About calibration GSS Psyleron

Check the connection between human consciousness and physical world
Дмитрий__
Разработчик
Posts:299
Joined:Wed Dec 23, 2009 13:49
Location:Институт высшей нервной деятельности и нейрофизиологии РАН
Has thanked: 2 times
Been thanked: 27 times
Contact:
About calibration GSS Psyleron

Post by Дмитрий__ » Thu May 12, 2011 14:34

Provided by the manufacturer for the procedure to verify the output quality of the "Pellerano". This description is intended to weed out questions about the influence of external artifacts on the results of the experiments with extrasensory effect.
<P>the Predecessor of REG-1 is the chip MicroREG successfully used in the Princeton lab of Jan. Improvements affected only the quality of the output random information and a new USB interface.</FONT></FONT></P>
<P>the Basic circuit elements of the device are:</FONT></FONT></P>
<P>1. 2 sources of noise arising from quantum tunneling of electrons in field-effect transistor.</FONT></FONT></P>
<P>2. Transistor signal amplifier</FONT></FONT></P>
<P>3. The comparator converts the signal into digital form</FONT></FONT></P>
<P>4. Mixing 2 signals to form a single data stream</FONT></FONT></P>
<P>5. Data output via USB port</FONT></FONT></P>
<P>"In all cases, the electronic process begins with a recognizable frequency distribution of white noise. For example, a test of the GSS, which had received the bulk of the data produces a flat spectrum with differences of no more than +/- 1 dB in the range from 50 Hz to 20 kHz. Next is the filtration process below 1000 Hz processing sampling rate. This is followed by amplifying and clipping the signal, which leads to wave trains of approximately rectangular shape with unpredictable timescales. Next is a gated 1-kHz pulses, producing an output sequence of random bits at equal intervals, convenient for a quick score. To eliminate possible environmental effects and aging of the electronic components to the stream of digital data is applied bit masks exclusive OR (XOR mask) with regular alternation of the structures of 1/0. This procedure also eliminates autocorrelation with the short period of the type of bits-bits and bytes-the bytes.</FONT></P>
<P>the output for the experiments are presented and recorded in the form of statistical "series", each of which represents a sequence of N "trials" (usually N = 200). Thus, the final test signal GSS is a sequence of conditioned bits, and in the later devices, of bytes output to the serial port of a computer presented as a sequence of "episodes", usually generated with SKO&shy;speed about 1 series per second. Calibration of the outputs of all devices fully meets the expectations of mathematical statistics by calculating the mean values, standard deviations, skusenosti and time for cumulative distributions of data, and time series of independent events. (1)</FONT></P>
<P>1) the “Mind/Machine Interaction Consortuim: PortREG Replication Experiments”, Robert G. Jahn, et al. * Research Report No. April 1, 2000</FONT></P>
<P>a Key objective in the creation of REG-1 is receiving data bits in the result of a process which are truly random in accordance with modern physical theory, but minimally influenced by known physical fields. For this we used electromagnetic protection, samplemovie signal simultaneously from multiple analog noise sources and subsequent processing to minimize or completely eliminate the influence of physical processes on the statistics of the first order, the most convenient in experiments with the device. Under normal conditions, psyleron REG-1 are all known the way&shy;statistical tests (for randomness).</FONT></P>
<P>furthermore, everyone ever made a version of the instrument was pre-tested by experienced operators and volunteers, for the subjective perception of their "sensitivity" of the device to psychic attack.</FONT></P>
<P><B>Security: </B>professional edition REG-1 contains at least 2 components of the electromagnetic protection; external aluminum housing surrounding the im&shy;direct electronic charge, and vnutrenni being permalloy mu-metal protection, isolating sensitive analog circuit elements. Internal protection inhibits the effect of both electric and magnetic fields outside and inside of the device.</FONT></P>
<P><B>Treatment</B><B>: </B>as additional procedures for ensuring the randomness of the output signal and preventing fluctuations due to changes in environmental conditions, this signal is logically multiplied by the bitwise (XORed) with polymorphic pseudorandom mask. This gives the effect of statistical Orav&shy;naveshivanija the data flow of the GSS, even in the case of physical deviations in the source&shy;Nike noise. </FONT></P>
<P><B>Sensitivity</B><B>: </B>thus treated output signal GSS is not sensitive to any known physical processes. In the case of the theoretically possible extreme conditions when affect the device in any way still, his behavior is likely to recall defective pseudo-random noise source, which will not pass calibration tests with an unbiased (however) the average. To control this highly improbable possibility of a new version of the device has a special led indication.</FONT></P>
<P><B>Statistical tests: </B>Before sending the consumer a Psyleron REG-1 is simply a special calibration tests. Each device is exposed as mi&shy;a minimum of 5 lots of 1 million 200-bit trials (equivalent to 1 million bits), carried out separately. On the basis of these results, each device is tested, fails, or is subjected to further testing.<B> the Device does not pass the test the first time (or sent to the addi&shy;tional testing) can no longer be labelled as </B><B>“Professional</B><B> </B><B>Edition” REG-1</B><B>.</B></FONT></P>
<P>the First battery of tests used to REG-1, comes from the PEAR lab. The first test operates with a 200-bit series and checks out their mean value, standard deviation, moment skewness and the bit population. Other tests include autocorrelation up to 25 lags, the test of arcsine, test length and sample tests on the mean and standard deviation of blocks of 100 and 1000 series. Each test is evaluated according to the minimum confidence level against the null hypothesis that the calibration data fit the theoretical distribution for a fully balanced process. For the success of REG-1 "to fail" not more than 5 (out of about 90) of these tests on a confidence level of 0.05, and none at the level of 0.001, and never to fail the test on the same category 5 more than twice. Devices not satisfy these requirements can no longer be "professional editions".</FONT></P>
<P>Each device is sent after the March 2009, also exposed a series of similar tests at the bit level: simple tests for the mean, autocorrelation, probability of bit transmission and distribution bytes. Not yet known a single case to failed device passed tests PEAR. In addition to the above devices REG-1 of each produced batch are sample tested third-party tests such as the test Dihard. To date, all devices "Psyleron REG-1 professional edition" these tests have been based on internationally accepted criteria. Also, this battery of tests is required for all Psyleron REG-1 Professional edition as of March 2009.</FONT></P>
<P><B>Methodology</B><B>: </B>In addition to the above users recommended a variety of methodological precautions to ensure the reliability of their experiments. For example, the reliability of any effects of the strengthened pre - and post-experimental calibration, and the manufacturer recommends users the possibility to carry out such experiments. </FONT></P>
<P><B>Opinion</B><B>: </B>subject to implementation of appropriate methodological precautions compromising the impact of the physical artifacts of the environment on the results of the experiments with REG-1.almost <B>cannot</B>.</FONT></P>
<P><BR></P>
<P><BR></P><

Post Reply