DocumentCode :
694476
Title :
Collision-driven random multiuser opportunistic spectrum access
Author :
Yumeng Wang ; Liang Shen ; Chenglong Xu ; Haitao Zhang ; Xiaohua Dong ; Yang Chen
Author_Institution :
Inst. of Commun. Eng., PLA Univ. of Sci. & Technol., Nanjing, China
fYear :
2013
fDate :
12-13 Oct. 2013
Firstpage :
792
Lastpage :
796
Abstract :
In this paper, the problem of distributed multiuser Opportunistic Spectrum Access is studied based on Partially Observable Markov Decision Process (POMDP), where the channels are partially observed by secondary users (SUs) constrained by hardware and energy. Considering the similarity of spectrum environment, the decision of every secondary user may be the same, which leads to collisions. Comparing with the previous works, we propose a “collision-driven random” strategy, where the activities of secondary users are relatively steady but changing over time. If a secondary collides with others when transmitting data in one time slot, it may take the action of selecting a channel randomly according to choosing probability at next time slot. Otherwise, the secondary user stays on the current channel. Observing the simulation results, the upgrade of performance is up to 100% compared with the existing strategies.
Keywords :
Markov processes; multi-access systems; probability; radio spectrum management; telecommunication congestion control; choosing probability; collision driven random strategy; collision driven spectrum access; distributed spectrum access; opportunistic spectrum access; partially observable Markov decision process; random multiuser spectrum access; time slot selection; Cognitive radio; Probability; Sensors; Simulation; Throughput; Vectors; Cognitive Radio; Distributed Multiuser; Opportunistic Spectrum Access; POMDP;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Science and Network Technology (ICCSNT), 2013 3rd International Conference on
Conference_Location :
Dalian
Type :
conf
DOI :
10.1109/ICCSNT.2013.6967226
Filename :
6967226
Link To Document :
بازگشت