IMPLEMENTASI JARINGAN SYARAF TIRUAN (JST) UNTUK PREDIKTOR DAYA PANCAR PADA SISTEM KOMUNIKASI BERGERAK
| Main Authors: | , Sigit Kusmaryanto, Ir, , Prof. Adhi Susanto, M.Sc., Ph.D. |
|---|---|
| Format: | Thesis NonPeerReviewed |
| Terbitan: |
[Yogyakarta] : Universitas Gadjah Mada
, 2012
|
| Subjects: | |
| Online Access: |
https://repository.ugm.ac.id/98042/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=53993 |
Daftar Isi:
- Mobile Communications Systems has evolved such that at present there has been Mobile Communications Systems Cellular, which utilizes multiple access technology Code Division Multiple Access (CDMA). Mobile Communications Systems CDMA Development emphasized on increasing the capacity of the channel. The capacity of the user terminal on Mobile Communications Systems CDMA systems such as DS / CDMA is limited by interference. One technique to overcome the interference in the system is to maintain the power emitted by each user terminal as low as possible and try to get the signals received by the Base Stations (BS) from all user terminals are at a relatively constant power and the same. To achieve this use the power settings to overcome the near-far effects. In this case, setting the feedback (Feedback Power Control) allows a BS sends a command to the user terminal power settings to lower or raise the level of transmission power such that the BS signal power levels can be maintained relatively constant and equal. The most important factor in the success of this is the process of handling the MS by the BS that is closely associated with the problem of predictive power on the BS. In this study discussed the application analyzes the structure of ANN is designed as a predictor of the transmit power level at Mobile Communications Systems DS / CDMA. The design begins with the generation of the MS signal at multipath channel model that meets the criteria the Poisson distribution and Reyleigh. Output signal of multipath channel model then used as ANN input. In this study discussed the application analyzes the structure of ANN is designed as a predictor of the power level at Mobile Communications Systems DS /CDMA. From the simulation results summarized ANN structures can be applied as a predictor of power of Mobile Communications Systems DS / CDMA, which has the ability to predict one step forward, as well as overcome the effects of near-far (far-near). ANN structure is formed by an Adaptive Linear Element (Adaline) and a Multi Layer Perceptron (MLP). Adaline structure was formed with 5 inputs and one output, while the MLP structure is formed with 20 inputs and 27 nodes on hidden layer and one output. Error obtained from simulation ANN structure that is error-noise = 1.72 percent and error MLP =1.98 percent