Android Open Source

Crypto tech Neural networks are a great tool for specific cryptography. On the transmitting side, a neural network is used (for example, on a computer or smartphone) to compose such text in which only certain bytes (characters) will be significant. For example, this would be an ad for a bulletin board. Even if the system cannot accurately select the words so that all characters occupy the right places in the text, the neural network will be able to restore the correct original text. It's just that extra characters that occupy the right positions in the text will be discarded when checking the syntax. Moreover, the text itself can already be encrypted. This is one way Crypto tech  

  Crypto tech There are many interesting ideas for programming. There is not enough time to prototype applications and test these ideas. Not to mention writing commercial versions of applications. For example, synchronous random number generators with fuzzy logic. It's almost impossible to fake them. But for this I still need to understand the programming of "large objects" and write a prototype. Vernam KeyGen

Crypto tech The prototype Android application in Java generates a set of "random" numbers for a given phrase. While the phrase is still just specified in the variable, the GUI has a primitive look, a fixed length key is generated. But the prototype can generate the same set of numbers on two (or more) identical devices. The process of forming a set of numbers has great potential for customization, which will be shown in the following projects. The created "key" can be used for direct encryption (over encrypted, for example RSA) text or used for a software "random number" generator as source data. Basically, you can create a commercial application with Open Source Vernam KeyGen      

  Dimension of numbers   One letter is 16 bits. But it's interesting if you use a conversion table (before encryption), in which letters are encoded by numbers of different bit depths - 18, 21, 34, 52, 63 bits ... How much will this complicate decoding? And if the bit depth of the letters (corresponding numbers) is changed by chance, in accordance with the second encryption key? český pepř  

  Czech pepper   If you have access to the same digital stream for subscriber No. 1 and subscriber No. 2 (sender and recipient of messages), then you can select some sequence of bytes from the stream to use as a key or initial data to create a key.   If the digital stream is not random (for example, it is an Internet radio stream), but you select data from it randomly, then ... it can also be a random set of "low quality" numbers. For example, your algorithm will choose numbers that will be the same several times in a row, because. at these moments, the signal level (the volume of the Internet radio stream) changed little. How to improve such a set of numbers? The simplest options: discard repetitions, skip (discard) close values. This is easy to do in the same way for the sender and receiver of messages and generate the same keys. Additionally, known techniques may be used. For example, capture a fragment of an Internet radio stream and archive it as a ZIP. Then selec

  Solutions   An interesting topic for a startup is the publication of news, for example, about some events or local people (or ad texts, or resumes, or vacancies), compiled in such a way that certain characters are in certain positions of the text. And the set of these characters is the key or ciphertext. In fact, not only the decoding of the cipher becomes a problem, but the very fact of its discovery. This is something like a public encrypted text broadcast. And, if the full text of the message is made up of several publications in different publications (websites), then the detection and "hacking" of such messages and answers is a big problem. It remains, perhaps, to track the activity of publication sources ... Well, or ... Entropy Crypto Tech   Entropy    

  Small digression   It is quite possible that some of what you read in the news already contains encrypted information. And you continue to be surprised by the strange style of modern journalists. And this is a very interesting topic. Text generators with feedback, flows... I even want to try. I'll start getting ready...     Crypto Tech  

Crypto Tech Open Source Entropy generators that can generate the same encryption key without using a sync channel is a very interesting programming topic. Ideally, the sender and receiver toss 50 coins and count the number of coins dropped by one side. And they do it the same way. Then they can use the Vernam code and be calm about the safety of information. But coins cannot fall in the same way in different places. If there is no synchronization channel. And such a channel is a security hole. What to do? Use quantum technologies? While it is too heavy, cumbersome, noticeable. You can create a commercial firm that will make a server with a hardware random number stream generator available to everyone. And can put it in the Internet radio or TV channel or distribute it from a satellite. Some Broadcast. No one but you (the owner) will know if this is really a stream of random numbers. Or only every seventh (or twenty-seventh) number in the stream is truly random. Your applicatio

  Digital Noise Channels Digital channels transmitting a stream of random numbers generated from the thermal noise of semiconductor devices can have a complex algorithm of operation. Apart from the owner of such a public channel, no one knows when digital noise is random and when it is synthesized Accordingly, if your application can choose real random data from the stream, you can get absolute encryption strength. At the same time, other users may work with less strong ciphers.   . The quality of the noise stream can be significantly improved by third-party users with the help of special processing using "huge programming objects" that perform asymmetric transformations. The output of such systems will have a different digital stream, the same for the transmitting and receiving sides. Examples of Java applications using "huge programming objects" can be created even on modern smartphones. Internet chats, announcements, news, Internet radio, counters, temporary r

Password   Text-to-speech is a great way to store passwords on your system. If you use your own audio codec (which will not be updated at the most inopportune moment), then you can always create the same password using the same phrase. But you can change it once a day with the same source text. In this case, if a hacker can copy the system, the saved password (encryption key) will change in a day and that's all! The original text is not stored anywhere.   Audio entropy

  Android Java The "quality" of the sound file as a source of random numbers is not high. In a small area, integers change less than we would like. To improve the quality of "random" numbers, special processing and selection algorithms are needed. You can use several algorithms at the same time. Because Since the source of integers (sound file) in the same assembly of the Operating System on the same smartphones turns out to be the same, it is important that the processing algorithms of the sender and recipient are the same. It is clear that such an encryption key cannot be considered ideal for the Vernam code. The strength of the cipher will not be absolute, but the labor costs for cracking can be huge. This mechanism can be used to further protect already encrypted data. Countermeasures for such an algorithm can be to correct the speech synthesis code - adding "random" values to the synthesis parameters. But you can develop your own codec for encr

  Key Generate Vernam's code is old, easy to program, low resource, fast.   But it depends on the quality of the random number generator. For commercial cryptography, you can use the digital stream of Internet TV. Better - news channels. Or Internet radio with news too. The fewer repetitions, the higher the probability that the bytes selected from the stream will be a random set.     In the application, you set the search in the stream for some sequence of bytes. For example from 10 values.   You need to select such a NUMBER of bytes for waiting (marker) that it happens 1-2 times a day.     After the length of the sequence is chosen, several of its varieties are written to the application.     The application is launched by the sender and receiver at the same time and listen to the same Internet stream. With a high probability, they will find the given sequence (marker) in the stream at the same time. After that, according to some algorithm, the same number of bytes are