PEMODELAN SPIKE GENERATOR PADA SPIKING NEURAL NETWORK MENGGUNAKAN FIELD PROGRAMMABLE GATE ARRAY

Hilmi, Kafin Sulthana (2025) PEMODELAN SPIKE GENERATOR PADA SPIKING NEURAL NETWORK MENGGUNAKAN FIELD PROGRAMMABLE GATE ARRAY. Bachelor thesis, Institut Teknologi Kalimantan.

[img] Text
04211037_cover.pdf
Restricted to Repository staff only until 5 October 2027.

Download (399kB) | Request a copy
[img] Text
04211037_statement_of_authenticity.pdf
Restricted to Repository staff only until 5 October 2027.

Download (249kB) | Request a copy
[img] Text
04211037_publishing_agreement.pdf
Restricted to Repository staff only until 5 October 2027.

Download (347kB) | Request a copy
[img] Text
04211037_approval_sheet.pdf
Restricted to Repository staff only until 5 October 2027.

Download (278kB) | Request a copy
[img] Text
04211037_preface.pdf
Restricted to Repository staff only until 5 October 2027.

Download (6MB) | Request a copy
[img] Text
04211037_abstract_id.pdf
Restricted to Repository staff only until 5 October 2027.

Download (509kB) | Request a copy
[img] Text
04211037_abstract_en.pdf
Restricted to Repository staff only until 5 October 2027.

Download (418kB) | Request a copy
[img] Text
04211037_table_of_content.pdf
Restricted to Repository staff only until 5 October 2027.

Download (6MB) | Request a copy
[img] Text
04211037_illustrations.pdf
Restricted to Repository staff only until 5 October 2027.

Download (6MB) | Request a copy
[img] Text
04211037_tables.pdf
Restricted to Repository staff only until 5 October 2027.

Download (511kB) | Request a copy
[img] Text
04211037_notations.pdf
Restricted to Repository staff only until 5 October 2027.

Download (396kB) | Request a copy
[img] Text
04211037_chapter_1.pdf
Restricted to Repository staff only until 5 October 2027.

Download (3MB) | Request a copy
[img] Text
04211037_chapter_2.pdf
Restricted to Repository staff only until 5 October 2027.

Download (5MB) | Request a copy
[img] Text
04211037_chapter_3.pdf
Restricted to Repository staff only until 5 October 2027.

Download (4MB) | Request a copy
[img] Text
04211037_chapter_4.pdf
Restricted to Repository staff only until 5 October 2027.

Download (8MB) | Request a copy
[img] Text
04211037_conclusions.pdf
Restricted to Repository staff only until 5 October 2027.

Download (829kB) | Request a copy
[img] Text
04211037_bibliography.pdf
Restricted to Repository staff only until 5 October 2027.

Download (779kB) | Request a copy
[img] Text
04211037_enclosure.pdf
Restricted to Repository staff only until 5 October 2027.

Download (731kB) | Request a copy
[img] Text
04211037_paper.pdf
Restricted to Repository staff only until 5 October 2027.

Download (720kB) | Request a copy
[img] Text
04211037_presentation.pdf
Restricted to Repository staff only until 5 October 2027.

Download (3MB) | Request a copy
[img] Text
04211037_Form. TA-020.pdf
Restricted to Repository staff only until 5 October 2027.

Download (289kB) | Request a copy

Abstract

Spiking Neural Network (SNN) merupakan model jaringan yang meniru mekanisme kerja otak manusia melalui representasi impuls listrik atau spike. Salah satu komponen penting dalam SNN adalah Spike Generator, yaitu unit yang mengubah sinyal analog menjadi spike train sebagai input bagi sistem. Penelitian ini bertujuan merancang dan mengimplementasikan Spike Generator berbasis Field Programmable Gate Array (FPGA) dengan memanfaatkan algoritma Step Forward Spike Encoding untuk mencapai efisiensi dan akurasi tinggi dalam proses encoding dan rekonstruksi sinyal. FPGA Cyclone V digunakan sebagai platform utama, dan pengujian dilakukan terhadap beberapa jenis sinyal, antara lain sinusoidal 3 Hz, kombinasi sinusoidal 6 Hz dan 16 Hz, sinyal konstan naik, serta sinyal konstan turun. Hasil pengujian menunjukkan bahwa algoritma mampu menghasilkan spike secara akurat dengan baseline adaptif mengikuti perubahan dinamika sinyal. Implementasi pada Quartus menghasilkan kualitas rekonstruksi sinyal yang lebih unggul dibandingkan MATLAB, dengan nilai Signal-to-Noise Ratio (SNR) yang signifikan: 36,68 dB untuk sinyal sinusoidal 3 Hz, 93,69 dB untuk sinyal konstan naik, dan 88,54 dB untuk sinyal konstan turun. Jumlah spike dibatasi hingga 200 untuk memastikan efisiensi dan keterukuran kinerja sistem. Penelitian ini memperlihatkan bahwa pendekatan berbasis FPGA tidak hanya hemat energi dan fleksibel, tetapi juga mampu menjaga kualitas sinyal dengan akurasi tinggi, sehingga berpotensi besar dalam pengembangan sistem SNN.

Item Type: Thesis (Bachelor)
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Jurusan Teknologi Industri dan Proses > Teknik Elektro
Depositing User: Kafin Sulthana Hilmi
Date Deposited: 10 Jul 2025 05:39
Last Modified: 10 Jul 2025 05:39
URI: http://repository.itk.ac.id/id/eprint/23517

Actions (login required)

View Item View Item