Specializing in Password Generating Software
iGBu has developed a number of advanced software concerning bilinear problems.
All are available for Windows, Mac and Linux platforms.
Detection of illicit fuels: EUREKA, EUREKA II, HEUREKA
Fuel adulteration is a widespread problem across the globe. It causes deterioration of fuel quality, degradation of engine performance, and significantly reduces air quality. Fuel adulteration results in customers paying more for illegally blended cheaper fuels. It also directly impacts the revenue and reputation of governments and oil firms. iGBu has developed a technique to detect illicit fuel mixtures that does not require chemical markers.
iGBu has developed the EUREKA program. Most modern techniques for detecting fuel adulteration use chemical markers, which are expensive and can
be removed via illegal fuel laundering. The EUREKA program is clearly superior to the current techniques of using markers for detecting illicit fuel
because it does not require the use of markers.
EUREKA II is a new version of software for detecting illicit fuels. It is under development with a new concept of algorithm for detecting the
HEUREKA is also under development using a new algorithm concept, which is different from those used for the EUREKA and EUREKA II
programs. This program can be applied to the fuel adulteration of both gasoline and diesel fuels.
Source apportionment of atmospheric aerosol or sediments of river or lakebed: SAFER, SMP, RESOLVE
The Chemical Mass Balance (CMB) model has been broadly used for aerosol source-apportionment studies. The CMB model may be the best model in estimating source contributions if accurate source compositions are available. However, reliable source compositions are virtually impossible to measure for some sources. In this case, a multivariate receptor model is generally applied to estimate source contributions from ambient data.
iGBu has developed three multivariate models, SAFER (source apportionment by factors with explicit restrictions), RESOLVE, and the SMP (Solver for Mixture Problem), and applied them to a number of particulates data sets. SAFER is the first multivariate receptor model that satisfies all five natural physical constraints (FNPCs) in estimating source compositions. RESOLVE is used to estimate the range of source compositions for unknown potential sources. It uses ambient data without any a priori information in estimating the range of unknown source compositions. SMP was recently developed and is the most advanced source apportionment model to date.
Good for when some a priori information is available for known sources
SAFER is a multivariate receptor model based on PCA analysis and uses a series of linear programming to estimate source compositions. The
SAFER model uses five Fundamental Natural Physical Constraints (FNPCs) and additional physical constraints (APCs) to estimate source
compositions. The SAFER model always guarantees physically meaningful source compositions. However, the SAFER model requires some a
priori information for each source as APCs, which is not always available for all sources. Therefore, accurate source apportionments could be
made by first applying the SAFER model to estimate source compositions of major sources such as fugitive soil dust, sea salt, biomass burning,
and secondary sources, and later followed by the CMB analysis. Details of the SAFER mode refer to the following references.
Published papers in relation to the SAFER and CMB model:
1. “Application of SAFER Model to the Los Angeles PM10 Data”, Atmospheric Environment, 34, 1747-1759.
2. “Extension of Self-Modeling Curve Resolution To Mixtures of More Than Three Components. Part III. Atmospheric Aerosol Data Simulation Studies,”
Chemometrics and Intelligent Laboratory Systems, 52, 145-154.
3. “Extension of Self-Modeling Curve Resolution To Mixtures of More Than Three Components. Part II. Finding the Complete Solution,” Chemometrics and
Intelligent Laboratory Systems, 49, 67-77.
4. "Extension of Self-Modeling Curve Resolution To Mixtures of More Than Three Components. Part 1. Finding the Basic Feasible Region," Chemometrics
and Intellegent Laboratory Systems, 8, 205.
5. “Diagnostics for Determining Influential Species in the Chemical Mass Balance Receptor Model”, Journal of Air & Waste Management Association, 49,
6. "The USEPA/DRI Chemical Mass Balance Receptor Model, Version 7," Environmental Software, 5, 38-49.
Good for when no a priori information is available for any source
In an effort to improve the SAFER model, the new multivariate model SMP has been developed, and applied to the source apportionment of
PM2.5, roadside particulate PAHs and PM10 data. The multivariate source apportionment model is bilinear and inherently ill posed.
Therefore, to restrict the feasible solution region into a smaller one, known physical constraints are imposed on the model. The SMP model
imposes all five FNPCs in the measurement space, while the SAFER model implements them in the eigenvector space. The five FNPCs are
minimum constraints that must be satisfied in the model. If any one of the five FNPCs is not implemented, the model results are questionable
Existing multivariate models have neglected the implementation of one of the important FNPCs (i.e., the sum of the model-estimated source
compositions must be less than or equal to one), which can lead to questionable results. Since this constraint is not implemented in the
existing models, the model result violates the mass balance equation of the receptor modeling, which leads to unreliable source compositions
and its corresponding pair of source contributions.
The SMP model clearly implements all five FNPCs using primal-dual interior point quadratic programming, subject to the inequality
constraints of all five known physical constraints. The SMP model does not require any a priori knowledge, although any additional available
information will indeed improve the results. The SMP is the most advanced state-of-the-art source apportionment model. The details of the
SMP model refer to the references below.
Published papers in relation to the SMP model:
1. “Transported vs. local contributions from secondary and biomass burning sources to PM2.5” Atmospheric Environment, 144, 24-36.
2. “A multivariate receptor modeling study of air-borne particulate PAHs: Regional contributions in a roadside environment”, Chemosphere, 144, 1270-1279.
3. “Source apportionment of PM10 mass and particulate carbon in the Kathmandu Valley, Nepal” Atmospheric Environment, 123, Part A, 190-199.
4. “Development of a New SMP Model Satisfying All Known Physical Constraints in Environmental Application”, Chemometrics and Intelligent Laboratory
Systems, 121, 57-65. (doi:10.1016/j.chemolab.2012.11.020)
Good for getting an idea of source compositions for unknown sources
RESOLVE estimates the range of source compositions for unknown potential sources from measured data alone. This is an invaluable tool
for gaining some concept of source compositions of potential sources when no information is available before detailed analysis of the data.
Password generator: EZPD
Information security has become a significant issue in today’s market. In general, important data is protected by a password, and internet accounts require password to login. Long and complex passwords can securely protect confidential information, but can be difficult to remember. In response to a high demand for secure password generation, iGBu has developed the unique password generator EZPD.
iGBu has developed a new next generation password generating software EZPD. It generates complicated long password whenever it is
needed to login or unlock password protected file. EZPD does not store the passwords; it generates unique password whenever it is needed.
Even there is no need to memorize or save it in a safe place. Furthermore, there is no need to type a long complex password, it types out for
you without typos. EZPD is a worry-free password generating software.
Encryption program: SENCRPT
Secure encryption program SENCRPT is under development and will be available soon. The SENCRPT program encodes confidential information in a text file using our unique complicated algorithm.
Facial recognition: FREC
Facial recognition software FREC is under development.
Customize or develop software for your bilinear problems:
iGBu has a wealth of expertise and experience concerning bilinear problems. We have also developed a number of advanced models and tools for providing solutions to these problems. With our experience and extensive resources, we can help you to create custom models for your specific bilinear problems. We can also develop software just for your specific bilinear application. If you need any assistance, please contact us at firstname.lastname@example.org.