Search

In this paper, we study the effect of channel estimation error on transmitter design in a multiple input multiple output (MIMO) channel. We assume that only knowledge of either the channel correlation or mean is available at the transmitter and find the transmitting strategy. Under covariance feedback, at the transmitter the channel is modelled as a matrix of zero mean circularly symmetric complex Gaussian (ZMCSCG) random variables with known covariances. We assume that only rows of channel matrix are correlated. Under mean feedback the covariance of channel is modelled as white. We determine the necessary and sufficient conditions under which a unit rank input covariance matrix, i.e. eigen-beamforming, can achieve capacity lower bound. Numerical simulations are conducted to corroborate theoretical results.

Mobility tracking yields efficient network resource management and enables very useful additional services, provided that accurate online estimations of both the location and the velocity of a mobile user are performed. We present a novel method for tracing a mobile user by subsequently estimating his/her position through a single base station positioning (SBSP) method designed for microcellular and third generation (3G) wireless communications systems. Raw data about user position are then smoothed through a linear regression setup which adaptively varies its inertia on the basis of the previous observations. This results in a tracking method which is able to estimate both the location and the velocity of a user with high accuracy, as proven by numerical results obtained in different microcellular scenarios.

The objective of the IST project MIND (Mobile IP based network developments) is to facilitate IP-based broadband services for mobile users in various operating environments. Activities within MIND concentrate on the development, optimisation and harmonisation of the appropriate layers involved. The physical (PHY) layer was chosen to base on the high performance radio local area network type 2 standard (HIPERLAN/2). The work presented here demonstrates the applicability of HIPERLAN/2 to scenarios with increased mobility. Initially, HIPERLAN/2 was designed for rather static environments providing hot spot data rates of up to 54 Mbps. Mitigation techniques to overcome the drawbacks of a fast fading environment are thus studied within IST-MIND. The extracted transceiver structures allow the deployment of HIPERLAN/2 in almost any environments currently envisaged. Particularly, the system deployment of HIPERLAN/2 to a vehicle-to-vehicle communication scenario is studied and preliminary system level performance results presented.

The previously introduced concept of virtual antenna arrays (VAA) promises a significant increase in link and system capacity. These therefore constitute a promising candidate for ad-hoc oriented 4G networks. VAA successfully emulates a multiple-input-multiple-output (MIMO) transmission channel to which space-time codes (STC) can be applied. The realisation of VAA is accomplished by letting spatially close mobile terminals communicate with each other, which overcomes the disadvantage of having only one receiving antenna in a mobile handset. The thus created ad-hoc network represents a general form of MIMO channels for which a new approach to link capacity calculation is required. The work presented establishes some fundamental tenets for the calculation of link capacity in the Shannon sense to ad-hoc networks, such as VAA. It shows that the capacity drastically increases over a network where each terminal maintains a single link to the base station. The results are somewhat preliminary since VAA constitutes only a minor subset of possible ad-hoc network configurations.

We investigate the impact of power control and choice of modulation index on the performance of a multi hop CSMA/CA system. We explicitly take into account the additional traffic generated due to multi-hop transmissions and the effect of frame error rates, the later being a function of the frame length, the modulation index, and the signal-to-noise ratio (SNR). An analytical model for the spectral throughput per user is presented with emphasis on the number of nodes dwelling in the area covered by a given transmission power and on the number of hops. Unlike other works, a closed form solution is derived for the spectral throughput performance of a multi hop CSMA/CA system as a function of the offered load, the nodes density, transmission power and frame error rate. This facilitates the PHY and MAC layers to be jointly optimised, e.g. the system designer may specify the optimum choice of modulation index for a given MAC configuration or, alternatively, the optimum persistency parameter for a given PHY.

Virtual antenna arrays were recently introduced that allow the application of multiple-input-multiple-output (MIMO) capacity enhancement techniques to mobile terminals with a limited number of antenna elements. This paper investigates the performance improvements of deployed higher order space-time block codes (STBCs) to such virtual antenna arrays. The main idea is to emulate (nT/sub x/ transmit antennas, m/sub Rx/ receive antennas) diversity scheme. This approach successfully deals with the problem of just one receiving antenna in a handset. Simulations results indicate a drastic improvement in error performance over (n/sub Tx/,1) schemes.

The performance of space time trellis codes (STTCs) can be further improved by utilizing the partial channel information and combining some precoding schemes at the transmitter. This paper presents a STTC scheme - hybrid STTC (HSTTC). For this scheme, the STTC signals are first linearly transformed and then transmitted through the single selected antenna. The upper bound on pairwise error probability (PEP) for HSTTC is derived. It is shown that by a proper code construction HSTTC can achieve not only a full spatial diversity order, but also a time diversity order, as well as a significant coding gain determined by the weighted square product distance (WSPD). A new design criterion for HSTTCs has been proposed and the optimum 4-state code, based on this criterion, has been found by systematic code search. Simulation results are provided for the quaternary phase-shift keying (QPSK) signal sets. They show that the proposed new code is superior by 9 dB and 11 dB to the conventional transmit antenna selection (TAS) and Tarokh-Seshadri-Calderbank codes (TSC), respectively, at a BER of 10e/sup -4/ over time-varying fading channels.

The aim of this paper is to quantify the performance of a W-CDMA system in dependency of a giver power delay profile (PDP) and channel estimation errors. The analysis extends the previously published work by Adachi [F. Adachi, Nov. 1997], which assumes ideal channel estimation and multipath tracking. For the analytical expressions a particular channel estimation technique was taken into account and a practical pilot symbol assisted technique was used to derive the channel estimation error. By combining both, the performance degradation due to multiple access interference (MAI) and due to channel estimation errors, an easy to use conditional SINR is derived. Assuming BPSK data modulation for simplicity, the downlink performance in terms of the average BER-vs.-SINR was evaluated by means of Monte Carlo simulation for different PDPs. It could be shown that the performance severely depends on the power distribution of the multipath components and the available number of pilot symbols, which determines the reliability of the estimated channel necessary to perform coherent demodulation.

Common radio resource management strategies are devoted to achieve an efficient usage of the pool of radio resources available in a heterogeneous radio access network context. This paper describes the functionalities associated with the common vision of radio access technologies, the different possibilities in the functional model split and the corresponding implementation considerations

A radio location system operates by measuring physical quantities related to radio signals traveling between a mobile terminal (MT) and a set of fixed base stations (BSs). In order to reduce the implementation complexity and the signalling overhead, the solution presented makes use of a single BS to locate unmodified MTs. The novel algorithm applies a triangulation technique supported by minimal information about the environment in the BS neighborhood. The proposed algorithm is compatible with the architecture of 3G cellular networks, as it relies on CDMA (code division multiple access) signalling and a BS array antenna for the estimation of the time delay and of the angle of arrival, respectively. Despite its modest implementation complexity, the proposed algorithm can predict the MT position with a fairly good accuracy, as demonstrated by numerical tests performed by resorting to a fully 3D deterministic propagation model developed at the University of Pisa.

An extended form of multi-hop communication systems has been introduced recently which allows the application of multiple-input-multiple-output (MIMO) capacity enhancement techniques over spatially separated relaying mobile terminals. It was further shown that such deployment yields significant capacity gains over direct communication or traditional single-input-single-output (SISO) relaying networks. The contribution of this paper is the derivation of the end-to-end bit-error rates (BER) for space-time block encoded M-PSK and M-QAM, assuming full and partial cooperation at each relaying stage. The theoretical analysis is corroborated by simulation results.

The recently introduced concept of virtual antenna arrays (VAAs) promises a significant increase in link and system capacity. VAAs therefore constitute a promising candidate for ad-hoc oriented 4G networks. VAAs successfully emulate a multiple-input multiple-output (MIMO) transmission channel to which space-time codes (STC) can be applied. The realisation of VAAs is accomplished by letting spatially close mobile terminals communicate with each other, which overcomes the disadvantage of having only one receiving antenna in a mobile handset. The thus created ad-hoc network represents a general form of MIMO channels for which different link level performance results apply. The work presented analyses the link level performance of a VAA system with deployed higher order space-time block codes (STBCs) in dependency of the prevailing channel conditions within the generated ad-hoc network. As such, the influence of Rayleigh, Rice and Nakagami flat fading channels with a varying parameter set onto the performance of generic STBCs with associated VAA is investigated. It is known that for close antenna spacing and/or high Rice K-factors (Nakagami m-factors) the correlation at the receiving end increases, which is shown to be successfully mitigated by the deployment of VAAs.

In the B3G system considered, which is based on the 3GPP UMTS architecture, the initial hypothesis made in this paper, is that the QoS management of the IP access network is taken in charge by a bandwidth broker. A pre-handover signalling for QaS-aware path selection is investigated with the goal of achieving the "always best connected" paradigm. Furthermore, the scalability of the proposed scheme is analysed taking into account the overhead due to IP micromobility, QoS routing, CRRM and bandwidth broker

Flexible network architecture is envisioned as one of the properties of the next generation mobile networks, and is also foreseen to significantly contribute to lowering latency, and addressing Ultra-Reliable Low Latency Communications (URLLC). In this paper, we study the flexibility in Radio Access Network (RAN) configuration in terms of splitting the radio and baseband functionalities between central cloud and distributed entities, and the impact of this flexibility on delivery of URLLC. Three different functionality splits have been implemented in cloud- RAN, and the impact of each of these splits on communication latency and jitter is examined. We modeled traffic based on the 3GPP traffic models and, for completeness, in addition to URLLC, the two cases of massive machine-type communications (mMTC) and enhanced Mobile Broadband (eMBB) models are also implemented. Hence, thorough analyses are performed and recommendation for split point between central cloud and distributed radio units are discussed. Index Terms-5G; Cloud-RAN; functionality split; URLLC; eMBB; mMTC.