|
 |
ISSN 1996-1065 [Online] |
|
|
Volume 1 (2) 2007 |
| Title: | Neural Network Based BER Prediction for 802.11 Network |
| Authors: | Gowrishankar and P.S. Satyanarayana |
| Published: | ©IJCIR Vol1 (2) 2007, PP. 19-30 |
| Language: | English |
Abstract:
Bit Error Rate (BER) will enumerate the Channel State Information (CSI) in wireless network. Accurate and timely estimation of CSI will guarantee the Quality of Service (QoS) by admission control, inter and intra network handovers. Here the BER of time varying 802.11 wireless channel is predicted by neural network system. The wireless channel is modeled as time variant nonlinear system. The neural network systems are the best suitable paradigm to predict and analyze the behaviors of time varying nonlinear system. In this framework BER is predicted by two different recurrent neural network architectures such as Recurrent Radial Basis Function Network (RRBFN) and Echo State Network (ESN). The Prediction accuracy RRBFN and ESN are in the range of 83.6 % to 98 % and 86.1% to 99.6 % respectively.
View full Article
| General Terms: | Measurement, Performance, Experimentation, Theory
Additional Key Words and Phrases: Bit error rate, Channel state information, Modulation, Nonlinear time varying system, Quality of service, Recurrent neural networks
|
| Categories and Subject Descriptors: | C.2.1 [Computer-Communication Networks]: Network Architecture and Design-wireless Networks; C.2.5 [Computer-Communication Networks]: Local and Wide-Area Networks- Access schemes; I.2.6 [Artificial Intelligence]: Learning – Connectionism and neural nets. |
|
|
|
|
