An expanded effort is underway to support the evolution of the UMTS standard to meet the rapidly developing needs associated with wireless Internet applications. The support of high speed downlink packet access (HSDPA) is provided by means of a new, shared channel called high speed downlink shared channel (HS-DSCH) which is enabled by a number of performance enhancing technologies such as adaptive modulation and coding (AMC), hybrid ARQ (HARQ), fat-pipe scheduling, fast cell selection (FCS) and multiple input multiple output (MIMO) antenna techniques. In this paper, the advantage of FCS is investigated. To avoid the difficulty in HARQ transmission-state synchronization, FCS is allowed only among the cells (sectors) belonging to the same base station. Two FCS approaches are compared in this paper. The user controlled cell selection (UCCS) selects the serving cell based only on the channel quality seen by the user. While a new-network controlled cell selection (NCCS) approach jointly considers channel quality and network loading. It is shown that FCS significantly improves the HSDPA performance at low mobility. Among the FCS schemes, NCCS is superior to UCCS when the load among the cells in the active set is unbalanced.

A link error prediction (LEP) methodology allows the abstraction of link performance into system level studies, so that system performance may be characterized efficiently. The quasi-static method (QSM) method (http://www.3gpp2.org) (R. Ratasuk et al., 2002) which assumes perfect channel quality estimates for demodulation, has been widely used for downlink system performance evaluation. However, unlike in the downlink, the per-user uplink pilot, which is power-controlled, is much weaker, so that channel estimation errors, and its effect on link performance (especially at high data rates), is considerably higher. In this paper we propose a simple and efficient LEP method, which we call the equivalent SNR method (ESM), for analytically capturing the effect of channel estimation errors on link performance. ESM is shown to he extremely accurate over a wide range of pilot SNRs. We note that ESM was adopted for LEP for cdma2000 (EVDV) and UMTS system simulations.

An expanded effort is underway to support the evolution of the UMTS and cdma2000-1x standards to meet the rapidly developing needs associated with wireless Internet applications. A number of performance enhancing technologies are proposed to ensure high peak and average packet data rates while supporting circuit-switched voice and packet data on the same spectrum. These techniques include adaptive modulation and coding (AMC), hybrid ARQ (H-ARQ) and fat-pipe scheduling. In order to enable these techniques downlink channel quality feedback (CQF) through explicit uplink signaling is necessary. Frequent CQF results in good estimates of downlink channel quality at the base station, which, in turn, improves downlink system performance. However, this comes at the expense of larger uplink signaling overhead, thereby impacting the overall uplink capacity. Infrequent CQF, on the other hand, reduces this signaling overhead in the uplink at the expense of larger errors in the channel quality estimates available at the base station, thereby leading to system performance degradation. In this paper, we present a variable rate CQF scheme that significantly reduces uplink signaling overhead without affecting downlink system performance. In addition to simulation results demonstrating the performance of the proposed scheme, we also compare the reduction in uplink signaling overhead of the said scheme with other known methods.

In anticipation of the high demand for wireless data services, several wireless standards are evolving to support packed data more efficiently. In UMTS, a new, shared channel called the high speed downlink shared channel (HS-DSCH) that multiplexes packet data users has been defined. Several performance-enhancing technologies are included in high speed downlink packet access (HSDPA): adaptive modulation and coding (AMC), hybrid ARQ (HARQ), scheduling etc. HARQ algorithm plays a critical role in HSDPA performance. HARQ techniques are classified in terms of the combining method (chase or incremental redundancy, IR), protocol timing (synchronous or asynchronous) and adaptivity (if modulation/code-rate can be changed for retransmission or not). In previous work (Das et al. 2001), an adaptive, asynchronous, incremental redundancy (A/sup 2/IR) scheme was proposed and shown to provide substantial gains over other HARQ options. In this paper, the design and performance aspects of A/sup 2/IR in the code domain (as against the time domain in Das et al.) are considered. Due to the higher bandwidth of the UMTS channels compared to cdma2000 or HDR, it is expected that code-division multiplexing (CDM) of users will be needed to "fill the data pipe" instead of pure time-division multiplexing. The paper also analyzes tradeoffs that are unique to implementing A/sup 2/IR with CDM. A/sup 2/IR with a fixed TTI is now part of UMTS Release 5.

An expanded effort is underway to support the evolution of the UMTS standard to meet the rapidly developing needs associated with wireless Internet applications. The support of high-speed downlink packet access (HSDPA) is provided by means of a new, shared channel called high speed downlink shared channel (HS-DSCH) which is enabled by a number of performance enhancing technologies such as adaptive modulation and coding (AMC), hybrid ARQ (HARQ), fat-pipe scheduling, fast cell site selection (FCSS) and multiple input multiple output (MIMO) antenna techniques. To fully benefit from these technologies, one must guarantee fast and reliable control message exchange between the base station and the user terminals. The downlink signaling is done through the use of shared control channels (SCCH) accompanied with each HS-DSCH. Since the SCCHs, HS-DSCH, and voice channels share the same resources (power, bandwidth), control signalling is often improved at the cost of system resources and capacity. In this paper, we evaluate the performance of the SCCH. The challenges and tradeoffs in the SCCH design will also be discussed.

We determine the capacity regions for a class of time-varying multiple-access channels (TVMACs), when the underlying channel state evolves in time according to a probability law which is known to the transmitters and the receiver. Additionally, the transmitters and the receiver have access to varying degrees of channel state information (CSI) concerning the condition of the channel. Discrete-time channels with finite input, output, and state alphabets are considered first. The special case of a TVMAC, with the channel state process being a time-invariant, indecomposable, aperiodic Markov chain, shows a surprising anomaly in that imperfect transmitter CSI can cause the capacity under some distributions for the initial state to be strictly larger than that under a stationary distribution for the initial state. We also study a time-varying multiple-access fading channel with additive Gaussian noise, when various amounts of CSI are provided to the transmitters and perfect CSI is available to the receiver, and the fades are assumed to be stationary and ergodic. Implications for transmitter power control are discussed.

Summary form only given. We address the capacity problem for a class of time-varying multiple-access channels (TVMAC), when the underlying channel state evolves in time according to a probability law which is known to the transmitters and the receiver. Additionally, the transmitters and the receiver have access to varying degrees of channel state information (CSI) concerning the condition of the channel. Discrete-time channels with finite input, output and state alphabets are considered first. The special case of a memoryless TVMAC, with the channel state process being a time-invariant, indecomposable, aperiodic Markov chain, shows a surprising anomaly in that imperfect transmitter CSI can cause the capacity under some distributions for the initial state to be strictly larger than that under a stationary distribution for the initial state. We also consider a time-varying multiple-access fading channel with additive Gaussian noise, when various amounts of CSI are provided to the transmitters and perfect CSI is available to the receiver, and the fades are assumed to be stationary and ergodic. Implications for transmitter power control are discussed.

An expanded effort is underway to support the evolution of the UMTS standard to meet the rapidly developing needs associated with wireless Internet applications. The support of packet-switched high-speed data users is provided by means of a new, shared channel called HS-DSCH (high speed downlink shared channel) that preferably serves one packet data user at a time in a time-multiplexed manner. A number of performance enhancing technologies are included in the high speed downlink packet access (HSDPA) proposal to ensure high peak and average packet data rates while supporting circuit switched voice and packet data on the same spectrum. These techniques include adaptive modulation and coding (AMC), hybrid ARQ (HARQ), fat-pipe scheduling, fast cell site selection (FCSS) and multiple input multiple output (MIMO) antenna techniques. In this paper, we provide the performance comparison of different HARQ schemes being considered for HSDPA.