Antenna arrays in multi-user detection of spread spectrum signals
In a code division multiple access (CDMA) spread spectrum system, each user is assigned a unique scrambling code. If these codes are not mutually orthogonal, the resulting multiple access interference (MAI) causes frame error rates to be unacceptably high. A number of multi-user detection algorithms...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2007
|
| Subjects: | |
| Online Access: | http://eprints.um.edu.my/8787/1/Antenna_arrays_in_multi-user_detection_of_spread_spectrum_signals.pdf http://eprints.um.edu.my/8787/ http://www.scopus.com/inward/record.url?eid=2-s2.0-47949125568&partnerID=40&md5=e0a4e7a772382746b9087bed3683b3aa ieeexplore.ieee.org/xpls/absall.jsp?arnumber=4454747 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In a code division multiple access (CDMA) spread spectrum system, each user is assigned a unique scrambling code. If these codes are not mutually orthogonal, the resulting multiple access interference (MAI) causes frame error rates to be unacceptably high. A number of multi-user detection algorithms have been suggested by various authors. Optimal receivers based upon maximum likelihood sequence decoding are quite complex, their complexity increasing exponentially with the number of users in the system. Sub-optimal receivers such as linear detectors or adaptive cancellers are not as complex, but the error rates increase with the number of users. In this paper, we suggest a new multi-user detection procedure using an antenna array at the receiver. First, we provide an upper bound on the minimum mean-squared error due to MAI, and show how the frame error rate increases with the number of users. The transmitted symbols are estimated by inverting the transfer function of the channel, multiplying it with the outputs of matched filters, and then passing the result through a maximum likelihood decoder. Our analysis indicates that if the channel impulse response is known with sufficient accuracy, this procedure leads to an optimal design. In those cases where the impulse response is only approximately known, we estimate the transmitted symbols and compute the resulting baseband SNR. If this SNR is above a threshold, the baseband signal is acceptable and ready for maximum likelihood sequence decoding. Otherwise, the impulse response is adjusted in small steps until the SNR is above that threshold. |
|---|