Showing papers by "Preben Mogensen published in 2005"

Extension of the ITU channel models for wideband (OFDM) systems

Abstract: This paper outlines a procedure to extend, or upsample, the ITU power delay profiles (PDP) to improve the frequency correlation properties while keeping the same mean delay and almost the same rms delay spread. Realistic frequency correlation properties are of particular importance for the evaluation of wideband system concepts with frequency dependent characteristics, e.g. frequency domain link adaptation and packet scheduling, both of which are likely to be part of future wideband systems such as based on OFDM. With the suggested procedure the frequency correlation can be kept approximately at or below 0.6 for frequency separations up to 25 MHz for the exemplified ITU Vehicular A and Pedestrian B channel profiles. The compatibility in terms of link level performance with WCDMA rake processing and the representation of the profiles on different sampling grids is briefly discussed in the paper. ITU Channel models; frequency correlation function; wideband systems; power delay profile; beyond 3G system concepts

Performance of VoIP on HSDPA

Abstract: This paper provides packet scheduler design and performance simulations for running VoIP services over high-speed downlink packet access (HSDPA) in WCDMA. The main challenge of supporting VoIP service on HSDPA is the tight delay requirement combined with the small VoIP packet size. A packet scheduler design incorporating VoIP packet aggregation and user multiplexing is proposed and the VoIP capacity is studied for a macro-cellular environment. Results are obtained for different delay budgets and packet scheduling settings, using either blind round robin or a slightly modified version of proportional fair scheduling. For proportional fair scheduling with code-multiplexing of 4-users, the downlink VoIP cell capacity on HSDPA is found to be in the range 72-104 users depending on whether the delay budget for the node-B scheduling and user reception equals 80 ms or 150 ms, respectively.

Mapping strategy for OFDM-based systems using H-ARQ

Abstract: A wireless communication device adapted so that, in response to a repeat request indicating a communication signal, the device wirelessly retransmits the communication signal using a remapping of bits into at least partially different frequencies in the retransmission compared to the original transmission.

UTRAN Long Term Evolution in 3GPP

Abstract: The 3rd Generation Partner Ship Project (3GPP) produced the first version of WCDMA standard in the end of 1999, which is the basis of the Universal Mobile Telephone System (UMTS) deployed in the field today. This release, called release 99, contained all the basic elements to meet the requirements for IMT-2000 technologies. Release 5 introduced the high speed downlink packet access (HSDPA) in 2002, enabling now more realistic 2 Mbps and even beyond with data rates up to 14 Mbps. Further Release 6 followed with high speed uplink packet access (HSUPA) in end of 2004, with market introduction expected in 2007. Alongside with on-going further WCDMA development, work on evolved universal terrestrial radio access (UTRA) has been initiated in 3GPP. The objective of evolved UTRA is to develop a framework for the evolution of the 3GPP radio-access technology towards wider bandwidth, lower latency and packet-optimized radio-access technology with peak data rate capability up to 100 Mbps. This paper introduces the requirements, the current state of progress in 3GPP, findings on the performance, agreed architecture as well as expected schedule for actual specification availability

Qos differentiation for wcdma services mapped onto an e-dch channel

Abstract: A SYSTEM HAS USER EQUIPMENT (1; 2) EACH HAVING BUFFERS (3, 4, 5, 6; 7, 8, 9, 10) FOR STORING PACKETS CLASSIFIED ACCORDING TO PRIORITY AND READY FOR TRANSMISSION OVER A RADIO INTERFACE TO A NETWORK ELEMENT (12) IN SUCH A SYSTEM, USER EQUIPMENT SENDS CAPACITY REQUEST SIGNALS (11; 13) INDICATIVE OF THE VARIOUS PRIORITIES OF THE BUFFERED PACKETS ON THE RADIO INTERFACE TO THE NETWORK ELEMENT RATHER THAN SEEKING A CAPACITY ALLOCATION DECISION FROM A HIGHER LEVEL IN THE CORE NETWORK, THE NETWORK ELEMENT ITSELF MAKES THE CAPACITY ALLOCATION DECISION AND PROVIDES A CAPACITY ALLOCATION SIGNAL (14) TO THE USER EQUIPMENT PACKETS THAT HAVE THUS BEEN ALLOCATED CAPACITY BY THE NETWORK ELEMENT ARE THEN SENT FROM THE USER EQUIPMENT TO THE NETWORK ELEMENT HAVING THE NETWORK ELEMENT MAKE THE CAPACITY ALLOCATION DECISIONS ITSELF is A MORE EFFICIENT SYSTEM SINCE THE NETWORK ELEMENT IS IN DIRECT COMMUNICATION WITH THE USER EQUIPMENT OVER THE RADIO INTERFACE AND THE DECISION IS MADE IN AN UNMEDIATED WAY

Resource allocation in a communication system

Abstract: A method of allocating resources in a communication system and a station for the communication system is disclosed. The communication system is such that a plurality of user equipment can communicate data on a dedicated channel. In the method a request for communication resources may be sent from a user equipment to the station on a first protocol layer. It may then be detected that priority based resource allocation is required, where after priority information is obtained at the station from a data flow on a second protocol layer. Resource allocation for comunicatioan on the first protocol layer may then at least partially be based on said priority information.

Transmitting data in a wireless network

Abstract: A method is described for transmitting data, particularly but not exclusively packet data, from a transmitting station to a receiving station via a wireless channel. The data is transmitted in transmission intervals or TTI slots. The method comprises estimating a utilisation factor representing usage of the transmission intervals, and scheduling the data for transmission to increase the utilisation factor. The effect is to reduce stochastic pattern interference for other users in a wireless communications network. A network node for implementing the above method is also disclosed.

Transmitting information using multiple carriers

Abstract: A communication network has a network element adapted to transmit data to a mobile station using at least two carriers. The network element transmits a first set of channels on a first main carrier, and a second set of channels on a second dependent carrier. The first and second set of channels may include a shared downlink channel, but are otherwise different each other.

RESPECT: a real-time emulator for service performance evaluation of cellular neTworks

Abstract: The evaluation and optimization of packet service performance in wireless networks is a complex process, consid- ering the number of heterogeneous entities and protocols that are involved. An emulation platform comes at hand for such performance investigations as it provides a means to see the performance from an end-2-end (E2E), user-perceived Quality Of Service (QoS) point of view. In this paper, the design and implementation of RESPECT, an easily configurable network emulator is described. RESPECT was originally geared towards Universal Mobile Communications System (UMTS) networks, but thanks to its modular and scalable design, it is being extended for generic heterogeneous networks. Using RESPECT, QoS studies can be carried out to study the behavior of different services in different network conditions, identify generalized service dependent performance metrics for already existing services and predict network environment requirements for future services.

Interference cancellation and 4-branch antenna diversity for WCDMA uplink packet access

Abstract: This paper addresses the performance gain of interference cancellation (IC) and 4-branch antenna diversity for enhanced packet data access in WCDMA uplink. A theoretical analysis is derived showing that the cell throughput gain from IC mainly depends on three factors: 1) the efficiency of the multi-user detection (MUD) receiver, 2) the other-to-own cell interference ratio, and 3) the uplink load. The gain from 4-branch antenna diversity and IC is then estimated bv means of system level simulations for two different node B scheduling concepts. Given specific outage constraints for both network load and user performance, the capacity gain from 4-branch compared to 2-branch antenna diversity is approx. 100% The cell throughput gain from IC is approx. 20% for an IC efficiency of 30% and 50% for an IC efficiency of 70% The estimated gain numbers from IC are almost independent of the considered scheduling scenario and are in accordance with the theoretical results. In the most optimistic case, the total cell throughout gain from fast node B scheduling, IC and 4-branch antenna diversity is 364% compared to a system implementation with RNC scheduling, no interference cancellation and 2-branch antenna diversity.

Investigation of frequency-domain link adaptation for a 5-MHz OFDMA/HSDPA system

Abstract: In this paper, we investigate frequency domain link adaptation (FDLA), e.g. utilizing the frequency selectivity of the channel in an OFDMA system. To make the study specific and based on realistic parameters, we re-use the specifications from a recent 3GPP 5-MHz OFDMA study item. The link adaptation and the frequency domain link adaptation are developed in a way compliant with the basic HSDPA specifications. With FDLA we show up to 75% cell throughput gain over the OFDMA reference system at the cost of increased uplink channel quality signaling overhead for frequency selective channels. We find that optimum waterfilling power distribution only provides a marginal gain over a simpler on/off equal power distribution algorithm per sub-carrier pool when signaling imperfections are taken into account.

The Impact of RLC Delivery Sequence on FTP Performance in UMTS

Abstract: The Radio Link Control (RLC) protocol of Universal Mobile Telecommunication System (UMTS) provides an option for inor out-of-sequence delivery of Service Data Units (SDUs) to upper layers. In this paper, the impact of this setting on the performance of File Transport Protocol (FTP) sessions is investigated. The investigations are carried out using a real-time emulation platform, and Transmission Control Protocol (TCP) with and without Selective Acknowledgement/Forward Acknowledgment (SACK/FACK) options. The results show that out-of-sequence delivery has a negative impact at higher bandwidths and TCP FACK exacerbates the problem. This is mainly due to the aggressive retransmission strategy employed by FACK, which considers the gaps in a SACK report caused by reordering as lost segments and retransmits them immediately.

Emulation based performance investigation of FTP file downloads over UMTS dedicated channels

Abstract: The Radio Link Control (RLC) protocol of Universal Mobile Telecommunication System (UMTS) provides link layer reliability that could mitigate the effects of the hostile radio propagation channel on packet data transmission. In this paper, the impact of some of the RLC reliability mechanisms on the performance of File Transport Protocol (FTP) is investigated. The investigations are carried out using a real time emulation platform, which makes the results from this study more realistic than simulation or simplified analytical studies as the overall End-2-End performance is analyzed involving real world protocol implementations.

Impact of Cyclic Prefix length on OFDM system Capacity

Abstract: This paper is a study on the impact of the Cyclic Prefix (CP) length on the downlink Capacity in a base-band synchronized SISO-OFDM context. To measure this impact, the capacity, measured in bits per second per hertz, is chosen as quality parameter. The study shows how the lengthening of the CP affects Spectral Efficiency Loss (SEL) and Signal to Interference and Noise Ratio (SINR). SINR can be mapped into Capacity using Shannons formula. The optimum CP length is defined as the one that maximises the Capacity for a given set of system parameters. The parameters studied in this paper are: 1) the useful OFDM symbol duration, 2) the Signal to Noise Ratio (SNR) at the receiver and 3) the channel Power Delay Profile (PDP). Depending on the values of these parameters different optimum CP lengths are obtained. For a system using only one value of CP length we suggest an optimum value to be 4�Ps for an OFDM symbol length of 40�Ps and 6�Ps for an OFDM symbol length of 80�Ps.

Effects of Dominant Other-Sector Interference on Multi-Antenna HSDPA Performance

Abstract: Link level SINR simulation results and network level sector throughput simulation results that quantify the benefit of dual antenna MMSE reception in a macrocellular HSDPA system are provided. The dual antenna RAKE receiver serves as baseline reference. Link-level simulation results are confirmed by a novel analytical expression that in flat Rayleigh fading and for uncorrelated rx antenna branches provides spatial interference suppression mean SINR gain as a function of the dominant other-sector interference ratio (DIR). It is shown that the MMSE receiver’s spatial interference suppression gain depends heavily on the amount of experienced DIR. The higher the DIR the higher is the SINR gain. Nevertheless, seen on network level SINR gain turns into moderate sector throughput gain, well below 50%. This is due to the fact that high DIR situations are rare in the investigated macrocellular scenario. Moreover, dynamic range limitations hinder translation of the full SINR gain into sector throughput.