Abstract This paper presents embedding-code that can be used to realize a mobile fuzzy logic controller FLC. The software code is developed using MS Visual C++. The code can be downloaded into any microcontroller board MCB that supports C++. A short review of the current MCBs is included. Some mobile-FLC-applications are cited. The software-code is an FLC-simulator. It can be used... This paper presents embedding-code that can be used to realize a mobile fuzzy logic controller FLC. The software code is developed using MS Visual C++. The code can be downloaded into any microcontroller board MCB that supports C++. A short review of the current MCBs is included. Some mobile-FLC-applications are cited. The software-code is an FLC-simulator. It can be used to simulate an FLC off-line or it can be downloaded into a mobile-microcontroller so it can operate in autonomously. The embedding-code has 4-modules, 3-input and 1-output. The three input modules represent system description in the form: of a rule-based knowledge base; a set of input/output variables with their corresponding membership functions; and a set of sensed input variable values. The output module contains the generated control signals. It also generates continuous step-by-step output simulation trace illustrating the detailed sequential steps executed by the FLC’s inference engine. While operating off-line, the three input modules are represented by three input data files and the output module by a generated output file. While operating in real-time, the knowledge-base and the i/o membership functions are merged into the FLC’s C++ code. The sensed input signals and the computed output control signals are mapped to the target microcontroller’s i/o-ports. In real-time mode the output simulation trace can be disabled. We have used this software to implement two RT-FLCs: One for precise estimation of the state of charge SoC of a battery using what is called the impedance interrogation method; and the other to implement the classical controller for balancing of an inverted pendulum. Read More Read Less

Abstract The binary natural and the Gray binary reflected codes have been widely applied in many communication systems. In particular, Gray binary codes have been found to be optimum in reducing the error compared to binary natural codes. the Gray codes outperform the binary codes. In this paper the exact gain offered by Gray codes over binary codes is computed for... The binary natural and the Gray binary reflected codes have been widely applied in many communication systems. In particular, Gray binary codes have been found to be optimum in reducing the error compared to binary natural codes. the Gray codes outperform the binary codes. In this paper the exact gain offered by Gray codes over binary codes is computed for the first time and found that the gain is dependent on the code-word length (constellation size), however at longer code-words the gain due to Gray labeling tends to remain at 3dB. Read More Read Less

Abstract TIn earlier days of humanity when there was no currency goods and services were exchanged through barter system1. Gold and other valuable metals were also exploited as medium of exchange. With the invent of paper currency (PC) trade transactions became very easy. In spite of its advantages paper currency has three major flaws: first, the holder of currency is always... TIn earlier days of humanity when there was no currency goods and services were exchanged through barter system1. Gold and other valuable metals were also exploited as medium of exchange. With the invent of paper currency (PC) trade transactions became very easy. In spite of its advantages paper currency has three major flaws: first, the holder of currency is always at risk due to lawlessness culture in many societies of the world. Second, counterfeit currency is a big issue for governments and currency issuing authorities. Third, printing and transferring PC causes heavy costs. Private organizations have introduced digital currencies like bitcoin2 but none of them is guaranteed or governed by any government. That is why they are used in a closed environment against particular commodities. In this paper we introduce implementable digital currency system at larger scale, i.e., at state level. Our proposed digital currency is issued and controlled by the state or central bank of the country and we name it official digital currency (ODC). The process of issuing ODC is almost same as that of conventional paper currency (CPC) but the controlling system is different. ODC coins or notes are digitally produced hence eliminate the need for any metal or paper. Here we propose the system for issuing, distributing and controlling ODC. The proposal also explains the country-wide process of day to day transactions in trade through ODC. Our proposed currency aims to be the real currency at macro and micro levels. The ODC is more secure, reliable, economical and easy to use. In this paper we introduce just the idea and compulsory modules of ODC system and not the implementable framework. We will present the implementable framework in a separate publication. Read More Read Less

Abstract Pakistan hosts a competitive and fluid telecommunication market and for a company to sustain, create customer value and increase economic efficiency, it needs to better understand its customers. The purpose of clustering or customer segmentation is to deliver actionable results for marketing, product development and business planning. In this paper, we focus on customer segmentation using clustering algorithms on real... Pakistan hosts a competitive and fluid telecommunication market and for a company to sustain, create customer value and increase economic efficiency, it needs to better understand its customers. The purpose of clustering or customer segmentation is to deliver actionable results for marketing, product development and business planning. In this paper, we focus on customer segmentation using clustering algorithms on real data of a telecommunication company in Pakistan. After choosing appropriate attributes for clustering, we used two-step clustering algorithm and k-Means algorithm for creating different customer segments. The results of both algorithms are compared and the insights obtained from each segment were analyzed before suggesting marketing strategies. Read More Read Less