IIIT Bangalore


IIIT Bangalore

Time filter
Source Type

Chaudhuri S.,IIIT Bangalore | Baig I.,IIIT Bangalore | Das D.,IIIT Bangalore
Computer Communications | Year: 2017

Classical wireless cellular networks configure semi-static values of handover decision parameters, which are usually been modified manually by the operator at a certain time interval post analyzing the network performance report. To reduce the operator's manual intervention, 3rd Generation Partnership Project (3GPP) proposed Self-Organizing Network (SON) concept for Handover (HO) decision parameters optimization for the LTE-Advanced (LTE-A) network. However, considering high-speed mobility requirements and ever-growing complexity from the contemporary LTE-A network, neither the classical manual approach nor the existing SON based approach from the prior-art will be able to meet the stringent HO performance goal set by the standard. In addition, unconventional handover algorithms and their decision parameters setting can affect user throughput, increases call drops and finally degrades the mobility performance of the overall LTE-A network. Therefore, in this paper, we propose a novel and efficient self-optimizing HO detection along with HO execution and decision parameter optimization algorithm which is named as Handover Detection Self-Organizing-HO Parameters (HD-SOHP) based on Reinforcement Learning (RL) concept. HD-SOHP improves the users’ mobility performance by achieving effective session handovers thereby reducing call drop, HO failures and ping-pong at the cell level. Compared with the different SON based handover techniques as stated in Table 2, Table 3 and Table 4, HD-SOHP gives two times performance improvement with respect to call drop, handover failure and ping-pong for the users’ moving with speed up to 120kmph. © 2017 Elsevier B.V.

Kumar P.,IIIT Bangalore | D'Souza M.,IIIT Bangalore
2017 9th International Conference on Communication Systems and Networks, COMSNETS 2017 | Year: 2017

Evolution of Internet of Things (IoT) demands interconnection of many autonomous and heterogeneous devices. Several such devices have very limited power. Every bit transmission consumes power and it is critical. The efficient power usage is a challenge. In this paper, we model an IoT device as a simple Hidden Markov Model (HMM) with a finite number of states and well determined emission probabilities. States in our HMM indicates the status of a device. We use the HMM to efficiently orchestrate the heartbeat duration of an IoT system. Our approach can identify the device anomaly with high accuracy and also save the end device power, by intelligently transmitting heartbeats based on HMM analysis. Our experimental result shows that, determination of device anomaly can be as high as 98%. © 2017 IEEE.

Aadhithan N.A.,IIIT Bangalore | Wali P.K.,PESIT South Campus | Das D.,IIIT Bangalore
2017 9th International Conference on Communication Systems and Networks, COMSNETS 2017 | Year: 2017

The exponentially increasing demand for higher data rates has pushed the mobile network operators and researchers to scout for ways to improve the system bandwidth, most notably by using the unlicensed band which already hosts long-standing technologies like Wi-Fi. Of late, some effort has gone into exploring the performance of Licensed-Assisted Access (LAA) which can boost the bandwidth of LTE by opportunistically anchoring a secondary carrier in the unlicensed band in addition to the primary licensed carrier. Duty-cycle based schemes are being looked at to fasttrack the commercial deployment of LAA, and falling along this line of thought, the current state of the art scheme in the literature proposes a proportionally fair duty-cycle based LAA scheme. In this work, we investigate the interesting question of whether a proportionally fair duty-cycle based LAA scheme affects the performance of Wi-Fi more than it should. © 2017 IEEE.

Bapna A.,IIIT Bangalore | Srinivasaraghavan G.,IIIT Bangalore
Proceedings - International Conference on Machine Learning and Cybernetics | Year: 2017

We propose a novel method for Bayesian learning of the parameters of a mixed belief network. Given the structure of a network, the parameters of conditional distribution of a node based on its type (discrete or continuous) and the types of its parents are learnt from the data. This node-wise updating scheme puts no restriction on the number and type of parents any node can have. We also extended the traditional algorithm for learning pure Gaussian networks to (i) deal with conditional Gaussian nodes, (ii) allow continuous nodes to be multivariate Gaussian and (iii) be able to converge to actual mean, covariance and weights of the network with which we generated the data. © 2016 IEEE.

JayPrakash L.T.,IIIT Bangalore
2013 IEEE 1st International Workshop on Communicating Business Process and Software Models: Quality, Understandability, and Maintainability, CPSM 2013 | Year: 2013

We propose a technique for analysis of impact of design changes using dynamic slicing of UML models. For a software architecture specified using UML, we first transform a given model into a graph representation which we have named Model Dependency Graph (MDG). MDG combines information available in the interaction model along with the relevant information available in class model into an integrated model. For a given slicing criterion, our slicing algorithm traverses the constructed MDG to identify the relevant model elements forming the dynamic slice. We use these slices to identify the model changes, and their impact. Our impact analysis methodology is based on using the difference between the dynamic slices computed before and after changes are made to the model. Our methodology's novelty lies in performing impact analysis based on a graph metamodel as against independently processing separate UML models. © 2013 IEEE.

Haloi M.,Indian Institute of Technology Guwahati | Jayagopi D.B.,IIIT Bangalore
IEEE Intelligent Vehicles Symposium, Proceedings | Year: 2015

In this work, we have developed a robust lane detection and departure warning technique. Our system is based on single camera sensor. For lane detection a modified Inverse Perspective Mapping using only a few extrinsic camera parameters and illuminant Invariant techniques is used. Lane markings are represented using a combination of 2nd and 4th order steerable filters, robust to shadowing. Effect of shadowing and extra sun light are removed using Lab color space, and illuminant invariant representation. Lanes are assumed to be cubic curves and fitted using robust RANSAC. This method can reliably detect lanes of the road and its boundary. This method has been experimented in Indian road conditions under different challenging situations and the result obtained were very good. For lane departure angle an optical flow based method were used. © 2015 IEEE.

Agrawal P.,IIIT Bangalore | Rao S.,IIIT Bangalore
e-Energy 2015 - Proceedings of the 2015 ACM 6th International Conference on Future Energy Systems | Year: 2015

It is known that energy-minimal scheduling is strictly harder than minimal-makespan scheduling, but it is not well understood. This paper is a study of the problems of scheduling to minimize the energy consumed by a system of dissimilar machines (with no restriction on the makespan). The different system types considered are for machines with identical speeds but different power ratings, and for machines with different speeds and power ratings but no idle power consumption. The results give insights for further studies classifying various types of energy-minimal scheduling problems, and for deriving algorithms for exact or approximate solutions for such problems. © 2015 ACM.

Das D.,IIIT Bangalore
Proceedings of the 2016 IEEE International Conference on Wireless Communications, Signal Processing and Networking, WiSPNET 2016 | Year: 2016

Apart from research being done on using SDN in wired networks, there have been attempts made to extend similar concepts in wireless domain as well, to improve network performance. There have been several architectural proposals put forward which includes concepts like reconfigurable data plane in base-stations e.g. OpenRadio to different Radio Access Technology (RAT). These ideas can be extended further to improve mobility of the UEs. It is known fact that intra-RAT mobility (e.g. LTE to LTE) is far more efficient than inter-RAT (e.g. LTE to UMTS) counterparts in terms of latency and complexity. Hence, it would be useful in a futuristic network if inter-RAT mobility procedures can be converted to intra-RAT ones by configuring some of base-stations dynamically to required RAT based on pattern of UE mobility. In this paper, we show how deployment of reconfigurable base-stations improve the overall mobility of UEs in a multi-RAT cellular network by increasing number of intra-RAT mobility procedures and in turn decreasing the time consuming inter-RAT mobility procedures. Simulation results reveal, as deployment of reconfigurable base-stations increases there is a maximum increase of 11% in probability of intra-RAT mobility procedures undergone by UEs in the network. Also, this increase in intra-RAT mobility saturates beyond a certain percentage of deployment of reconfigurable base-stations. © 2016 IEEE.

Ganesan R.,Infosys | Sarkar S.,Infosys | Narayan A.,IIIT Bangalore
Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012 | Year: 2012

Sharing of physical infrastructure using virtualization presents an opportunity to improve the overall resource utilization. It is extremely important for a Software as a Service (SaaS) provider to understand the characteristics of the business application workload in order to size and place the virtual machine (VM) containing the application. A typical business application has a multi-tier architecture and the application workload is often predictable. Using the knowledge of the application architecture and statistical analysis of the workload, one can obtain an appropriate capacity and a good placement strategy for the corresponding VM. In this paper we propose a tool iCirrus-WoP that determines VM capacity and VM collocation possibilities for a given set of application workloads. We perform an empirical analysis of the approach on a set of business application workloads obtained from geographically distributed data centers. The iCirrus-WoP tool determines the fixed reserved capacity and a shared capacity of a VM which it can share with another collocated VM. Based on the workload variation, the tool determines if the VM should be statically allocated or needs a dynamic placement. To determine the collocation possibility, iCirrus-WoP performs a peak utilization analysis of the workloads. The empirical analysis reveals the possibility of collocating applications running in different time-zones. The VM capacity that the tool recommends, show a possibility of improving the overall utilization of the infrastructure by more than 70% if they are appropriately collocated. © 2012 IEEE.

Rao M.,IIIT Bangalore
Advanced Materials Research | Year: 2014

A novel way of three dimensional (3D) chip stacking has been designed in a view to improve heat dissipation across the layers. Chip stacking using vertical interconnections forms microscale channels for coolant to circulate through the gaps. Solder-based self assembled (SBSA) 3D structures have been designed as posts on simulated through silicon vias (TSVs) to prove the processing concept. The processing of SBSA structures using a low temperature solder alloy and dip soldering method is described. Additional processing steps to fabricate interconnected 3D structures were demonstrated. Mechanical grinding of the 3D structures shows that soldered SBSA structures were void free and robust enough to be used as a connection post for chip stacking. SBSA structures provide a solder bump that serves as a connection path in the integration of dissimilar electronic technologies. Conventional copper posts, developed in a previous project, can be an effective approach to integrated circuit (IC) stacking. However, the SBSA post provides more variety in size and shape with a potential to serve as a reservoir for solder to aid in chip bonding. The solder bumps are heat resistant and uniform thicknesses were obtained across a large array of SBSA structures. The electrical durability of SBSA posts were determined by completing I-V measurements after thermal treatments. Fabricated SBSA posts were subjected to thermal cycling with temperatures ranging from room temperature to 300 °C. The interconnected SBSA posts are shown to be stable until 165 °C with little variation in measured resistance. © (2014) Trans Tech Publications, Switzerland.

Loading IIIT Bangalore collaborators
Loading IIIT Bangalore collaborators