Difference between revisions of "Riset-disertasi"
Jump to navigation
Jump to search
| Line 7: | Line 7: | ||
#* Payload size (bytes dikirim). | #* Payload size (bytes dikirim). | ||
#* Retransmisi (via pcap). | #* Retransmisi (via pcap). | ||
| + | |||
# QoE (Quality of Experience) | # QoE (Quality of Experience) | ||
| − | * SSIM/PSNR area non-ROI (harus ≈1 / ∞ → fidelity utuh). | + | #* SSIM/PSNR area non-ROI (harus ≈1 / ∞ → fidelity utuh). |
| − | * SSIM area ROI (rendah → privasi terlindungi). | + | #* SSIM area ROI (rendah → privasi terlindungi). |
| + | |||
# Cryptographic footprint | # Cryptographic footprint | ||
| − | * Distribusi bytes terenkripsi antar gambar/resolusi. | + | #* Distribusi bytes terenkripsi antar gambar/resolusi. |
| + | |||
# Privasi visual | # Privasi visual | ||
| − | * Apakah ROI (misalnya wajah) tak terbaca setelah enkripsi. | + | #* Apakah ROI (misalnya wajah) tak terbaca setelah enkripsi. |
------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------ | ||
| Line 19: | Line 22: | ||
==[[Riset-2]] (App-layer vs Transport-layer vs Hybrid) → Paper Q1 [port 5002 - 6002]== | ==[[Riset-2]] (App-layer vs Transport-layer vs Hybrid) → Paper Q1 [port 5002 - 6002]== | ||
Tujuan: membandingkan keamanan di layer aplikasi (ROI+Ascon) vs layer transport (OSCORE) vs kombinasi keduanya (cross layer). | Tujuan: membandingkan keamanan di layer aplikasi (ROI+Ascon) vs layer transport (OSCORE) vs kombinasi keduanya (cross layer). | ||
| − | + | # Perbandingan overhead antar layer | |
| − | + | #* Latency (CoAP vs OSCORE vs ROI+Ascon+OSCORE). | |
| − | + | #* Payload size (besar header tambahan OSCORE). | |
| − | + | #* CPU usage: ROI+Ascon (app) vs OSCORE (transport). | |
| − | + | ||
| − | + | # Reliabilitas komunikasi | |
| − | + | #* Packet retransmission / packet loss. | |
| − | + | #* Goodput (gambar/MBps efektif). | |
| − | + | ||
| − | + | # Security | |
| − | + | #* Transport: full payload protected, tapi tidak selective. | |
| − | + | #* App-layer: selective encryption, lebih hemat. | |
| − | + | #* Kombinasi: double protection, overhead lebih tinggi. | |
| − | + | ||
| + | # Platform diversity | ||
| + | #* Evaluasi di VM (GNS3) vs Raspberry Pi vs ESP32. | ||
| + | #* Apakah constraint hardware mempengaruhi performa signifikan. | ||
------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------ | ||
| Line 39: | Line 45: | ||
Tujuan: membuktikan NanoCipher sebagai AEAD ringan khusus multimedia. | Tujuan: membuktikan NanoCipher sebagai AEAD ringan khusus multimedia. | ||
Evaluasi inti (kriptografi + sistem): | Evaluasi inti (kriptografi + sistem): | ||
| − | + | # Benchmark kriptografi (MCU/CPU level) | |
| − | + | #* Cycles/byte (ukur dengan benchmark Ascon/PRESENT/ChaCha20). | |
| − | + | #* Throughput (MB/s). | |
| − | + | #* Code size (flash, KB). | |
| − | + | #* RAM usage. | |
| − | + | #* Energy/byte (estimasi via current measurement di Raspberry Pi/ESP32). | |
| − | + | ||
| − | + | # Keamanan algoritma | |
| − | + | #* Correctness: Known Answer Test (KAT). | |
| − | + | #* Nonce misuse resistance (uji basic). | |
| − | + | #* Konsistensi constant-time (cek dengan alat timing). | |
| − | + | ||
| − | + | # Perbandingan dengan cipher lain | |
| − | + | #* Ascon. | |
| − | + | #* PRESENT. | |
| − | + | #* ChaCha20. | |
| − | + | #* AES. | |
| − | + | ||
| + | # Integrasi | ||
| + | #* Kirim gambar (via libcoap) dengan NanoCipher. | ||
| + | #* Latency, payload, CPU, QoE → dibandingkan Ascon/AES/ChaCha20/PRESENT. | ||
------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------ | ||
| Line 63: | Line 72: | ||
Tujuan: menguji apakah NanoCipher bisa jadi alternatif AEAD di OSCORE (transport security). | Tujuan: menguji apakah NanoCipher bisa jadi alternatif AEAD di OSCORE (transport security). | ||
Evaluasi: | Evaluasi: | ||
| − | + | # Integrasi OSCORE | |
| − | + | #* Modifikasi libcoap → NanoCipher jadi AEAD opsional di COSE. | |
| − | + | #* Kinerja encode/decode OSCORE. | |
| − | + | ||
| − | + | # Perbandingan OSCORE-AES vs OSCORE-Ascon vs OSCORE-NanoCipher | |
| − | + | #* Latency per message. | |
| − | + | #* Payload overhead. | |
| − | + | #* CPU usage. | |
| − | + | #* Energy consumption. | |
| − | + | ||
| − | + | # Robustness | |
| − | + | #* Correctness (uji KAT + interop message decrypt). | |
| − | + | #* Replay protection (uji ulang id_context, sequence number). | |
| − | + | ||
| + | # Resource impact (IoT) | ||
| + | #* Memory footprint tambahan bila OSCORE pakai NanoCipher. | ||
| + | #* Apakah Raspberry/ESP32 masih muat (RAM/flash). | ||
Revision as of 09:17, 29 October 2025
Contents
Riset-1 (ROI + Ascon via CoAP) → Paper Conference [port 5001 - 6001]
Tujuan: menunjukkan keunggulan selective encryption untuk multimedia di IoT. Evaluasi:
- Kinerja sistem
- CPU usage / execution time enkripsi (per gambar, per resolusi).
- Latency end-to-end (ClientA → ServerB).
- Payload size (bytes dikirim).
- Retransmisi (via pcap).
- QoE (Quality of Experience)
- SSIM/PSNR area non-ROI (harus ≈1 / ∞ → fidelity utuh).
- SSIM area ROI (rendah → privasi terlindungi).
- Cryptographic footprint
- Distribusi bytes terenkripsi antar gambar/resolusi.
- Privasi visual
- Apakah ROI (misalnya wajah) tak terbaca setelah enkripsi.
Riset-2 (App-layer vs Transport-layer vs Hybrid) → Paper Q1 [port 5002 - 6002]
Tujuan: membandingkan keamanan di layer aplikasi (ROI+Ascon) vs layer transport (OSCORE) vs kombinasi keduanya (cross layer).
- Perbandingan overhead antar layer
- Latency (CoAP vs OSCORE vs ROI+Ascon+OSCORE).
- Payload size (besar header tambahan OSCORE).
- CPU usage: ROI+Ascon (app) vs OSCORE (transport).
- Reliabilitas komunikasi
- Packet retransmission / packet loss.
- Goodput (gambar/MBps efektif).
- Security
- Transport: full payload protected, tapi tidak selective.
- App-layer: selective encryption, lebih hemat.
- Kombinasi: double protection, overhead lebih tinggi.
- Platform diversity
- Evaluasi di VM (GNS3) vs Raspberry Pi vs ESP32.
- Apakah constraint hardware mempengaruhi performa signifikan.
Riset-3 (NanoCipher – AEAD baru) → Paper Q1 [port 5003 - 6003]
Tujuan: membuktikan NanoCipher sebagai AEAD ringan khusus multimedia. Evaluasi inti (kriptografi + sistem):
- Benchmark kriptografi (MCU/CPU level)
- Cycles/byte (ukur dengan benchmark Ascon/PRESENT/ChaCha20).
- Throughput (MB/s).
- Code size (flash, KB).
- RAM usage.
- Energy/byte (estimasi via current measurement di Raspberry Pi/ESP32).
- Keamanan algoritma
- Correctness: Known Answer Test (KAT).
- Nonce misuse resistance (uji basic).
- Konsistensi constant-time (cek dengan alat timing).
- Perbandingan dengan cipher lain
- Ascon.
- PRESENT.
- ChaCha20.
- AES.
- Integrasi
- Kirim gambar (via libcoap) dengan NanoCipher.
- Latency, payload, CPU, QoE → dibandingkan Ascon/AES/ChaCha20/PRESENT.
Riset-4 (NanoCipher + OSCORE) → Paper Q1 [port 5004 - 6004]
Tujuan: menguji apakah NanoCipher bisa jadi alternatif AEAD di OSCORE (transport security). Evaluasi:
- Integrasi OSCORE
- Modifikasi libcoap → NanoCipher jadi AEAD opsional di COSE.
- Kinerja encode/decode OSCORE.
- Perbandingan OSCORE-AES vs OSCORE-Ascon vs OSCORE-NanoCipher
- Latency per message.
- Payload overhead.
- CPU usage.
- Energy consumption.
- Robustness
- Correctness (uji KAT + interop message decrypt).
- Replay protection (uji ulang id_context, sequence number).
- Resource impact (IoT)
- Memory footprint tambahan bila OSCORE pakai NanoCipher.
- Apakah Raspberry/ESP32 masih muat (RAM/flash).
