Implementasi dan Analisis Performa Algoritma Kriptografi Hybrid Elliptic Curve Diffie-Hellman (ECDH) dan ChaCha20 untuk Keamanan Pesan

Fadillah, Aini Rini Yara (2025) Implementasi dan Analisis Performa Algoritma Kriptografi Hybrid Elliptic Curve Diffie-Hellman (ECDH) dan ChaCha20 untuk Keamanan Pesan. Other thesis, Instittut Teknologi Sepuluh Nopember.

Text 5002211173-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB)

Abstract

Keamanan komunikasi menjadi isu utama, terutama dalam menjaga kerahasiaan data pribadi. Algoritma konvensional seperti RSA dan Diffie-Hellman rentan terhadap serangan kriptografi modern serta kurang efisien dalam waktu dan memori. Sebagai solusi, algoritma Elliptic Curve Diffie-Hellman (ECDH) dan ChaCha20 menawarkan keamanan dan efisiensi lebih baik dalam segi waktu dan memori. ECDH digunakan untuk menghasilkan shared secret, yang selanjutnya diproses menggunakan HKDF-SHA256 dan kemudian digunakan dalam enkripsi simetris ChaCha20. Penelitian ini menggunakan kurva SECP256R1 dan diimplementasikan melalui simulasi Python GUI dengan input keyboard dan file, dan diuji pada dua perangkat dengan spesifikasi CPU berbeda. Parameter pengujian meliputi waktu eksekusi, memori, segmentasi panjang plaintext, dan Avalanche Effect. Hasil menunjukkan waktu perhitungan shared secret sebesar 0,139 ms dan 0,184 ms. Enkripsi ChaCha20 menunjukkan korelasi tinggi antara waktu dan ukuran plaintext (R2 = 0,9966 dan 0,8984). Memori puncak ECDH memperoleh nilai sebesar 0,27 KB setelah inisialisasi awal dan memori puncak ChaCha20 bersifat konsisten dengan ukuran data, serta uji Avalanche menunjukkan nilai ideal 49–50%, membuktikan sensitivitas tinggi algoritma terhadap perubahan input. Penelitian ini menujukkan bahwa kombinasi ECDH dan ChaCha20 menawarkan waktu eksekusi yang relatif cepat dan penggunaan memori yang efisien dalam sistem kriptografi, serta menjadikan algoritma kriptografi ini layak digunakan dalam sistem komunikasi aman.

========================================================================================================================

Communication security is a major issue, especially in maintaining the confidentiality of personal data. Conventional algorithms such as RSA and Diffie-Hellman are vulnerable to modern cryptographic attacks and are inefficient in terms of time and memory. As a solution, the Elliptic Curve Diffie-Hellman (ECDH) and ChaCha20 algorithms offer better security and efficiency in terms of time and memory. ECDH is used to generate a shared secret, which is then processed using HKDF-SHA256 and subsequently used in ChaCha20 symmetric encryption. This study uses the SECP256R1 curve and is implemented through a Python GUI simulation with keyboard and file inputs, and tested on two devices with different CPU specifications. Test parameters include execution time, memory, plaintext length segmentation, and Avalanche Effect. Results show shared secret calculation times of 0.139 ms and 0.184 ms. ChaCha20 encryption shows a high correlation between time and plaintext size (R2 = 0.9966 and 0.8984). The peak memory of ECDH reached 0.27 KB after initialization, and the peak memory of ChaCha20 was consistent with the data size. The Avalanche test yielded an ideal value of 49–50%, demonstrating the algorithm's high sensitivity to input changes. This study shows that the combination of ECDH and ChaCha20 offers relatively fast execution time and efficient memory usage in cryptographic systems, making this cryptographic algorithm suitable for use in secure communication systems.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Kriptografi, Elliptic Curve Diffie Hellman, ChaCha20, Shared Secret, Enkripsi Cryptography, Elliptic Curve Diffie Hellman, ChaCha20, Shared Secret, Encryption
Subjects:	Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA76.9.A25 Computer security. Digital forensic. Data encryption (Computer science)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics
Depositing User:	Aini Rini Yara Fadillah
Date Deposited:	04 Aug 2025 03:56
Last Modified:	04 Aug 2025 03:56
URI:	http://repository.its.ac.id/id/eprint/126630

Actions (login required)

View Item