Farideh Khaleghi

Transmission of high speed data in cdma2000

Requirements for CDMA high-speed data (HSD) services make one of the distinguish marks between second generation and third generation (3G) wireless systems. cdma2000 provides 3G capacity, high data rates of up to 2 Mb/s and services. This paper describes a generalized frame structure for the support of sub-frames within cdma2000. This structure is parameterized such that it can be configured for efficient operation, depending on the radio environment and mobile station complexity, in order to provide several degrees of freedom. Impact on both convolutional code and turbo-code physical layer implementation is investigated. The results show the benefits of using the sub-frames technique.

On symbol-based turbo codes for cdma2000

Turbo codes have enjoyed a great attention. Among different soft-output decoding algorithms of turbo codes only maximum a posteriori (MAP) decoding as an iterative soft-output decoder (BCJR) allows for achieving an acceptable BER performance at Eb/No levels within only 1 dB of the value corresponding to the Shannon capacity. The drawback of the MAP algorithm is the excessive memory required. Here, we investigate the so-called symbol-based turbo codes. These codes allow for a reduced required memory by 30% for the BCJR method.

A new algorithm to reduce the deviation in the base stations transmitted powers during soft handoff in CDMA cellular systems

Soft handoff (SHO) is one of the critical components that determine the performance of a CDMA cellular system. The mobile station in 3G systems send power control commands to the base stations to adjust their powers. These power control commands can be received in error by on of the base stations in the active set serving the mobile. This results in the base stations transmitted powers deviating. Reducing the deviation in the base stations transmitted powers helps to achieve the desired diversity gain from SHO. We propose to change the way the base stations respond to power control commands. This does not require changing the way the mobile issues the power control commands. From the simulation results we see that at a speed of 5 km/h, the probability that the difference in the two base stations transmitted powers being greater than 5 dB is about 25% using the conventional scheme while it is only 8% for the proposed scheme (majority voting over two commands). We notice also that the base stations average transmitted power is reduced by employing the majority voting scheme.