Title :
Comparing DSP realizations of correlator and SVM receivers for chaos-based multi-user DS-SS
Author :
Vali, Ramin ; Kao, Johnny W. H.
Author_Institution :
Dept. of Comput., Unitec Inst. of Technol., Auckland, New Zealand
Abstract :
This contribution presents a comparison between two realizations of chaos-based multi-user direct-sequence spread spectrum (DS-SS) system in digital signal processor (DSP) technology. The first realization is a conventional correlator receiver and the second is a new support vector machine (SVM) receiver. The two realizations were aimed at examining the tradeoffs between the two receiver types and their suitability for implementation in large scale wireless communication networks. The performance benchmarks used are the bit-error-rate (BER) and system speed. In both systems the information is spread using chaotic sequences resulting in physical layer security in the time and frequency domains. The hardware platform chosen was a 32-bit floating-point TigerSHARC DSP which proved to be sufficient for the representation of chaos-based spreading codes. Both implementations show excellent agreement with the previous theoretical findings in terms of BER performance. This agreement of results proves that the finite number resolution of the DSP does not have a significant effect on the system performance. Overall it is shown that the SVM receiver outperforms the correlator receiver in terms of the BER. However this improvement is at the expense of the processing speed.
Keywords :
chaos; code division multiple access; correlators; digital signal processing chips; electronic engineering computing; floating point arithmetic; radio receivers; spread spectrum communication; support vector machines; telecommunication computing; SVM receiver; TigerSHARC DSP; bit error rate; chaos based multiuser DS-SS; chaotic sequence; correlator receiver; digital signal processor technology; direct sequence spread spectrum system; floating point DSP; physical layer security; support vector machine receiver; system speed; wireless communication; Bit error rate; Chaotic communication; Digital signal processing; Receivers; Support vector machines; Training; BER performance; Chaos-based DS-SS systems; DSP realization; machine learning; physical level security; support vector machine;
Conference_Titel :
Digital Signal Processing (DSP), 2014 19th International Conference on
Conference_Location :
Hong Kong
DOI :
10.1109/ICDSP.2014.6900708