WO2016070855A1

WO2016070855A1 - Method for generating variable codes

Info

Publication number: WO2016070855A1
Application number: PCT/CR2014/000005
Authority: WO
Inventors: Carlos Luis CORDERO JIMÉNEZ
Original assignee: Monitoreo Tecnológico, S.A.
Priority date: 2014-11-06
Filing date: 2014-11-06
Publication date: 2016-05-12

Abstract

The invention relates to a method for generating codes from asynchronous or pseudo-asynchronous events, consisting in harvesting numbers from numerous sources, each source transmitting the number asynchronously by means of a manual or automatic command. The harvested collection of numbers is grouped to form asynchronous codes of different sizes. The source is a piece of software that takes the least significant bit from a computer clock and transmits it to the harvester via Internet protocols. This software can be any instant messaging system, such as MSN Messenger, ICQ, Yahoo! Messenger, AOL Instant Messenger or Google Talk. The source can transmit numbers of a larger size.

Description

"VARIABLE CODE GENERATION PROCESS"

FIELD OF THE INVENTION The present invention relates to the technological field as it is a process of generating codes from asynchronous or pseudo-asynchronous events by means of a manual or automatic command.

BACKGROUND OF THE INVENTION

1. Asynchrony and asynchronous events:

According to the Dictionary of the Royal Spanish Academy, asynchrony is defined as "Lack of temporal coincidence in the facts." In engineering it is considered "asynchronous", a phenomenon or event that is completely isolated and without continuation or correspondence in time with another event, whether or not similar.

The asynchronous event-based model makes the advantages of multithreaded applications available to the user, while avoiding many of the complex problems inherent in multithreaded design. Using a class that supports this model can allow you to:

• Perform tasks that require a lot of time (such as downloads and database operations) "in the background", without interrupting the application.

• Simultaneously execute several operations and receive the corresponding notification when each of them ends.

• Wait until resources are available without the application stopping (do not respond).

• Establish communication with asynchronous operations pending using the usual model of events and delegates.

1.1 Boolean functions:

When you have a combinational digital system, each output can be described by a Boolean function of its input variables. That is, the outputs of a combinational network depend only on their inputs.

i A Boolean variable is a symbol, which can be substituted by an element of the set B = {0,1}. The element that is replaced by the variable symbol is called the variable value. A Boolean constant is a value belonging to the set {0,1}.

In sequential systems, the values of the outputs, at a given time, do not depend exclusively on the values of the inputs at that time, but also depend on the previous state or internal state.

The simplest sequential system is the bistable, and of this, the type D (or bolt) is the most commonly used today.

Most sequential systems are governed by clock signals. These are called "synchronous" or "synchronous", unlike "asynchronous" or "asynchronous" which are those that are not controlled by clock signals.

2. Series of unrepeatable numbers in random order: These series are constructed under the following principle: if an "X" number comes out and one more number is desired, the probability that the same number will appear will be the same compared to the rest of existing data

2.1 Definition of random number:

A random number is generated by a mathematical formula or with the help of software that is responsible for generating it through an algorithm.

A set of variable numbers can be said to be obtained in such a way that every number obtained has the same probability of being chosen and the choice of one does not depend on the choice of the other. Whenever random numbers are generated, there is talk of a set of numbers in a range (1, n) in which the probability that one of those numbers is chosen will always be 1 / n.

Variable numbers are normally used to represent reality through mathematical models such as that of the daily climate (with certainty it is not known what the weather of a particular day will be), since the weather of one day does not depend on the weather of the previous day, so the probability of rain today does not influence the probability of rain tomorrow. 2.2. Utilization:

Although they are constantly used without being aware of this, they are especially suited for calculations of integrals, variance, fashion or errors.

In these cases, average points are used and plotted on a Cartesian plane, to later draw the line that allows identifying which are the closest points and in this way have a minimum range and margin of error on the calculation that is being made at a variable.

Variable number sets are very useful in probability and statistics, because their use allows empirically estimating various probabilities and hopes.

Variable numbers are always being used in web applications with which you interact every day, a way to simulate variable numbers can be using "javascript" with the random method.

Variable numbers generated with JavaScript can be used for countless things on a web page: display a random banner, a different header, a phrase, generate a security code, etc. 2.3. Types of random numbers:

Two classifications can be found:

- Uniformly distributed numbers - Non-uniform distributed numbers

In the case of non-uniforms, it is also necessary to take a uniform number and then invert the variable so that a separate process can be continued. The Cauchy distribution, the Gaussian distribution or the Bernulli distribution can be used for these procedures. For uniformly distributed, it should be noted that number generators have their own characteristics that identify them as such and are:

• These types of numbers should not fall into cycles

• The series of numbers generated must be reproducible

• Speed in acquiring the numbers · The generator must try to have a minimum storage

• All the numbers that are generated have to be uniformly distributed so that their probability of exit has to be the same. · All the elements generated must be independent of those generated previously

3. Methods of generating series of unrepeatable numbers in random and pseudo-random order:

The modern approach is to use a computer to successively generate pseudorandom numbers. These constitute a succession of values that, although produced in a deterministic way, have all the appearance of being uniform and independent random variables in (0,1).

In practice, no function produces true random data. Functions produce pseudorandom numbers.

Most of the generating methods start with an initial number (seed, X ₀ ), a certain procedure is applied to this number and thus the first "random number" is found. Using this number as input, the procedure is repeated to achieve a next "random number" X _n n> 1.

A value generator for a random variable is achieved in two steps: • First you have to define a random number generator evenly distributed between 0 and 1.

• This sequence is then transformed to generate the values of the desired random variable.

Generator Requirements:

• The generated series must be reproducible; Thus, in different experiments exactly the same sequence of random numbers can be used to reproduce the same conditions.

• It must allow the generation of several separate sequences of random numbers. A sequence is a segment of the numbers produced by the generator: a sequence ends where the next one begins.

Thus, a different sequence can be dedicated to each source of randomness of the model.

Basic Principle:

A variable number generator consists of a function that returns the values of a sequence of real numbers, (ui, U2,..., _N ), where each u is in the range [0, 1]

The first element of the sequence is called the initial seed of the series, and from it it must be possible to generate the rest of the sequence so that it is reproducible.

Properties of a generator:

Among the properties that a good generator must fulfill to perform simulations, we would highlight the following:

• The ui values must be independent and identically distributed.

• The generated sequence must be quite long, in order to allow long simulations and / or with many random variables. • It must be computed very efficiently, since each simulation could require the generation of a considerable number of variable numbers, of the order of millions. The condition of uniformity must also be met for all sub-sequences of size k of the generated sequence. That is, the u¡ must be distributed evenly in [0, 1] the pairs (u¡, u¡ + i) must be distributed evenly in the plane [0, 1] x [0, 1] the triplets (u¡, u ¡ ₊ I, u¡ + 2) must be distributed evenly in the cube [0, 1] ³ . Ensuring the k-uniformity of the sequences is crucial for simulation experiments.

Independence: A series is independent if it cannot be specified by an algorithm that requires fewer bits than the series itself.

3.1. Analog Computing:

The series are obtained with physical means.

Advantages: It is very fast and the series obtained are truly random. Disadvantages: Those that are obtained cannot be repeated.

3.2. Boards:

100,000 numbers can be obtained.

Advantages: The series obtained can be replayed.

Disadvantages: It is very delayed and demands a lot of storage space 3.3. Digital computing

It has a function and an initial value, of which the numbers are generated randomly. Advantages: It is fast, the series obtained can be repeated and does not take up much storage space.

Disadvantages: The numbers obtained depend on the previous ones

3.4. Manual

You can take as an example, throw a dice and make extraction of numbered balls inside an urn.

Advantages: The series obtained are truly random

Disadvantages: It is very slow, large storage is needed since each series that is obtained has to be saved and the series that is obtained cannot be repeated

3.5 Random number generators:

3.5.1 Types of Generators: · Physical: These are physical devices that use external sources: the decay of a radioactive material or electrical noise, to generate random numbers.

The most frequently used are based on electrical circuits equipped with a noise source (often a resistor or a semiconductor diode) that is amplified, sampled and compared with a reference signal to produce bit sequences. For example, if the noise is less than the reference a 0 is produced and if a 1 is greater.

Frequently, in the bit sequences thus obtained the 1's do not have the same probability as the 0's, so they must be post-processed to obtain random bit sequences.

For your application, these random bits are joined to form bytes, integers or real numbers, as required. • Arithmetic: They are deterministic algorithms that are executed by the computer.

Good arithmetic generators produce sequences of numbers that are indistinguishable from independent realizations of uniform variables. There are two large families of arithmetic generators: linear and nonlinear generators.

In simulation the most commonly used generators are linear.

0 3.5.2 Generator Models:

There are several models that allow generating random numbers, among which the most cited in the literature are:

5 · GCL (Linear Congruence Generators)

These generators are the most used and the best known. They are based on the recurrence relationship: N¡ + 1 = (aN¡ + c) mod m where "a" is the multiplier and "m" the module.

0 · There are "m" possible values of N¡, between 0 and m - 1.

• The sequence is periodic: when a number reappears a second time, the sequence is repeated. The period depends on the values of a, c and m, as well as the initial value; the maximum possible is m.

;5

• Bit Shift Generators

In these generators each new random integer Ni, is obtained by manipulating the bits of the previous number, Ni— 1. In C language, this can easily be done 0 using operators on bits, »,«, ^Λ , |, &.

• Fibonacci generators They are very fast generators that have a very long period. Fibonacci generators are based on a recurrence of the type N¡ = (N¡- _r or N¡_ _s ) mod m, where r <s are given integers and I denote some of the operations +, -, χ, ^Λ . This type of generator needs to start (with another generator) and keep a list of the last generated numbers.

• Medium squares This method begins with an initial number (seed). This number is squared. The middle digits of this new number (according to the desired digits) are chosen and placed after the decimal point. This number forms the first "random number".

• Exponential Distribution

To get Y with distribution

Then the equation remains:

Then, the appropriate change will be y = - J- In (1— x)

• Normal distribution

To obtain Y e N (0, 1), we have the following equation:

To solve this equation, the Box-Müller algorithm (by density of probabilities) is applied.

• Acceptance and rejection

It is a simple and general method, although sometimes it is not very efficient.

Let p (x) be defined in a finite interval, (a, b), and M an upper bound of p (x).

Two random values x e U (a, b) and p e U (0, M) are generated. So:

{x is accepted if p (x)> p

x is rejected if p (x) <p

The value x appears with a probability density p (x), although not all the values we obtain in the method are used.

This is the only method available when the distribution of probabilities is complicated.

It is the basis of the Metropolis algorithm, the most used in computational physics.

• Convolution Method:

The principle of this method is that, to obtain the values of X it will be enough to generate 2 values for Yi and Y ₂ , and return their sum. The sum of a set of random variables (convolution) is different from the sum of its distribution functions (composition).

There are several distributions that can be generated by convolution: • A variable with normal distribution can be obtained by adding a sufficient number of random variables with any distribution.

• A variable with chi-square distribution with k degrees of freedom can be obtained by adding the squares of k variables with distribution N (0, 1).

• A variable with Erlang-k distribution can be obtained by adding k variables with exponential distribution.

• The sum of two variables with distribution U (0, 1) forms a variable with triangular distribution.

All the mentioned models, allow the generation of numbers either uniformly distributed or non-uniform distributed.

4. Numerical sequence tests:

These consist of performing two types of tests, empirical and theoretical. If the generators pass these tests, we can ensure that we have a fairly competent random number generator.

Empirical tests (on the sequence sample):

- Uniformity test: the values need to be uniformly distributed in [0, 1]. An X ² test can be performed. Alternatively, we can estimate the moments of order k, X ^k = 1 / N∑ ¡(x¡) ^k and check that they approximate their corresponding theoretical values, 1 / (k + 1).

- Serial test: pairs of values (x1, x2) are generated, and it is checked whether they are distributed evenly in the square [0, 1] * [0, 1].

- Correlation test: the correlation between separate numbers k places in the sequence is determined, C (k) = 1 / N∑¡ x¡x¡ _{+ k} . Its value would have to approach 1/4.

Theoretical tests (over the entire sequence): The simplicity of linear congruence generators allow to demonstrate important properties:

- To determine values of a, c and m, sequences of maximum period m are obtained. For example, if m is a power of 2, it would be enough that c is odd and "a" equals

5 a multiple of 4 plus 1.

- Spectral test: if vectors with consecutive values are formed, u _n = (x _n , x _n + i. ■ ■ x _n + k), they form parallel hyperplanes in the k-dimensional space. The separation dk between the planes has to be the minimum possible.

0

Among the best known tests to verify these series are:

- The Frequency test: it is used to verify that the data is evenly distributed.

5

- The autocorrelation test: verifies the correlation between random numbers and compares them with the desirable zero correlation.

- The GAP test (of gaps or distance): is used to ensure that the recurrence ¡0 of each particular digit in a number flow occurs with a random interval. The KS test is then used to compare these intervals with the expected length of gaps.

- The Poker test: test groups of numbers together like a poker hand and! 5 compare each hand with the expected hand using the Chi-square test.

- The test run up and down: it is usually the main test used to verify the dependence. Detects if a statistically unacceptable pattern that increases or decreases exists between adjacent numbers in a flow of numbers.

- Series Test: Measures the correlation between adjacent elements in a sequence of random numbers.

5 Description of the Invention

The invention is mainly used in the field of statistics, systems simulation, key generation for information encryption.

5

Number banks are currently generated by two main methods, the first is through algorithms that take a "seed" which is a presumed number chosen from a selection and with that seed through specialized algorithms it generates a set of numbers which actually they are predictable) by what they are called pseudo variables.

The second method is through devices that take numbers generated by variable physical phenomena, these phenomena can be white noise, thermal noise, cosmic noise, atmospheric variables, etc. These values are unpredictable or

> depend on so many variables that in practice cannot be predicted. Although its degree of variability is extremely high, the generation speed is extremely low, so large tables of variable numbers required for key generation or cryptography cannot be generated.

) There are numerous companies that offer good quality number banks, but being previously known they are useless in cryptography although they work for simulation processes.

The behavior of complex living beings is determined by events

> asynchronous, that is, they happen at any time and that moment has no relationship or has a very distant relationship to another event. For example, the time it takes for the same animal to fall to its next prey depends on so many variables that the same animal cannot control, so that duration between prey is random. In the same way human behavior is random in

) its small details, for example how often we yawn or how often we press the search button on an internet search engine.

The present invention consists of a method to generate codes from asynchronous or pseudo-asynchronous events consists in collecting numbers from numerous 5 sources, each source sends the number asynchronously by means of a command manual or automatic The collection of numbers collected is grouped into groups to form asynchronous codes of different sizes. The source is software that takes the least significant bit of a computer clock and sends it via Internet protocols to the collector. This software can be any instant messaging system such as MSN Messenger, ICQ, Yahoo! Messenger, Aol Instant Messenger, Google talk. The source can send larger numbers.

The process generates codes or keys for communications, system simulation or encrypted transactions from asynchronous events. For this, the process starts from the digitization of an event, the specific time of occurrence of the event is recorded. That time is sent to a server and when it is on that server the least significant bit of the time sent consisting of a one or a zero is selected. The union of all events constitutes an unrepeatable row of codes whose flow must be continuous.

This process allows the collection and grouping of events from various remote sources. To generate asynchronous events, software is used that takes an event registered in an electronic device and sends it to a central computer which groups it together to form large chains of events.

The process sends the recorded event using a manual command or an automatic pseudo asynchronous, asynchronous or synchronous command. You could use one or more bits taken from the clock of an electronic device as a recorded event. The method could use as a source of the registered number the value of the clock generated by an integrated circuit of a computer equipment, cell phone, appliances; the watch

generated by a computer equipment software, a random number generated by a computer equipment software, a random number generated by a variable number generating device.

Additionally, the method uses a computer, a cell phone, a smart device, an electronic tablet, an electronic phone book, microprocessor devices, microcontrollers, machines as a sending device for the registered number. sequential logics, slot machines, ATMs, video game consoles.

The software that facilitates it can be instant messaging systems which requires human, animal or machine interaction, instant messaging systems MSN Messenger, ICQ, Yahoo! Messenger, Aol Instant Messenger, Google talk, search engines and internet, social networks and use as a means of communication the

Internet, computer networks, digital communication networks,

analog communication networks, telephone communication networks, conventional mail, Courier mail.

The events generated are grouped in the order of arrival, in random order, in groups of variable size, in arrangements sequentially or randomly. The number of events sent to the central computer is greater than one and an array of centralized or distributed servers is used as the central computer.

There is an event trigger which interacts with a computer, at the time of the event the least significant bit of the device clock in which the event occurred is recorded. Then the registered number is sent to the central computer.

Multiple people through their devices will be sending numbers to the central computer which generates a large row of numbers, these numbers are filtered if necessary where sequence 10 becomes 1, sequence 01 becomes 0, sequence 00 becomes discards and sequence 11 is discarded. The new row is the expected result and contains the variable numbers sequentially.

Claims

1. A process that generates codes or keys for communications, system simulation or encrypted transactions from asynchronous events consisting of:

From the digitization of an event the specific time of occurrence of the event is recorded,

that time is sent to a server,

when it is on that server the least significant bit of the time sent consisting of a one or a zero is selected,

the union of all the events constitutes a row of unrepeatable codes whose flow must be continuous.

2. A process of claim 1, which allows the collection and grouping of events from various remote sources.

3. A process of claim 1, wherein to generate asynchronous events a software is used that takes an event registered in an electronic device and sends it to a central computer which groups it to form large chains of variable events.

4. A process of claim 1, which sends the recorded event by a manual command or an asynchronous, asynchronous or synchronous pseudo automatic command.

5. A process of claim 1, which uses as recorded event one or more bits taken from the clock of an electronic device.

6. A method to capture asynchronous events using as a source of the registered number the value of the clock generated by an integrated circuit of a computer equipment, cell phone, appliances; the clock generated by a computer equipment software, a random number generated by a computer equipment software, a random number generated by a variable number generating device.

7. The method of claim 6, which uses a computer, a cell phone, a smart device, an electronic tablet, an electronic phone book, microprocessor devices, microcontrollers, sequential logic machines, slot machines, ATMs as a device for sending a registered number. , video game consoles.

8. A method of claim 6, which uses instant messaging systems as software which requires human, animal or machine interaction, instant messaging systems MSN Messenger, ICQ, Yahoo! Messenger, Aol Instant Messenger, Google talk, search engines and the Internet, social networks and uses the Internet, computer networks, digital communication networks, analog communication networks, telephone communication networks, conventional mail, Courier mail as a means of communication.

9. A method of claim 6, which generates events by grouping them in the order of arrival, in random order, in groups of variable size, in arrangements sequentially or randomly.

10. A method of claim 6, wherein the number of events sent to the central computer is greater than one and an array of centralized or distributed servers is used as the central computer.