A one-shot LDD with pre-SDC/post-MRC for random CDMA over asynchronous Rayleigh channels
| Autor: | Seong Rag Kim, Sok-Kyu Lee |
|---|---|
| Rok vydání: | 2002 |
| Předmět: |
Computer Networks and Communications
Computer science Code division multiple access business.industry Detector Multiuser detection Spread spectrum Signal-to-noise ratio Bit error rate Maximal-ratio combining Fading Electrical and Electronic Engineering Telecommunications business Decorrelation Algorithm Computer Science::Information Theory Communication channel |
| Zdroj: | IEEE Journal on Selected Areas in Communications. 20:257-263 |
| ISSN: | 0733-8716 |
| Popis: | It is well known that conventional detectors of direct sequence-code division multiple access systems are interference-limited. Multiuser detection has been considered as a promising approach to combat this problem. Since an optimal multiuser detector is too complex to implement, several suboptimal multiuser detector techniques have been developed. The one-shot linear decorrelating detector (LDD) was proposed to reduce the computational complexity and delay required in the conventional LDD, but it still require high complexity. Another one-shot LDD with low complexity was proposed. To improve its performance, this detector uses a maximal ratio combining (MRC) at the two outputs of the left piece of the present bit and the right piece of the next bit in an one-shot window; this detector is called as a one-shot LDD with post-MRC. In spite of the aid of MRC, the performance of the one-shot LDD with post-MRC is worse than the one-shot LDD. To overcome this problem, we propose the one-shot LDD with pre-SDC/post-MRC that employs selection diversity combining (SDC) with multiple antenna elements in front of the one-shot LDD with post-MRC. In asynchronous random code division multiple access systems over fading channels, the largest signal-to-noise ratio (SNR) of SDC, near-far resistance (NFR) of the one-shot LDD, and the output SNR of MRC are random variables. We derive the probability density functions (PDFs) for these random variables and calculate the bit error rate) of the proposed detector using the obtained PDFs. In a comparison of mathematical analysis and computer simulation results, we observe good agreement. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|