Sastry Jammalamadaka | Kl University (original) (raw)

Uploads

Papers by Sastry Jammalamadaka

In this paper the design of a transmitter for a frequency selective Raleigh fading Wireless Commu... more In this paper the design of a transmitter for a frequency selective Raleigh fading Wireless Communication channel is discussed. The proposed model is using Differential Phase Shift Keying (DPSK) modulation in which Synchronization and Doppler correction are taken care by including the preamble and a special sub carrier in the transmitted signal. The modulated data has been encrypted. The signals are made to pass through the Rayleigh Faded channel after getting dispersed. Three paths are assumed here in this paper. Doppler correction is taken to overcome the phasor noise. Additive White Gaussian Noise effect is considered while designing the transmitter model.

Web users surf the WEB sites quite frequently for accessing frequently required information. Ther... more Web users surf the WEB sites quite frequently for accessing frequently required information. There is no specific information stored in the WEB logs with which a user is uniquely identified. Every user navigates the WEB site leading to certain patterns of surfing. The web users being novice can only be recognised through patterns of surfing the web. Many algorithms have been presented in the literature that considers clustering the users based on some criteria generally leading to finding favourite links. The algorithms do not use any navigational patterns and the favourite links provided in the side bars do not provide any useful information. User behaviour cannot be predicated using just the favourite links. Most of the algorithms take more time to form clusters and present them to the user the favourite links in a side bar. This paper presents a method that quickly recognizes frequent navigational patterns of the novice user derived from a WEB log, and clicks rendered by the user for navigating through URL links. The user's interestingness is captured and used to find the interesting patterns. The customised URLs are presented as toolbar icons using which the user can directly surf the interesting areas. The method has been built into large web site and it has been proved that the surfing has been made efficient when compared with other methods presented in the literature.

Power management within intelligent Tags is one of the most important issues that must be given w... more Power management within intelligent Tags is one of the most important issues that must be given with utmost importance as the Tags are driven by battery power and the longevity of the battery must be increased so as to reduce the requirement of reducing frequency of charging or replacing the battery. Intelligent tags have many built-in functions that are required for supporting many of the intelligent issues such as identification, locating the tags etc. Each built-in intelligence requires certain amount of power. An event taking place either internally or externally involves usage of a set of intelligent modules which all adds to the requirement of certain amount of power. Power management has a direct impact on the latency with which the intelligent functions can be supported. This paper presents some of the strategies for power saving when a set of relevant functions are to be activated keeping in view of power and latency reduction.

Search engines are frequently used for obtaining some information hosted on different WEB sites. ... more Search engines are frequently used for obtaining some information hosted on different WEB sites. The correctness of the information searched is very much dependent on the keying in of the most appropriate WEB snippets and the order in which the WEB snippets are presented. Search engines literally visits the metadata of all the WEB sites before the content that matches the keyed in WEB snippets is presented to the end user through a set of URLS.

Intelligent Tags communicates with mobile devices using one of the wireless communication protoco... more Intelligent Tags communicates with mobile devices using one of the wireless communication protocols for transmitting the environment related to Mobile devices which may be situated at remote locations. The data transmitted to the mobile devices is quite sensitive and therefore must be secured. Building security mechanisms with the Tags will affect the design of the tags especially effecting the response time and throughput. More intelligence has to be built into Tags if attacking is to be sensed and then enforcing the counter attack mechanism in which the response time and throughput may temporarily get effected because security enforcement is under taken only when attacking takes place. However there should be in process implementation of security enforcement without effecting the response time and throughput. The paper presents a method of enforcing in-cloud security by using a PC in the neighborhood of a PC.

Finding the location of a TAG is very important especially when objects to which the tags are att... more Finding the location of a TAG is very important especially when objects to which the tags are attached are to be tracked. Tracking of the objects is a complicated process especially when there is fall down of the objects in the unknown places during airline crashers, atmospheric disasters etc. The objects must be tracked accurately, just finding the distance of an object from a master mobile device is not sufficient. It is also necessary to find the location of the object in relation to directions (North, North-East, West, South-West etc.,) in addition to the distance of the object from the master (hand held mobile device) accurately. In this paper a novel method has been presented using which the location of an object can be accurately determined.

WEB sites that make available the URLS which are frequently and usually accessed by the users for... more WEB sites that make available the URLS which are frequently and usually accessed by the users for immediate access without the need for frequent navigation backward and forward is most importantly required. The usage of such type of web sites will be phenomenally very high. WEB logs provide a good source to determine the pattern of navigation that different users adapt for surfing. Many techniques exist in the literature that uses the concept of clustering through usage of one of the available clustering algorithms. All of them suffer from huge amount of time that it requires to generate the clusters and also none have recommended a method of customizing the user interface that render the frequently required URL links right in the HOME page of large WEB sites. This paper presents a method that helps quickly recognizing the frequently navigated URLS with the help of the first elementary level of click rendered by the user. The method is simple, fast and stands the proof of concept. The method has been applied on to large web site and it has been proved that it is efficient when compared other methods presented in the literature.

An accurate assessment of the performance of a newly developed radio system can be done through r... more An accurate assessment of the performance of a newly developed radio system can be done through repeated tests of the system over an actual channel. When comparison is to be made between two or more systems over a real channel, they must all be tested simultaneously. The channel characteristics and transmission conditions vary uncontrollably. Hence tests can't be repeated at other times. Moreover it is not possible to test a system repeatedly for the same channel conditions. The computer simulation of channel model using C++ on Linux shows good degree for accuracy between input and output. The signal to noise ratio for different channel conditions can be calculated easily for rating the channel.

When a complex-valued signal is transmitted through a frequency selective fading channel with com... more When a complex-valued signal is transmitted through a frequency selective fading channel with complex low-pass equivalent, the time variant channel impulse response will have two components viz. a direct component and a dispersive Rayleigh faded component. The received signal, which is a random process, will consist of vector sum of different waves received after reflections by Ionosphere layers from a number of paths. When the impulse response of channel is modeled as zero means complex valued Gaussian process, the envelope of it at any instant "t" is Rayleigh distributed.

Networking heterogeneous embedded systems is a challenge. Every distributed embedded systems requ... more Networking heterogeneous embedded systems is a challenge. Every distributed embedded systems requires that the network is designed specifically considering the heterogeneity that exits among different Microcontroller based systems that are used in developing a distributed embedded system. Communication architecture, which considers the addressing of the individual systems, arbitration, synchronisation, error detection and control etc. needs to be designed considering a specific application. The issue of configuring the slaves has to be addressed. It is also important that the messages, flow of the messages across the individual ES systems must be designed. Every distributed embedded system is different and needs to be dealt with separately. This paper presents an approach that addresses various issues related to networking distributed embedded systems through use of universal serial bus communication protocol (USB). The approach has been applied to design a distributed embedded that monitors and controls temperatures within a Nuclear reactor system.

Distributed embedded systems are being used for many purposes especially in the field of automobi... more Distributed embedded systems are being used for many purposes especially in the field of automobile automation, safety critical systems and the process systems which need high performance and response. Many communications systems are being used which include I 2 C, CAN, and USB etc., for interconnecting distributed embedded systems. These communication systems will function effectively if all the distributed embedded systems are homogeneous. However these systems need many of the conversions when the networking of heterogeneous embedded systems is needed. Every distributed embedded system requires the design of networking, communication system architecture, design of flow of messages through proper addressing and arbitration, synchronization, error detection and control etc. These issues become complicated due to the use of heterogeneous embedded systems. This paper presents networking of a distributed embedded system through use of CAN protocol which is to be built around heterogeneous microcontroller based systems, a novel system for implementation of arbitration, the architecture for message flow and the design of data flow across the distributed embedded systems. The inventions presented in the paper have been applied to a pilot distributed system that monitors and controls the temperatures within a Nuclear reactor system.

Realizing a distributed heterogeneous embedded system using I 2 C requires investigations and fin... more Realizing a distributed heterogeneous embedded system using I 2 C requires investigations and findings related to designing of networking, architecture, message design and flow for prioritization, and datagram design. The existing methods for effecting communication among the heterogeneous embedded systems interconnected through I 2 C communication system have not addressed the heterogeneity issues. To suffice, new methods for designing Heterogeneous Embedded networks, communication systems, message flow systems and the message design have presented. The methods have been applied to an existing distributed embedded system and the results obtained proved to be exact as required. Every distributed embedded system is different and the kind of Hardware and protocol conversions required is dependent on the type of distributed embedded system. It has been shown in the paper the kind of conversions that must be undertaken considering I 2 C as the communication method and a set of selected heterogeneous embedded systems that are used for developing an application to monitor and control the temperatures within a Nuclear reactor system. The way the communication must be effected is dependent on the number of masters and the slaves that must be supported on the network. A communication system architecture that suits the pilot project has been presented. A design flow method which uses priority queues has been presented to effect the communication according to the flow required. The design of datagrams required for effecting the communication has also been presented in the paper.

Pair-wise testing is a combinatorial test criterion technique based on specification, which requi... more Pair-wise testing is a combinatorial test criterion technique based on specification, which requires each pair of input variables and every combination of their valid values should be covered by at least one test case. Pairwise test case generation is the process of producing a subset of all possible test cases that can test the system to a reasonable acceptance. Pairwise testing can be adapted when exhaustive testing is not possible or prohibitively expensive. Some of pairwise test techniques like AETG (Automatic Efficient Test Generator), IPO (In-Parameter Order), SBC (Simulated Bee Colony) etc. generate test cases by pairing the input variables. The methods generates the test cases which are less in number compared to total test cases that can be used to test the system exhaustively. These methods also take lesser time to generate the test cases. This paper proposes an artificial neural network (ANN) to generate optimized test cases based on the considerations of the pairs of input variables (Test pair generation). The method presented in this paper demonstrates that the test cases generated through associative mapping of test pairs to the neural network provides optimized solution when compared to other methods especially when test cases are generated to test a WEB applications. A tool called ANN-PTCG (Artificial Neural Network-Pair wise Test Case Generator) has been developed based on the method proposed in this paper. Test cases are generated for testing a WEB based application using ANN-PTCG and other similar pairwise testing methods which include AETG (Automatic Efficient 11820 Lakshmi Prasad Mudarakola et al Test Generator), IPO (In-Parameter Order), and SBC (Simulated Bee Colony). The number of test cases that are generated, the extent of coded covered and the amount of time taken to generate the test cases has been used to compare the ANN-PTCG with other related test case generation methods. It is seen that ANN-PTCG method is quite good compared to other methods due to generation of minimal test cases in less time and the test cases covering entire source code.

Wireless communication channel is being used as primary means of communication nowadays. Data is ... more Wireless communication channel is being used as primary means of communication nowadays. Data is DPSK modulated and hamming encoded before being given as input to the Rayleigh Faded Wireless Communication channel. This paper mainly focuses on the data re-generation techniques from phase modulated data at the source side. The generated data is made to pass through the channel by undergoing various transmission impairments. To accurately regenerate the data at the receiver side equalizers are recommended for usage. Decision feedback equalizer is one such equalizer that reproduces the data accurately. It takes care of all types of the noise components and suppresses the AWGN also. It counters the Doppler Effect and ISI most accurately.

Decision tree analysis involves forecasting future outcomes and assigning probabilities to those ... more Decision tree analysis involves forecasting future outcomes and assigning probabilities to those events. One of the most basic fundamental applications of decision tree analysis is for the purpose of option pricing. The binomial tree would factor in multiple paths that the underlying asset's price can take as time progresses. The price of the option is calculated using the discrete probabilities and their associated pay-offs at maturity date of the option. In this work we came up with an approach to build a binomial decision tree that can be used to price European, American and Bermudian options and a methodology to train the decision tree using a clustering based learning algorithm that minimizes the mean square error (MSE) between the observed and predicted option prices. The training methodology involves clustering the options based on moneyness and fit a linear equation for each cluster to calculate the confidence that needs to be used in building the binomial decision tree for a particular strike price within the cluster. It is observed that the MSE for option price using the proposed model is less when compared to the Black-Scholes model for the proposed learning algorithm.

RS485 based networking of distributed embedded systems is most effective as every microcontroller... more RS485 based networking of distributed embedded systems is most effective as every microcontroller based embedded system has built-in serial communication interface. Establishing a network connecting heterogeneous microcontroller based systems become quite simple when RS485 based communication is used. However RS485 suffers from too much of overhead due to the need for transmitting the addresses along with Data. HUB based RS485 based communication eliminates transmission overheads. In this paper HUB based RS485 based networking of heterogeneous embedded systems that suits to a pilot project that monitors and controls the temperature within a Nuclear reactor system has been presented. Many other related inventions that include message flow architecture, method of addressing the devices on the network, and data packet design that are related to a pilot project have also been presented.

Options traded over-the-counter are associated with credit risk and are called vulnerable options... more Options traded over-the-counter are associated with credit risk and are called vulnerable options. Collateral can be taken to mitigate the credit risk. However the amount of collateral should be such that it shouldn't take into account the almost not plausible states of the underlying which when considered increases the required collateral amount. This paper proposes a methodology to calculate the optimum collateral amount that is required from the seller of a vulnerable option. The calculated collateral makes the vulnerable option as risky as the exchange traded option. The algorithm to calculate the optimum collateral uses the novel binomial decision tree built without any assumption on the underlying distribution. The study reveals that the price of a vulnerable option converges to an exchange traded option as the collateral amount reaches a certain optimum value. The proposed methodology will be of interest to the option seller so that the excess collateral above the optimum collateral can be used for some other purposes.