Search

Today's 4G LTE systems bring unprecedented mobile broadband performance to over a billion of users across the globe. Recently, work on a 5G mobile communication system has begun, and next to a new 5G air interface, LTE will be an essential component. The evolution of LTE will therefore strive to meet 5G requirements and to address 5G use cases. In this article, we provide an overview of foreseen key technology areas and components for LTE Release 14, including latency reductions, enhancements for machine-type communication, operation in unlicensed spectrum, massive multi-antenna systems, broadcasting, positioning, and support for intelligent transportation systems.

The next major step in the evolution of LTE targets the rapidly increasing demand for mobile broadband services and traffic volumes. One of the key technologies is a new carrier type, referred to in this article as a Lean Carrier, an LTE carrier with minimized control channel overhead and cell-specific reference signals. The Lean Carrier can enhance spectral efficiency, increase spectrum flexibility, and reduce energy consumption. This article provides an overview of the motivations and main use cases of the Lean Carrier. Technical challenges are highlighted, and design options are discussed; finally, a performance evaluation quantifies the benefits of the Lean Carrier.

We develop a unified performance metric and detailed analysis for hybrid automatic repeat request (HARQ) schemes based on incremental redundancy (IR) and Chase combining (CC). The general result is applicable to both symbol-based and bit-interleaved coded modulations, to HARQ processes based on rateless and fixed-rate mother codes, and IR schemes with and without a self-decodability restriction. The analysis shows that IR over CC coding gains tend to increase with the initial coding rate, but decrease with the signal-to-noise (SNR) variation between retransmissions. The gains can also be diminished when a fixed-rate mother code is used or when the self-decodability criterion is imposed. The theoretical prediction is compared with observed gains at 10% codeword error rates based on turbo-coding simulations. For scenarios with moderately varying SNR between retransmissions, the analytical model tracks actual simulation results very well. However, when the SNR varies widely and the systematic part of the turbo codeword is effectively erased, the CC scheme could, in fact, outperform some IR schemes. For these scenarios, which can be induced by fast time-varying fading or long retransmission delays, the self-decodable IR and the CC schemes prove to be more robust without much performance comprise. Finally, we discuss adaptive improvement to the conventional IR schemes based on the analytical result

Dynamic base station downlink discontinuous transmission (DTX) operations have been proposed and under discussion in the standardization body as a candidate to reduce interference in the LTE network to support rapidly increasing demand for mobile broadband services and traffic volumes [1]. In this paper, we analyze the energy efficiency improvement of the dynamic base station DTX operations from three perspectives. First, we discuss the complementary operation aspect of the dynamic base station DTX and existing LTE dormant mode. Secondly, we provide an analysis showing substantial energy efficiency benefits at all loads of dynamic base station DTX operation with near- future base station hardware components. Thirdly, we provide an analytical framework to assess the energy saving potential of the dynamic base station DTX concept for all possible future hardware improvement possibilities.

To evolve the LTE network to cope with ever and rapidly increasing demand for mobile broadband services and traffic volumes, interference avoidance designs have been identified as one major area to address in the next LTE release. In both traditional homogeneous macro network and heterogeneous network with mix of macro and small cells, intercell interference can be substantially reduced by temporarily halting unneeded downlink transmission. In this paper, we analyze several design aspects of such base station downlink discontinuous transmission (DTX) operations. We provide extensive performance evaluation results and analysis on (1) the time scale of such DTX operations; and (2) whether the DTX operations should be restricted to the small cells only. The results provide guideline for an ideal combination of these design dimensions.

Carrier aggregation has enabled LTE to efficiently utilize wide and fragmented spectrum to provide high data rate services to end users' benefits. We investigate the challenges of providing reliable multi-carrier hybrid automatic repeat request (HARQ) and timely periodic channel state information (CSI) feedbacks from terminals aggregating multiple component carriers in LTE networks. We consider unified and reliable uplink control channel designs to carry multi-carrier HARQ and periodic CSI feedback messages simultaneously. Care is taken in the designs to reduce the detection complexity on the network side, which also improves its detection reliability relative to existing LTE signalling schemes.

Cellular wireless systems have focused, to date, on system capacity, user data rates, latency and coverage as primary design criteria. Energy efficiency has recently received increased attention as another important criterion to be considered. The potential for energy savings in 3GPP LTE systems has been investigated in [1]-[3]. An important design element that determines the potential for energy efficient transmission is the structure of reference signals that need to be transmitted independent of the load in the network, such as the cell-specific reference signals (CRS) in 3GPP LTE systems. This paper discusses modifications to the reference signal structure in LTE that can enable greater energy efficiency. The time and frequency estimation performance implications of such modifications are evaluated and discussed.

This paper discusses the design, performance optimization and potential extensions of the LTE HARQ operations. The HARQ retransmission behavior of all 764 possible LTE code blocks is classified into 12 types based on two easily and already computed quantities in the standard. Using with the unified information-theoretical quantification of various HARQ gains discussed in the companion paper, the optimal retransmission settings for the LTE system are designed. The analytical framework is further applied to optimize potential modifications to the LTE standard for enhanced retransmission performance. Numerical results are provided to verify these optimized settings.

We develop a general model based on random coding bounds to explain and predict the coding performance of various type-I and type-II hybrid automatic repeat request (HARQ) schemes based on the bit-interleaved coded modulation (BICM) in 3GPP HSPA. We explore how the intrinsic properties of BICM interact with various transmitter-side (e.g., whether to use the same bit mapping for retransmission) and receiver-side HARQ options (e.g., bit vs. symbol level processing and Log-MAP vs. Log-Max demodulation). The accuracy of the theoretical model is verified with extensive link simulations. The performance model can be incorporated into radio resource management algorithms for efficient wireless network operations. In a companion paper, the unified analytical model is further applied to investigate performance of LTE HARQ operations, which consist of varying mixture of type-I/II and bit remapping HARQ features.

This paper discusses circular buffer rate matching (CBRM) algorithms for the turbo code in the long term evolution (LTE) of the WCDMA-based air interface. To enhance performance at high code rates, systematic bit puncturing is incorporated in conjunction with the CBRM. The RM algorithm is further optimized based on the algebraic properties of the QPP interleavers and the 8-state recursive systematic convolutional code of the LTE turbo code.

The design of LTE turbo coding chain suitable for flexible parallel and pipelined hardware implementations is presented. The hierarchical data structure further offers an opportunity to design an efficient two-level early stopping algorithm. An analytical model is developed to compute the distribution of actual decoding iterations performed by the proposed algorithm. It is then shown the scheme can reduce both the total hardware requirement at the design phase as well as the average power consumption at run-time of turbo decoders.

The error detection performance of CRC coding in LTE with general two-level early stopping algorithms for turbo decoding is investigated. Analytical models for the probability of block error and undetected block error at the code block and transport block levels were developed. Simulations were used to verify the model for shorter CRC lengths and block sizes. The analytical models show the setting of 24-bit CRCs in LTE allows low-complexity, robust and reliable early stopping algorithms to reduce turbo decoding complexity.

Work is ongoing in 3GPP to significantly extend the performance of UMTS in the work item long term evolution (LTE). LTE contains a new system architecture with fewer infrastructure nodes and it has been decided to terminate the ARQ functionality in the evolved node B. This paper describes the requirements that exists for an LTE ARQ concept and outlines a solution that fulfills these requirements in the new LTE architecture. The solution builds on two layered ARQ feedback mechanisms that provide sufficient reliability with a low resource cost for the feedback The paper contains thorough stepwise argumentation how we arrived at the proposed ARQ concept.

Performance of conventional decoders with inaccurate channel estimation input is analyzed to construct an accurate and computationally efficient model for dynamic system simulations. The model abstracts general channel conditions experienced by a codeword into a collection of repetitive states, caused by type-I HARQ or diversity receive antennas, and non-repetitive states, caused by fast fading, type-II HARQ or OFDM transmissions. An effective SNR accounting for losses due to channel estimation errors is derived for the repetitive states. The contribution of all non-repetitive states is summarized in the normalized conditional information rate, which can then used to read off the correct codeword error rate from a look-up table. The accuracy of the proposed model is verified with extensive simulation results.

We present a new methodology for constructing accurate physical-layer (radio link) performance models for dynamic wireless network system simulations. Our approach emphasizes the consistent treatment of the residual fading effects caused by inaccurate channel estimation across slot, transmission time interval, and hybrid automatic repetition request levels. The proposed performance models can he implemented using simple manipulations of certain pre-calculated values. Numerical results are provided to illustrate the accuracy for individual models as well as for multiple models combined together.

We present the potential performance improvements attainable by new enhancements to the WCDMA uplink. Improvements can be expected in both system capacity and end-user perceived performance. In this paper we focus on the capacity gains of short TTI, hybrid ARQ and scheduling in node-B. The performance is studied from both the link-level and system-level perspectives. The link performance results focus on the gains obtained by the introduction of short TTI and hybrid ARQ with soft combining in node-B. In the system performance results, we also evaluate node-B based rate scheduling for both 2 ms TTI and 10 ms TTL. Our studies indicate that with the introduction of enhanced uplink the gain in uplink capacity can be in the order of 70-100%.

We develop an information-theoretic model to explain and predict the coding gains of the incremental redundancy (IR) over the chase combining (CC) scheme using random coding bounding techniques. From extensive simulations, the model is found to provide accurate coding gain predictions in conjunction with either optimal or practical repeated low-rate codes. Contrary to a widely held belief that IR is superior to CC for all circumstances, we demonstrate that this is not always the case when parallel concatenated turbo codes with severe puncturing are used in IR on fading channels. Based on the theoretical model, we further propose an adaptive algorithm to remedy this anomaly.

We consider the maximum-likelihood (ML) decoding performance of concatenated space-time (ST) random coded communication utilizing multiple input and multiple output (MIMO) antennas on quasi-static fading channels. The ensemble-averaged codeword error rates of such systems can be upper bounded by the outage probabilities of the corresponding conditional cutoff rates. We refer to these as the information cutoff probabilities and evaluate them numerically for a wide range of ST coding schemes. The slope behaviors of these cutoff probabilities are found to be consistent with the maximum achievable diversity orders given by a new unified bound. With simulations of a class of ST coding schemes based on bit-interleaved coded modulations and iterative spatial/temporal decoders, we further show these cutoff probabilities can provide accurate predictions for total performance gains.

For pt. see ibid., p.1177-81(2002).We continue the consideration of the maximum-likelihood decoding performance of MIMO space-time random coded communication on quasi-static fading channels. First, we generalize the formulation to frequency-selective fading channels. By the equivalence of multipath fading and delay diversity coding, we show the system can be analyzed with an equivalent flat fading representation. We then present numerical results of the p% cutoff capacities, which are defined as the highest supportable rates for target cutoff probabilities at p%. Our discussion show how this new notion can provide a unified view of three main theoretic results of MIMO space-time coding: rate enhancements, achievable diversity gains, and performance of specific codes. As a result, this analysis of cutoff capacities reveals the relative strengths and weaknesses of several space-time coding schemes in interesting and surprising ways.

We develop an information-theoretic model to explain and predict the coding gains of the incremental redundancy over the Chase combining scheme. From this analysis, we identify a new approach to improve the spectral efficiency of the conventional incremental redundancy scheme, which uses the same modulation and coding formats and, most notably, the same amounts of channel resources for retransmissions as those for the initial transmissions. Depending on the relative channel conditions between the transmissions, fewer incremental redundant bits than those employed in the first transmission might be enough to guarantee high probability of successful decoding. Hence, we propose a new scheme that adapts the modulation and coding formats according to simple feedback information to minimize unnecessary channel usage. Our simulation results demonstrate system throughput gains up to 25% in certain conditions can be achieved by the proposed scheme without modifying the existing WCDMA specifications.

A new coding design with large coding gains and complexity-reduced decoders is proposed. The coding strategy is based on a multilevel coding construction with rate-compatible punctured convolutional codes as building blocks. A new sub-optimal decoding is proposed where only one decoder using Viterbi algorithm is drafted. Typical example codes for quadrature amplitude shift keying (QASK) constellations an constructed and shown to outperform Ungerboeck's codes, but with higher decoding delay. Computer simulation is also performed to verify the results. >

The methodology of multi-stage iterative decoding with exchange of soft information has previously been applied to a wide range of error-controlling codes with remarkable success, including parallel/serial concatenated convolutional/block codes (Berrou et al., 1993) and generalized concatenated convolutional codes (Cheng). A theoretical explanation of this decoding mechanism has been proposed (MacKay et al.) via the belief propagation algorithm (BPA) (Pearl 1988). It is shown that, by representing a turbo code with a Bayesian belief network, the application of BPA directly leads to the turbo decoding algorithm. In fact, it is argued that the decoding algorithms for all the aforementioned variations of the turbo coding construction as well as a wide range of low-density parity check codes can be explained by this principle. The author goes on to discuss low density generator matrix codes.

Several parallel concatenated coding schemes (turbo codes) based on multimemory (MM) convolutional codes (more specifically, a (2,1,4,7) code) were proposed to achieve near Shannon-limit error correction performance with reasonable decoding complexity. On the other hand, in many cases of interest, unit-memory (UM) codes have been demonstrated to have larger free distances than the MM codes with the same rate and the same number of memory elements. New turbo codes based on the (8, 4, 3, 8) UM Hamming code are developed and shown to possess better performance potential in some senses. The standard turbo decoding algorithms, however, do not appear to achieve this potential.

Different floating-point and fixed-point implementation issues of the log-MAP algorithm for decoding turbo codes are studied. A theoretical framework is proposed to explain why the log-MAP turbo decoders are more tolerant of overestimation of SNR than underestimation. A linear approximation (the Log-Lin algorithm) that allows for efficient ASIC implementations and results in negligible performance loss in floating-point models is proposed. It is further shown that a fixed-point implementation of this Log-Lin algorithm, including clipping and quantization of the received samples to 6 bits, results in a loss of performance of less than 0.05 dB.

The following topics were dealt with: Shannon theory; universal lossless source coding; CDMA; turbo codes; broadband networks and protocols; signal processing and coding; coded modulation; information theory and applications; universal lossy source coding; algebraic geometry codes; modelling analysis and stability in networks; trellis structures and trellis decoding; channel capacity; recording channels; fading channels; convolutional codes; neural networks and learning; estimation; Gaussian channels; rate distortion theory; constrained channels; 2D channel coding; nonparametric estimation and classification; data compression; synchronisation and interference in communication systems; cyclic codes; signal detection; group codes; multiuser systems; entropy and noiseless source coding; dispersive channels and equalisation; block codes; cryptography; image processing; quantisation; random processes; wavelets; sequences for synchronisation; iterative decoding; optical communications.

EDGE has enabled the GSM network to offer third generation services as part of the UMTS core network. This is due to the introduction of 8PSK modulation and the recent specification of interfaces from the GERAN to the UMTS core network. This paper studies further improvements to the spectral efficiency and capacity of EDGE through the use of more sophisticated adaptive modulation schemes based on QAM, and the use of novel turbo coding schemes. Furthermore, this work considers a novel transmit diversity scheme adapted for use in dispersive channels. We provide performance results for the enhanced system showing tremendous gains. We also use theoretical predictions of outage analysis to predict the gain in performance obtained by the use of higher order modulations and transmit diversity. The gains from theoretical predictions match our simulations almost exactly.