An approach to optimal soft decoding for vector quantization (VQ) over a code-division multiple-access (CDMA) channel is presented. The decoder of the system is soft in the sense that the unquantized outputs of the matched filters are utilized directly for decoding (no decisions are taken), and optimal according to the minimum mean-squared error (MMSE) criterion. The derived decoder utilizes a priori source information and knowledge of the channel characteristics to combat channel noise and multiuser interference in an optimal fashion. Hadamard transform representations for the user VQs are employed in the derivation and for the implementation of the decoder. The advantages of this approach are emphasized. Suboptimal versions of the optimal decoder are also considered. Simulations show the soft decoders to outperform decoding based on maximum-likelihood (ML) multiuser detection. Furthermore, the suboptimal versions are demonstrated to perform close to the optimal, at a significantly lower complexity in the number of users. The introduced decoders are, moreover, shown to exhibit near-far resistance. Simulations also demonstrate that combined source-channel encoding, with joint source-channel and multiuser decoding, can significantly outperform a tandem source-channel coding scheme employing multiuser detection plus table lookup source decoding.