In recent years, Internet of Things has developed rapidly, and now has penetrated into human life and industrial production. It is speculated that the internet of things will become ubiquitous in the future, which will bring a series of problems. First, the large number of things will lead to operated system and software updates consuming a lot of manpower and resources. Another problem is the Internet of things facing security issues, in recent years for the means of Internet of things and tools have been increasing largely. Therefore, to achieve a secure automatic update on the Internet of Things is essential. This report will follow such an automatic update system based on Internet of things to expand. First it elaborated on the main motive of this problem, found three existing related works and three security methods for communication to analyze. Then combined results of analysis, put forward own a secure automatic update solution: manager and devices connect and mutual authentication in real time, at the same time, the manager will regularly check the database to see if there is new version application. When the administrator uploads a new version, the manager will download the version and then sends to all devices, then device installs and finally restart itself. Next, the report described how to implement this system in detail and evaluated it. In the end, this report summarized and introduces the future work.

Οι κρυπτογραφικές συναρτήσεις κατακερματισμού αποτελούν στις μέρες μας ένα από τα δημοφιλέστερα συστατικά των κρυπτογραφικών συστημάτων, λόγω των ιδιαίτερων ιδιοτήτων τους. Λαμβάνοντας υπόψη τη συνεχή αύξηση του όγκου δεδομένων και των ταχυτήτων επικοινωνίας, η χρήση μιας συνάρτησης κατακερματισμού με χαμηλή ρυθμοαπόδοση μπορεί να επιβραδύνει το συνολικό ψηφιακό τηλεπικοινωνιακό σύστημα. Ο σχεδιασμός ενός δεδομένου αλγορίθμου κατακερματισμού ώστε να έχει τη βέλτιστη ρυθμοαπόδοση αποτελεί ζήτημα μεγάλης σημασίας. Στη συγκεκριμένη διπλωματική εργασία παρουσιάζεται μια μεθοδολογία σχεδιασμού με στόχο τη βέλτιστη ρυθμοαπόδοση κρυπτογραφικών αλγορίθμων που βασίζονται σε συγκεκριμένη επαναληπτική μορφή. Για το σκοπό αυτό αναπτύχθηκε ένα λογισμικό, που συνδυάζει δύο τεχνικές, τον επαναχρονισμό και την ξεδίπλωση, παράγοντας το βέλτιστο σχεδιαστικό αποτέλεσμα. Η μεθοδολογία εφαρμόστηκε σε δύο δημοφιλείς συναρτήσεις κατακερματισμού, τις SHA-1 και SHA-256. Οι μετασχηματισμένοι αλγόριθμοι συνθέθηκαν και υλοποιήθηκαν σε FPGA, επιβεβαιώνοντας την αποτελεσματικότητα της μεθόδου. / Nowadays, cryptographic hash functions are one of the most popular primitive components in the cryptographic systems, due to their key features. Considering that data sizes and communication speeds are increasing every year, the use of a hash algorithm with low throughput can be a bottle neck in the digital communication system. Designing a given hash algorithm to be throughput optimum is a critical issue. In this diploma thesis a design methodology is presented which oprimizes the throughput of cryptographic hash functions that rely on a specific iterative structure. For this purpose, a software was designed combining two techniques, retiming and unfolding, that generates the optimal throughput design. The methodology was applied to two popular hash algorithms, SHA-1 and SHA-256. The transformed algorithms were synthesized and implemented in a FPGA device, confirming its effectiveness.

Spacecrafts are extensively used by public and private sectors to support a variety of services. Considering the cost and the strategic importance of these spacecrafts, there has been an increasing demand to utilize strong cryptographic primitives to assure their security. Moreover, it is of utmost importance to consider fault tolerance in their designs due to the harsh environment found in space, while keeping low area and power consumption. The problem of recovering spacecrafts from failures or attacks, and bringing them back to an operational and safe state is crucial for reliability. Despite the recent interest in incorporating on-board security, there is limited research in this area. This research proposes a trusted hardware module approach for recovering the spacecrafts subsystems and their cryptographic capabilities after an attack or a major failure has happened. The proposed fault tolerant trusted modules are capable of performing platform restoration as well as recovering the cryptographic capabilities of the spacecraft. This research also proposes efficient fault tolerant architectures for the secure hash (SHA-2) and message authentication code (HMAC) algorithms. The proposed architectures are the first in the literature to detect and correct errors by using Hamming codes to protect the main registers. Furthermore, a quantitative analysis of the probability of failure of the proposed fault tolerance mechanisms is introduced. Based upon an extensive set of experimental results along with probability of failure analysis, it was possible to show that the proposed fault tolerant scheme based on information redundancy leads to a better implementation and provides better SEU resistance than the traditional Triple Modular Redundancy (TMR). The fault tolerant cryptographic primitives introduced in this research are of crucial importance for the implementation of on-board security in spacecrafts.

Spacecrafts are extensively used by public and private sectors to support a variety of services. Considering the cost and the strategic importance of these spacecrafts, there has been an increasing demand to utilize strong cryptographic primitives to assure their security. Moreover, it is of utmost importance to consider fault tolerance in their designs due to the harsh environment found in space, while keeping low area and power consumption. The problem of recovering spacecrafts from failures or attacks, and bringing them back to an operational and safe state is crucial for reliability. Despite the recent interest in incorporating on-board security, there is limited research in this area. This research proposes a trusted hardware module approach for recovering the spacecrafts subsystems and their cryptographic capabilities after an attack or a major failure has happened. The proposed fault tolerant trusted modules are capable of performing platform restoration as well as recovering the cryptographic capabilities of the spacecraft. This research also proposes efficient fault tolerant architectures for the secure hash (SHA-2) and message authentication code (HMAC) algorithms. The proposed architectures are the first in the literature to detect and correct errors by using Hamming codes to protect the main registers. Furthermore, a quantitative analysis of the probability of failure of the proposed fault tolerance mechanisms is introduced. Based upon an extensive set of experimental results along with probability of failure analysis, it was possible to show that the proposed fault tolerant scheme based on information redundancy leads to a better implementation and provides better SEU resistance than the traditional Triple Modular Redundancy (TMR). The fault tolerant cryptographic primitives introduced in this research are of crucial importance for the implementation of on-board security in spacecrafts.