With the advent of feature rich smartphone platforms such as Android and iOS, people can now enjoy a wide variety of applications
on-the-go. Among these applications, games are one of the most desired types. However, as bigger screens, faster CPUs and interfaces supporting higher bandwidth (WiFi, 3G, LTE) consume more power, battery lifetime becomes a bottleneck on such devices. In this paper, we present novel techniques that combine 3D spatial subdivisioning, Potentially Visible Set (PVS) and Visual Perception
based Localisation (VPL) methods to estimate the non-critical game states to save wireless interface energy with minimum processing penalty. Our techniques and algorithms are realised in a commercial game and can save up to 57% of wireless interface energy while retaining game play quality.
Chaotic Block Image Scheme using Large Key Space and Message Digest AlgorithmCSCJournals
In this paper, chaotic block image scheme using large key space and message digest algorithm. Cat map intended for confusion and 2D-Sine Tent Composite map (2D-STCM) key generator intended for diffusion. Confusion is implemented by 2D Cat map with arbitrary block size. In the first tendency, 2D cat map use for local shuffling of indexes inside blocks, while in the second tendency, 2D cat map used for global shuffling of whole image indexes. The designed algorithm executes two confusions and one diffusion in each iteration. To increase the security level, the message digestion algorithm is used as a fingerprint for the plain image that creates the initial value of the key. After that 2D-STCM generates a large key stream. Diffusion implementation takes place by XOR operation; between a key stream and confused image. Experimental results, show that security level increases due to integration of confusion and diffusion. On the other side large key space and the high sensitivity of secret keys have been given a guarantee for the performance of the security. Performance measures reach to the top value among those in the similar researches. To verify the obtained results, authors implemented inverse chaos. All the tests are processed by MATLAB 2015a.
Research Inventy : International Journal of Engineering and Scienceinventy
Research Inventy : International Journal of Engineering and Science is published by the group of young academic and industrial researchers with 12 Issues per year. It is an online as well as print version open access journal that provides rapid publication (monthly) of articles in all areas of the subject such as: civil, mechanical, chemical, electronic and computer engineering as well as production and information technology. The Journal welcomes the submission of manuscripts that meet the general criteria of significance and scientific excellence. Papers will be published by rapid process within 20 days after acceptance and peer review process takes only 7 days. All articles published in Research Inventy will be peer-reviewed.
An Approach Towards Lossless Compression Through Artificial Neural Network Te...IJERA Editor
An image consists of significant info along with demands much more space within the memory. The particular significant info brings about much more indication moment from transmitter to device. Any time intake is usually lowered by utilizing info compression techniques. In this particular method, it's possible to eliminate the repetitive info within an image. The particular condensed image demands a lesser amount of storage along with a lesser amount of time for you to monitor by means of data from transmitter to device. Unnatural neural community along with give food to ahead back again propagation method can be utilized for image compression. In this particular cardstock, this Bipolar Code Method is offered along with executed for image compression along with received the higher results as compared to Principal Part Analysis (PCA) method. However, this LM protocol can be offered along with executed which will acts as being a powerful way of image compression. It is seen how the Bipolar Code along with LM protocol fits the very best for image compression along with control applications.
A NOVEL SCHEME FOR DEVIATION DETECTION IN ASYNCHRONOUS DISTRIBUTED PRICINGIJNSA Journal
Modelling resource allocation problems in the form of non-cooperative pricing games takes into account the difference between how much a given performance metric is valued and how much is paid for it. For the convergence of the sum of all users’ payoff to a global maximum, the determination of the utility function is essential. Although supermodularity conditions have been previously defined and determined to obtain suitable utility functions, different utilities have significantly varying performance characteristics under similar network parameters. In an ad-hoc framework, absence of a central authority leads to uncontrollability of unfairness. Users could misbehave by broadcasting high price coefficients to force other users to transmit at a lower power. This paper proposes an adaptation of the Asynchronous Distributed Pricing Algorithm with a Deviation Detection Block that re-aligns the deviated system back into the algorithm.
Chaotic Block Image Scheme using Large Key Space and Message Digest AlgorithmCSCJournals
In this paper, chaotic block image scheme using large key space and message digest algorithm. Cat map intended for confusion and 2D-Sine Tent Composite map (2D-STCM) key generator intended for diffusion. Confusion is implemented by 2D Cat map with arbitrary block size. In the first tendency, 2D cat map use for local shuffling of indexes inside blocks, while in the second tendency, 2D cat map used for global shuffling of whole image indexes. The designed algorithm executes two confusions and one diffusion in each iteration. To increase the security level, the message digestion algorithm is used as a fingerprint for the plain image that creates the initial value of the key. After that 2D-STCM generates a large key stream. Diffusion implementation takes place by XOR operation; between a key stream and confused image. Experimental results, show that security level increases due to integration of confusion and diffusion. On the other side large key space and the high sensitivity of secret keys have been given a guarantee for the performance of the security. Performance measures reach to the top value among those in the similar researches. To verify the obtained results, authors implemented inverse chaos. All the tests are processed by MATLAB 2015a.
Research Inventy : International Journal of Engineering and Scienceinventy
Research Inventy : International Journal of Engineering and Science is published by the group of young academic and industrial researchers with 12 Issues per year. It is an online as well as print version open access journal that provides rapid publication (monthly) of articles in all areas of the subject such as: civil, mechanical, chemical, electronic and computer engineering as well as production and information technology. The Journal welcomes the submission of manuscripts that meet the general criteria of significance and scientific excellence. Papers will be published by rapid process within 20 days after acceptance and peer review process takes only 7 days. All articles published in Research Inventy will be peer-reviewed.
An Approach Towards Lossless Compression Through Artificial Neural Network Te...IJERA Editor
An image consists of significant info along with demands much more space within the memory. The particular significant info brings about much more indication moment from transmitter to device. Any time intake is usually lowered by utilizing info compression techniques. In this particular method, it's possible to eliminate the repetitive info within an image. The particular condensed image demands a lesser amount of storage along with a lesser amount of time for you to monitor by means of data from transmitter to device. Unnatural neural community along with give food to ahead back again propagation method can be utilized for image compression. In this particular cardstock, this Bipolar Code Method is offered along with executed for image compression along with received the higher results as compared to Principal Part Analysis (PCA) method. However, this LM protocol can be offered along with executed which will acts as being a powerful way of image compression. It is seen how the Bipolar Code along with LM protocol fits the very best for image compression along with control applications.
A NOVEL SCHEME FOR DEVIATION DETECTION IN ASYNCHRONOUS DISTRIBUTED PRICINGIJNSA Journal
Modelling resource allocation problems in the form of non-cooperative pricing games takes into account the difference between how much a given performance metric is valued and how much is paid for it. For the convergence of the sum of all users’ payoff to a global maximum, the determination of the utility function is essential. Although supermodularity conditions have been previously defined and determined to obtain suitable utility functions, different utilities have significantly varying performance characteristics under similar network parameters. In an ad-hoc framework, absence of a central authority leads to uncontrollability of unfairness. Users could misbehave by broadcasting high price coefficients to force other users to transmit at a lower power. This paper proposes an adaptation of the Asynchronous Distributed Pricing Algorithm with a Deviation Detection Block that re-aligns the deviated system back into the algorithm.
AN ENHANCED SEPARABLE REVERSIBLE DATA HIDING IN ENCRYPTED IMAGES USING SIDE M...Editor IJMTER
This paper proposes a scheme for Enhanced Separable Reversible Data Hiding in
Encrypted images Using Side Match. In the first step the original image is encrypted using an
encryption key. Then additional data is embedded into the image by modifying a small portion of the
encrypted image using a data hiding key. With an encrypted image containing additional data, if a
receiver has the data hiding key, he can extract the additional data. If the receiver has the encryption
key, he can decrypt the image, but cannot extract the additional data. If the receiver has both the data
hiding key and encryption key, he can extract the additional data and recover the original content by
exploiting the spatial correlation in natural images. The accuracy of data extraction is improved by
using a better scheme for measuring the smoothness of the received image, and uses the Side Match
scheme to further decrease the error rate of extracted bits.
Gamelets - Multiplayer Mobile Games with Distributed Micro-Clouds [Full Text]Anand Bhojan
In this paper, we propose a novel distributed micro-cloud infrastructure with a next generation device called Gamelet to mitigate the limitations in traditional cloud system for multiplayer cloud gaming on resource limited mobile devices. The paper also investigates the opportunities, issues and possible solutions for Gamelet infrastructure for mobile games with a demonstrable prototype.
Experimental Applications of Mapping Services in Wireless Sensor NetworksM H
Wireless sensor networks typically gather data at a number of locations. However, it is desirable to be able to design applications and reason about the data in more abstract forms than points of data. This paper examines one way in which this can be done. By bestowing the ability to predict inter-node values upon the network, it is proposed that it will become possible to build applications that are unaware of the concrete reality of sparse data. This interpolation capability is realised as a service of the network. We present an implementation of this service and discuss its merits and shortcomings. Additionally, we present an initial application of the service in the form of isopleth generation. That is, the delineation of contours of constant parameter value. Finally, we discuss the improvements required to create more sophisticated applications and services and examine the benefits these improvements would bring.
CONFIGURABLE TASK MAPPING FOR MULTIPLE OBJECTIVES IN MACRO-PROGRAMMING OF WIR...ijassn
Macro-programming is the new generation advanced method of using Wireless Sensor Network (WSNs), where application developers can extract data from sensor nodes through a high level abstraction of the system. Instead of developing the entire application, task graph representation of the WSN model presents simplified approach of data collection. However, mapping of tasks onto sensor nodes highlights several problems in energy consumption and routing delay. In this paper, we present an efficient hybrid approach of task mapping for WSN – Hybrid Genetic Algorithm, considering multiple objectives of optimization – energy consumption, routing delay and soft real time requirement. We also present a method to configure the algorithm as per user's need by changing the heuristics used for optimization. The trade-off analysis between energy consumption and delivery delay was performed and simulation results are presented. The algorithm is applicable during macro-programming enabling developers to choose a better mapping according to their application requirements.
Protecting the data in a safe and secure way which does not impede the access of an authorized authority is an immensely difficult and very interesting research problem. image cryptography is a special type of encryption technique to obscure image-based secret information which can be decrypted by Human Visual System. Communication is the process of transmitting information from source to destination. The exchanging information should not be stolen by unauthorized parties like hackers while sending or receiving via channel. To avoid this stealing of the information visual cryptography techniques are used. This paper proposes a novel method for key generation by using nearest prime pixels. Further 2’s complement and logical operations are performed to generate decrypted image. The final decrypted image is generated by representing pixels in matrix form and data is retrieved in column wise.
Graphical Password Authentication Using Modified Persuasive Cued Click-Pointpaperpublications3
Abstract: There are many authentication systems which are used for computer based authentication purpose. Generally user have put username and password in alphanumeric form. But alphanumeric password which is easy for remember can be guess by any hacker/attacker. But strong password given by user which is hard to remember sometime. This paper introduces modified persuasive cued click point authentication system. In persuasive cued click point authentication system, due to viewport chances of hotspot creation is increased. Due to elimination of viewport, hotspot creation is reduced in the modified PCCP, which gives more security to system.Keywords: Cued Click Points, Graphical Password, Modified PCCP, Server side images, and security.
Title: Graphical Password Authentication Using Modified Persuasive Cued Click-Point
Author: Mohite Sandhya, Kare Rohini, Bhongale Pooja, Bhosale Priyanka, Prof. Parchure S.V.
International Journal of Recent Research in Mathematics Computer Science and Information Technology
ISSN 2350-1022
Paper Publications
A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...cscpconf
This paper presents a hybrid approach for images and video super-resolution. We have proposed the approach for enhancing the resolution of images and low resolution, under
sampled videos. We exploited the shift and motion based robust super-resolution (SR)algorithm [1] and the diffusion image regularization method proposed in [2] to obtain the alias free and jerk free smooth SR image.We presented a framework for obtaining super-resolution video thatis robust,even in the presence of fast changing video frames. Wecompare our hybrid
approach framework’s simulation results with different resolution enhancement techniques i.e. Robust Super-resolution, IBP and Interpolation methods reported in the literature. This
approach shows good results in term of different quality parameters.
Introduction to the Special issue on ‘‘Future trends in robotics and autonomo...Anand Bhojan
Robotics is an extremely dynamic field with thriving advancement in its technology. As research progresses in robotic systems, more and more aspects of vision based processing, GPS enabled services, Autonomous techniques, very far distance communication in robots, dynamic environment handling, mobility techniques, multi-agent control and coordination techniques, multi-robot communication and coordination are explored to make robotics intelligent and to do specific tasks. Vision has helped in many areas for better services and fastens the process for localized results. Advancements in communication, positioning and localization techniques brought the robotics beyond the controlled industrial environments to more dynamic outdoor environments. Research in autonomous and other intelligent techniques has made robots capable of taking decisions in complex environments. The book covers future trends in robotics research topics including motion path planning, routing in dynamic environments, multi-agent control techniques, nature inspired algorithms and synchronization techniques with interesting applications.
ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...Anand Bhojan
The use of recording applications for teaching is increasingly popular, with tools such as Ink2Go and ShowMe being available on many platforms and at reasonable cost. However, most recording applications available are typically native applications and do not work within the web browser. In this work, we study the feasibility of implementing a recording solution in the web browser environment.
Poster: http://www.comp.nus.edu.sg/~bhojan/papers/showNtell15.png
Details and Demo @ www.sntboard.com
In this paper, we present simple and novel method to procedurally generate the game maps for multiplayer shooter games faster (in order of seconds) without compromising the expected features of a good multiplayer shooter environment.
AN ENHANCED SEPARABLE REVERSIBLE DATA HIDING IN ENCRYPTED IMAGES USING SIDE M...Editor IJMTER
This paper proposes a scheme for Enhanced Separable Reversible Data Hiding in
Encrypted images Using Side Match. In the first step the original image is encrypted using an
encryption key. Then additional data is embedded into the image by modifying a small portion of the
encrypted image using a data hiding key. With an encrypted image containing additional data, if a
receiver has the data hiding key, he can extract the additional data. If the receiver has the encryption
key, he can decrypt the image, but cannot extract the additional data. If the receiver has both the data
hiding key and encryption key, he can extract the additional data and recover the original content by
exploiting the spatial correlation in natural images. The accuracy of data extraction is improved by
using a better scheme for measuring the smoothness of the received image, and uses the Side Match
scheme to further decrease the error rate of extracted bits.
Gamelets - Multiplayer Mobile Games with Distributed Micro-Clouds [Full Text]Anand Bhojan
In this paper, we propose a novel distributed micro-cloud infrastructure with a next generation device called Gamelet to mitigate the limitations in traditional cloud system for multiplayer cloud gaming on resource limited mobile devices. The paper also investigates the opportunities, issues and possible solutions for Gamelet infrastructure for mobile games with a demonstrable prototype.
Experimental Applications of Mapping Services in Wireless Sensor NetworksM H
Wireless sensor networks typically gather data at a number of locations. However, it is desirable to be able to design applications and reason about the data in more abstract forms than points of data. This paper examines one way in which this can be done. By bestowing the ability to predict inter-node values upon the network, it is proposed that it will become possible to build applications that are unaware of the concrete reality of sparse data. This interpolation capability is realised as a service of the network. We present an implementation of this service and discuss its merits and shortcomings. Additionally, we present an initial application of the service in the form of isopleth generation. That is, the delineation of contours of constant parameter value. Finally, we discuss the improvements required to create more sophisticated applications and services and examine the benefits these improvements would bring.
CONFIGURABLE TASK MAPPING FOR MULTIPLE OBJECTIVES IN MACRO-PROGRAMMING OF WIR...ijassn
Macro-programming is the new generation advanced method of using Wireless Sensor Network (WSNs), where application developers can extract data from sensor nodes through a high level abstraction of the system. Instead of developing the entire application, task graph representation of the WSN model presents simplified approach of data collection. However, mapping of tasks onto sensor nodes highlights several problems in energy consumption and routing delay. In this paper, we present an efficient hybrid approach of task mapping for WSN – Hybrid Genetic Algorithm, considering multiple objectives of optimization – energy consumption, routing delay and soft real time requirement. We also present a method to configure the algorithm as per user's need by changing the heuristics used for optimization. The trade-off analysis between energy consumption and delivery delay was performed and simulation results are presented. The algorithm is applicable during macro-programming enabling developers to choose a better mapping according to their application requirements.
Protecting the data in a safe and secure way which does not impede the access of an authorized authority is an immensely difficult and very interesting research problem. image cryptography is a special type of encryption technique to obscure image-based secret information which can be decrypted by Human Visual System. Communication is the process of transmitting information from source to destination. The exchanging information should not be stolen by unauthorized parties like hackers while sending or receiving via channel. To avoid this stealing of the information visual cryptography techniques are used. This paper proposes a novel method for key generation by using nearest prime pixels. Further 2’s complement and logical operations are performed to generate decrypted image. The final decrypted image is generated by representing pixels in matrix form and data is retrieved in column wise.
Graphical Password Authentication Using Modified Persuasive Cued Click-Pointpaperpublications3
Abstract: There are many authentication systems which are used for computer based authentication purpose. Generally user have put username and password in alphanumeric form. But alphanumeric password which is easy for remember can be guess by any hacker/attacker. But strong password given by user which is hard to remember sometime. This paper introduces modified persuasive cued click point authentication system. In persuasive cued click point authentication system, due to viewport chances of hotspot creation is increased. Due to elimination of viewport, hotspot creation is reduced in the modified PCCP, which gives more security to system.Keywords: Cued Click Points, Graphical Password, Modified PCCP, Server side images, and security.
Title: Graphical Password Authentication Using Modified Persuasive Cued Click-Point
Author: Mohite Sandhya, Kare Rohini, Bhongale Pooja, Bhosale Priyanka, Prof. Parchure S.V.
International Journal of Recent Research in Mathematics Computer Science and Information Technology
ISSN 2350-1022
Paper Publications
A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...cscpconf
This paper presents a hybrid approach for images and video super-resolution. We have proposed the approach for enhancing the resolution of images and low resolution, under
sampled videos. We exploited the shift and motion based robust super-resolution (SR)algorithm [1] and the diffusion image regularization method proposed in [2] to obtain the alias free and jerk free smooth SR image.We presented a framework for obtaining super-resolution video thatis robust,even in the presence of fast changing video frames. Wecompare our hybrid
approach framework’s simulation results with different resolution enhancement techniques i.e. Robust Super-resolution, IBP and Interpolation methods reported in the literature. This
approach shows good results in term of different quality parameters.
Introduction to the Special issue on ‘‘Future trends in robotics and autonomo...Anand Bhojan
Robotics is an extremely dynamic field with thriving advancement in its technology. As research progresses in robotic systems, more and more aspects of vision based processing, GPS enabled services, Autonomous techniques, very far distance communication in robots, dynamic environment handling, mobility techniques, multi-agent control and coordination techniques, multi-robot communication and coordination are explored to make robotics intelligent and to do specific tasks. Vision has helped in many areas for better services and fastens the process for localized results. Advancements in communication, positioning and localization techniques brought the robotics beyond the controlled industrial environments to more dynamic outdoor environments. Research in autonomous and other intelligent techniques has made robots capable of taking decisions in complex environments. The book covers future trends in robotics research topics including motion path planning, routing in dynamic environments, multi-agent control techniques, nature inspired algorithms and synchronization techniques with interesting applications.
ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...Anand Bhojan
The use of recording applications for teaching is increasingly popular, with tools such as Ink2Go and ShowMe being available on many platforms and at reasonable cost. However, most recording applications available are typically native applications and do not work within the web browser. In this work, we study the feasibility of implementing a recording solution in the web browser environment.
Poster: http://www.comp.nus.edu.sg/~bhojan/papers/showNtell15.png
Details and Demo @ www.sntboard.com
In this paper, we present simple and novel method to procedurally generate the game maps for multiplayer shooter games faster (in order of seconds) without compromising the expected features of a good multiplayer shooter environment.
mumble: Framework for Seamless Message Transfer on SmartphonesAnand Bhojan
This work explores the possibility of transferring data between mobile devices that are nearby each other without the need of pairing, authentication, superuser access and Internet connectivity. Such technology can be used for emergency broadcast, traffic congestion avoidance, IoT, smart city, social and dating systems and games. Except Bluetooth Low Energy technology, there is no other protocol or technology available that allows pairing-free data transfer without Internet connectivity. Bluetooth Low Energy is capable of doing so but at a relatively short range. Using Wi-Fi Direct's Service Broadcast and Discovery, a simple yet novel method is developed which allows for seamless long range (Wi-Fi range) data transfer without Internet connectivity.
The design details of FATS-mumble can be found in the paper (http://www.comp.nus.edu.sg/~bhojan/papers/mumble15.pdf) published in ACM Mobicom 2015 SmartObjects. Please use it for all the references of this mumble project.
The project is open sourced (Github Link): https://github.com/FatsGW/FATS-Mumble.
Demo: http://youtu.be/ACmNYLd2NlI
ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...Anand Bhojan
The use of recording applications for teaching is increasingly popular, with tools such as Ink2Go and ShowMe being available on many platforms and at reasonable cost. However, most recording applications available are typically native applications and do not work within the web browser. In this work, we study the feasibility of implementing a recording solution in the web browser environment.
The design details of ShowNTell can be found in the paper (http://www.comp.nus.edu.sg/~bhojan/papers/ShowNTell15.pdf) published in IEEE ASEE 45th Frontiers In Education 2015 Conference. Please use it for all the references of this ShowNTell project.
Poster: http://www.comp.nus.edu.sg/~bhojan/papers/showNtell15.png
Details and Demo @ www.sntboard.com
NB:- There are new features such as real-time collaborative drawing and non-linear editing are added in the latest versions. Register at www.sntboard.com to get regular updates.
"The greatest challenge of education in the digital world is to train the mind to focus with a strong feeling of purpose where the time shrinks and to re-focus at will" - Dr Bhojan Anand
Towards Fog-Assisted Virtual Reality MMOG with Ultra-Low LatencyIJCNCJournal
In this paper, we propose a method to realize a virtual reality MMOG (Massively Multiplayer Online Video Game) with ultra-low latency. The basic idea of the proposed method is to introduce a layer consisting of several fog nodes between clients and cloud server to offload a part of the rendering task which is conducted by the cloud server in conventional cloud games. We examine three techniques to reduce the latency in such a fog-assisted cloud game: 1) To maintain the consistency of the virtual game space, collision detection of virtual objects is conducted by the cloud server in a centralized manner; 2) To reflect subtle changes of the line of sight to the 3D game view, each client is assigned to a fog node and the head motion of the player acquired through HMD (Head-Mounted Display) is directly sent to the corresponding fog node; and 3) To offload a part of the rendering task, we separate the rendering of the background view from that of the foreground view, and migrate the former to other nodes including the cloud server. The performance of the proposed method is evaluated by experiments with an AWS-based prototype system. It is confirmed that the proposed techniques achieve the latency of 32.3 ms, which is 66 % faster than the conventional systems.
With the arrival of cloud technology, game accessibility and ub
iquity have a bright future; Games can be
hosted in a centralize server and accessed through the Internet by a thin client on a wide variety of devices
with modest capabilities: cloud gaming. However, current cloud gaming systems have very strong
requireme
nts in terms of network resources, thus reducing the accessibility and ubiquity of cloud games,
because devices with little bandwidth and people located in area with limited
and unstable
network
connectivity, cannot take advantage of these cloud services.
In this paper we present an adaptation technique inspired by the level of detail (LoD) approach in 3D
graphics. It delivers multiple platform accessibility
and network adaptability
, while improving user’s
quality of experience (QoE) by reducing the impa
ct of poor
and
unstable
network parameters (delay,
packet loss, jitter) on game interactivity. We validate our approach using a prototype game
in a controlled
environment
and characterize the user QoE in a pilot experiment. The results show that the propos
ed
framework provides a significant QoE enhancement
FPGA Based Pattern Generation and Synchonization for High Speed Structured Li...TELKOMNIKA JOURNAL
Recently, structured light 3D imaging devices have gained a keen attention due to their potential
applications to robotics, industrial manufacturing and medical imaging. Most of these applications require
high 3D precision yet high speed in image capturing for hard and/or soft real time environments. This
paper presents a method of high speed image capturing for structured light 3D imaging sensors with FPGA
based structured light pattern generation and projector-camera synchronization. Suggested setup reduces
the time for pattern projection and camera triggering to 16msec from 100msec that should be required by
conventional methods.
DISTRIBUTED SYSTEM FOR 3D REMOTE MONITORING USING KINECT DEPTH CAMERAScscpconf
This article describes the design and development ofa system for remote indoor 3D monitoring
using an undetermined number of Microsoft® Kinect sensors. In the proposed client-server
system, the Kinect cameras can be connected to different computers, addressing this way the
hardware limitation of one sensor per USB controller. The reason behind this limitation is the
high bandwidth needed by the sensor, which becomes also an issue for the distributed system
TCP/IP communications. Since traffic volume is too high, 3D data has to be compressed before
it can be sent over the network. The solution consists in self-coding the Kinect data into RGB
images and then using a standard multimedia codec to compress color maps. Information from
different sources is collected into a central client computer, where point clouds are transformed
to reconstruct the scene in 3D. An algorithm is proposed to conveniently merge the skeletons
detected locally by each Kinect, so that monitoring of people is robust to self and inter-user
occlusions. Final skeletons are labeled and trajectories of every joint can be saved for event
reconstruction or further analysis.
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
Mobile cloud game in high performance computing environmentTELKOMNIKA JOURNAL
Mobile cloud game is a solution to play high-end games in indigent thin clients with a diversity of end-user devices, and as real-time gaming, mobile cloud game hosting game engines in the cloud. Moreover, frequent change in network quality is another issue that should be limited to run the real fast cloud game. Thus, reliable software components between cloud and user devices as clients, including using artificial intelligence (AI) algorithms such as machine learning, deep learning and so on will enhance the game performance, particularly in multiplayer and real-time conditions. In this paper, we list the mobile cloud game architecture in the high-performance computing (HPC) environment, where a load of the game will be distributed between servers as cloud and clients. The server node as clouds or clients will consist of more than one server with many processors (cores) or sometimes can be recognized as distributed computing. Using HPC for cloud games will boost the game performance where the execution times will be dispersed not only in some node in servers and clients but in many cores of each server or client. The involvement of the internet of things (IoT) and ubiquitous access from heterogeneous devices will give benefit to enjoyment in the game itself.
Indoor 3D Video Monitoring Using Multiple Kinect Depth-Camerasijma
This article describes the design and development of a system for remote indoor 3D monitoring using an
undetermined number of Microsoft® Kinect sensors. In the proposed client-server system, the Kinect
cameras can be connected to different computers, addressing this way the hardware limitation of one
sensor per USB controller. The reason behind this limitation is the high bandwidth needed by the sensor,
which becomes also an issue for the distributed system TCP/IP communications. Since traffic volume is too
high, 3D data has to be compressed before it can be sent over the network. The solution consists in selfcoding the Kinect data into RGB images and then using a standard multimedia codec to compress color
maps. Information from different sources is collected into a central client computer, where point clouds are
transformed to reconstruct the scene in 3D. An algorithm is proposed to merge the skeletons detected
locally by each Kinect conveniently, so that monitoring of people is robust to self and inter-user occlusions.
Final skeletons are labeled and trajectories of every joint can be saved for event reconstruction or further
analysis.
Indoor 3 d video monitoring using multiple kinect depth camerasijma
This article describes the design and development of a system for remote indoor 3D monitoring using an
undetermined number of Microsoft® Kinect sensors. In the proposed client-server system, the Kinect
cameras can be connected to different computers, addressing this way the hardware limitation of one
sensor per USB controller. The reason behind this limitation is the high bandwidth needed by the sensor,
which becomes also an issue for the distributed system TCP/IP communications. Since traffic volume is too
high, 3D data has to be compressed before it can be sent over the network. The solution consists in selfcoding
the Kinect data into RGB images and then using a standard multimedia codec to compress color
maps. Information from different sources is collected into a central client computer, where point clouds are
transformed to reconstruct the scene in 3D. An algorithm is proposed to merge the skeletons detected
locally by each Kinect conveniently, so that monitoring of people is robust to self and inter-user occlusions.
Final skeletons are labeled and trajectories of every joint can be saved for event reconstruction or further
analysis.
DIVISION AND REPLICATION OF DATA IN GRID FOR OPTIMAL PERFORMANCE AND SECURITYijgca
Using Grid Storage, users can remotely store their data and enjoy the on-demand high quality applications and services from a shared networks of configurable computing resources, without the burden of local data storage and maintenance. In this project based on the dynamic secrets proposed design an encryption scheme for SG wireless communication, named as dynamic secret-based encryption (DSE). Dynamic encryption key (DEK) is updated by XOR the previous DEK with current DS. In this project based on the dynamic secrets proposed design an encryption scheme for SG wireless communication, named as dynamic secret-based encryption (DSE). The basic idea of dynamic secrets is to generate a series of secrets from unavoidable transmission errors and other random factors in wireless communications In DSE, the previous packets are coded as binary values 0 and 1 according to whether they are retransmitted due to channel error. This 0/1 sequence is called as retransmission sequence (RS) which is applied to generate dynamic secret (DS). Dynamic encryption key (DEK) is updated by XOR the previous DEK with current DS.
DIVISION AND REPLICATION OF DATA IN GRID FOR OPTIMAL PERFORMANCE AND SECURITYijgca
Using Grid Storage, users can remotely store their data and enjoy the on-demand high quality applications and services from a shared networks of configurable computing resources, without the burden of local data storage and maintenance. In this project based on the dynamic secrets proposed design an encryption scheme for SG wireless communication, named as dynamic secret-based encryption (DSE). Dynamic encryption key (DEK) is updated by XOR the previous DEK with current DS. In this project based on the dynamic secrets proposed design an encryption scheme for SG wireless communication, named as dynamic secret-based encryption (DSE). The basic idea of dynamic secrets is to generate a series of secrets from unavoidable transmission errors and other random factors in wireless communications In DSE, the previous packets are coded as binary values 0 and 1 according to whether they are retransmitted due to channel error. This 0/1 sequence is called as retransmission sequence (RS) which is applied to generate dynamic secret (DS). Dynamic encryption key (DEK) is updated by XOR the previous DEK with current DS.
DIVISION AND REPLICATION OF DATA IN GRID FOR OPTIMAL PERFORMANCE AND SECURITYijgca
Using Grid Storage, users can remotely store their data and enjoy the on-demand high quality applications and services from a shared networks of configurable computing resources, without the burden of local data storage and maintenance. In this project based on the dynamic secrets proposed design an encryption scheme for SG wireless communication, named as dynamic secret-based encryption (DSE). Dynamic encryption key (DEK) is updated by XOR the previous DEK with current DS. In this project based on the dynamic secrets proposed design an encryption scheme for SG wireless communication, named as dynamic secret-based encryption (DSE). The basic idea of dynamic secrets is to generate a series of secrets from unavoidable transmission errors and other random factors in wireless communications In DSE, the previous packets are coded as binary values 0 and 1 according to whether they are retransmitted due to channel error. This 0/1 sequence is called as retransmission sequence (RS) which is applied to generate dynamic secret (DS). Dynamic encryption key (DEK) is updated by XOR the previous DEK with current DS
Transformation and dynamic visualization of images from computer through an F...TELKOMNIKA JOURNAL
This article shows the implementation of a system that uses a graphic interface to load a digital image into a programmable logic device, which is stored in its internal RAM memory and is responsible for visualizing it in a matrix of RGB LEDs, so that This way, the LEDs show an equivalent to the image that was sent from the PC, conserving an aspect ratio and respecting as much as possible the color of the original image. To carry out this task, a Matlab script was designed to load the image, convert and format the data, which are transmitted to the FPGA using the RS232 protocol. The FPGA is in charge of receiving them, storing them and generating all the signals of control and synchronization of the system including the control of the PWM signals necessary to conserve the brightness of each one of the LEDs. This system allows the visualization of static images in standard formats and, in addition, thanks to the flexibility of the hardware used, it allows the visualization of moving images type GIF.
Metaverse - The 'Killer App' for 5G, 6G and BeyondAnand Bhojan
"Metaverse - The 'Killer App' for 5G, 6G and Beyond":
With the growing popularity of X-Reality (VR/AR/MR/Metaverse) generation of global media-data is expected to accelerate at much higher rate and they will drive innovations and advancements in mobile networks. In this talk, challenges in Metaverse and we will explore suitable system architectures for streaming and analytics for Metaverse with the 5G edge nodes.
Multimedia Analytics with 5G Edge NodesAnand Bhojan
Presentation at ACM-MM-19 TPC meeting: https://sites.google.com/comp.nus.edu.sg/mm19-tpc-meeting/workshop
With the growing popularity of X-Reality (VR/AR/MR) generation of global media-data is expected to accelerate at much higher rate. Most of these data need real-time processing such as real-time facial features/emotion recognition to environmental context recognition/mapping beyond the capabilities of an MR glass. The demand for high communication bandwidth and ultra-low latency (<15ms motion to photon latency) will drive the community to exploit the capabilities of the telco edge nodes and push telcos to build more powerful edge nodes. The real-time analytics of media data with strict real-time constraints will be the ‘killer application’ of 5G networks and edge computing. In addition, we believe, multimedia analytics and 5G/edge computing will be the key to push for mass adoption of XR devices. In this talk, we will explore some suitable media analytics methods and architecture to run media analytics with the 5G edge nodes.
Adaptive Video Content Manipulation for OLED Display Power ManagementAnand Bhojan
Emerged as a choice of display for vivid colours and power efficiency in expensive Smartphones, in recent days Organic Light Emitting Diode (OLED) displays are increasingly replacing Liquid Crystal Displays (LCD) and Plasma in Smartphones and large screen Televisions. Yet, OLED displays are one of the most energy consuming modules in modern Smartphones. Power consumption of OLED displays have strong correlation with the colours illuminated by its pixels. Improvements in pixel density by manufacturers makes the displays more power hungry. In this paper, we present Human Visual System (HVS) aware adaptive colour transformation and darkening methods to create power-efficient videos for OLED displays while preserving visual fidelity of the videos. Our measurements and subjective evaluations show that we can save up to 60% of energy consumption with minimum distortion to visual quality of the video. As our colour transformation process uses pre-computed colour-map and the darkening process uses the gamma-correction function which is already implemented in most of the display hardware, runtime computation overheads are minimum and hence, our method can be used to manipulate realtime/live video streams as well.
CloudHide: Towards Latency Hiding Techniques for Thin-client Cloud Gaming Anand Bhojan
[This is a short introduction pitch to the full paper.]
Though there is an explosion of cloud services in recent years, cloud gaming industry is still way behind in making a strong breakthrough. Despite many benefits of cloud gaming (video or pixel streaming of rendered game scenes), the primary factor that hinders the growth is "response latency", a well known issue. In this paper, we introduce a novel predictive paradigm of cloud gaming in order to mitigate latency that is inherent in cloud gaming systems. The main idea of the predictive paradigm is to pre-generate and send future outcome frames in advance to the thin-client so that the thin-client can respond immediately to any user input. The approach in this work seeks to exploit the near unlimited processing power of the cloud and ever increasing network bandwidth. The paper provides a generic and comprehensive theoretical model that shows the relationship between various parameters (network bandwidth, round-trip-time and processing time for game logic, predicting future outcomes, rendering, capturing, encoding and decoding) of predictive cloud gaming and response latency. The theocratical model design is started with an ideal server with unlimited resources for ease of understanding and then it is enhanced with resource constraints of a practically feasible non-ideal server. The model demonstrates significant amount of response latency reduction in predictive cloud gaming when compared to conventional cloud gaming. The model is analysed further to estimate the resource requirements for predictive cloud gaming. A simple game is built to demonstrate key aspects of the model and to conduct subjective evaluation of quality of experience (QoE) of the game play. The evaluations show significant gain in QoE. The resource requirements of the predictive cloud gaming gaming system grows exponentially with respect to prediction period of future frames. The paper illustrates some optimisation methods to reduce exponential growth and discusses directions for further works in this area.
SuperStreamer: Enabling Progressive Content Streaming in a Game EngineAnand Bhojan
This technical demonstration presents the SuperStreamer project, which enables progressive game assets streaming to players while games are played, reducing the startup time required to download and start playing a cloud-based game.
SuperStreamer modifies a popular game engine, Unreal Engine 4, to support developing and playing games with progressive game asset streaming. With SuperStreamer, developers can mark the minimal set of files, containing only the game content essential to start playing the game. When a player plays the game, these minimal set of files will be downloaded to the player's device. SuperStreamer also generates low resolution textures automatically when a developer publishes a game, and these low resolution textures are transmitted first into the game client. As players move through a game level, high quality textures required for the game will be downloaded. In our demo game, we are able to decrease the time taken to startup and load the first game level by around 30%.
Game balancing with ecosystem mechanismAnand Bhojan
To adapt game difficulty upon game character’s strength, Dynamic Difficulty Adjustment (DDA) and some other learning strategies have been applied in commercial game designs. However, most of the existing approaches could not ensure diversity in results, and rarely attempted to coordinate content generation and behaviour control together. This paper suggests a solution that is based on multi-level swarm model and ecosystem mechanism, in order to provide a more flexible way of game balance control.
Game Balancing with Ecosystem MechanismAnand Bhojan
To adapt game difficulty upon game character’s strength, Dynamic Difficulty Adjustment (DDA) and some other learning strategies have been applied in commercial game designs. However, most of the existing approaches could not ensure diversity in results, and rarely attempted to coordinate content generation and behaviour control together. This paper suggests a solution that is based on multi-level swarm model and ecosystem mechanism, in order to provide a more flexible way of game balance control.
TIṬAL – Asynchronous multiplayer shooter with procedurally generated mapsAnand Bhojan
Would it be possible to bring the promise of unlimited re-playability typically reserved for Roguelike games to competitive multiplayer shooters? This paper tries to address this issue by proposing a novel method to dynamically generate maps at run-time almost as soon as players press the Play button, while ensuring the features what players would expect from the genre. The procedures are simple and practically
feasible to be employed in actual computer games. In addition, the work experiments the possibility of incorporating asynchronous game-play element into a multiplayer shooter with human imitating bots where the players can let their bot/avatar replace them when they are not around. The algorithms are implemented and evaluated with a playable game. The evaluations prove that playable 3D dynamic mapscan be generated in order of seconds using game context data to initialise the parameters of the algorithm. The paper also shows that asynchronous game-play element is a possible feature for consideration in next generation multiplayer shooters.
mumble: Framework for Seamless Message Transfer on SmartphonesAnand Bhojan
This work explores the possibility of transferring data between mobile devices that are nearby each other without the need of pairing, authentication, superuser access and Internet connectivity. Such technology can be used for emergency broadcast, traffic congestion avoidance, IoT, smart city, social and dating systems and games. Except Bluetooth Low Energy technology, there is no other protocol or technology available that allows pairing-free data transfer without Internet connectivity. Bluetooth Low Energy is capable of doing so but at a relatively short range. Using Wi-Fi Direct's Service Broadcast and Discovery, a simple yet novel method is developed which allows for seamless long range (Wi-Fi range) data transfer without Internet connectivity.
The design details of FATS-mumble can be found in the paper (http://www.comp.nus.edu.sg/~bhojan/papers/mumble15.pdf) published in ACM Mobicom 2015 SmartObjects. Please use it for all the references of this mumble project.
The project is open sourced (Github Link): https://github.com/FatsGW/FATS-Mumble.
Demo: http://youtu.be/ACmNYLd2NlI
PARVAI - HVS Aware Adaptive Display Power Management for Mobile Games [Full T...Anand Bhojan
Displays are one of the major system energy consumers in smartphones. Power consumption of Organic Light-Emitting Diode (OLED) displays have direct relationship with the colour and intensity of the contents being displayed. Games are one of the most popular smartphone applications that consume relatively more resources and power. In this paper, we present Human Visual System (HVS) aware algorithms and techniques to reduce OLED display power consumption up to 45% while retaining the perceived quality of game content. We have implemented and evaluated our solution in the game Quake III and its Android port Kwaak3.
Energy Efficient Mobile Applications with Mobile Cloud Computing ( MCC )Anand Bhojan
http://www.icghia2014.in/keynotespeakers.htm
Conference Keynote Talk: International Conference on Graph Algorithms, High Performance Implementations and Applications Coimbatore (ICGHIA'14)
ARENA - Dynamic Run-time Map Generation for Multiplayer ShootersAnand Bhojan
In this paper, we present simple and novel method to procedurally generate the game maps for multiplayer shooter games faster (in order of seconds) without compromising the expected features of a good multiplayer shooter environment.
PARVAI - HVS Aware Adaptive Display Power Management for Mobile GamesAnand Bhojan
Displays are one of the major system energy consumers in smartphones. Power consumption of Organic Light-Emitting Diode (OLED) displays have direct relationship with the colour and intensity of the contents being displayed. Games are one of the most popular smartphone applications that consume relatively more resources and power. In this paper, we present Human Visual System (HVS) aware algorithms and techniques to reduce OLED display power consumption up to 45% while retaining the perceived quality of game content. We have implemented and evaluated our solution in the game Quake III and its Android port Kwaak3.
Mobile Cloud Computing for Games - Gamelet Anand Bhojan
In recent years, cloud computing services have been increasing in greater pace. High penetration rate of mobile devices and resource limited devices escalate the demand for cloud services further. Even though the cloud industry continues to grow exponentially, the cloud gaming service has been left behind due to the limitations in today's technology. There are three well known reasons for the slower growth - latency, server scalability (esp. bandwidth) and lack of game data at client side to use latency hiding and synchronisation techniques such as Dead-reckoning. In this paper, we propose a novel distributed micro-cloud infrastructure with a next generation device called Gamelet to mitigate the limitations in traditional cloud system for multiplayer cloud gaming on resource limited mobile devices. The paper also investigates the opportunities, issues and possible solutions for Gamelet infrastructure for mobile games with a demonstrable prototype.
Energy Efficient Multi-player Smartphone Gaming using 3D Spatial Subdivisioning and PVS Techniques [Full Text]
1. Energy Efficient Multi-player Smartphone Gaming using
3D Spatial Subdivisioning and PVS Techniques
Bhojan Anand, Zeng Qiang, Ananda L Akkihebbal
School of Computing, National University of Singapore
banand@comp.nus.edu.sg, zen1986@gmail.com, ananda@comp.nus.edu.sg
ABSTRACT
With the advent of feature rich smartphone platforms such as An-
droid and iOS, people can now enjoy a wide variety of applications
on-the-go. Among these applications, games are one of the most
desired types. However, as bigger screens, faster CPUs and inter-
faces supporting higher bandwidth (WiFi, 3G, LTE) consume more
power, battery lifetime becomes a bottleneck on such devices.
In this paper, we present novel techniques that combine 3D spa-
tial subdivisioning, Potentially Visible Set (PVS) and Visual Percep-
tion based Localisation (VPL) methods to estimate the non-critical
game states to save wireless interface energy with minimum pro-
cessing penalty. Our techniques and algorithms are realised in a
commercial game and can save up to 57% of wireless interface en-
ergy while retaining game play quality.
Categories and Subject Descriptors
C.2 [Computer-Communication Networks]: Wireless Commu-
nication; K.8 [Personal Computing]: Games
General Terms
Algorithms, Design, Experimentation, Human Factors
Keywords
Energy; Power; Wireless; Game; Mobile; Smartphone
1. INTRODUCTION
Energy. Growth of battery technology is not on par with the
rest of the technologies in modern smartphones. As most of us
are aware, very few systems today fulfill Mark Weiser’s vision of
“several days of continuous use” for the “Computer for the 21st
Century”. The three main sources of power consumption in smart-
phones are, 1) the processor (CPU/GPU), 2) the display, and 3)
the network interfaces. For the HTC Magic (Android samrtphone),
the measurements show 35-40%, 45-50% and 4-15% power con-
sumption by the wireless interface, the LCD and the rest (CPU,
memory...) respectively [3]. In this paper, we focus on conserving
power at the network interface level.
Games. Games are one of the most-downloaded categories of
smartphone applications. Mobile games are resource-intensive ap-
plications and they consume more power than other common ap-
plications. For example, in our measurements on the HTC Desire
Permission to make digital or hard copies of all or part of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed
for profit or commercial advantage and that copies bear this notice and the full cita-
tion on the first page. Copyrights for components of this work owned by others than
ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or re-
publish, to post on servers or to redistribute to lists, requires prior specific permission
and/or a fee. Request permissions from permissions@acm.org.
MM’13-IMMPD’13, October 22, 2013, Barcelona, Spain.
Copyright 2013 ACM 978-1-4503-2399-4/13/10 ...$15.00.
http://http://dx.doi.org/10.1145/2505483.2505491.
HD (Android smartphone), the battery provides 8 hrs of GSM talk
time, while it lasts for only 1hr 50min when the 3D game Kwaak3
[11] is played 1. In this paper, we focus on conserving power at
the network interface level in networked multiplayer 3D gaming
environments.
Network Interface. The key mechanism to reduce the power
consumption of wireless interfaces is to put the interface into sleep
mode whenever possible. The challenge is to decide, 1) when to
put the wireless interface into sleep mode? and, 2) how long can
it be in sleep mode without affecting the quality of the game play
adversely?. An obvious answer would be to sleep when the game
state is not important, from the player’s perspective. Almost all
player interactions in a game are with other players who are in a
perceptible range. Hence, the game state of a client is important
when there are other players in its perceptible range. However,
predicting the near-future game state for a client is challenging.
In this paper, we introduce a novel technique that uses the 3D
rendering engine’s knowledge to determine the importance of the
game state. Modern game engines use occlusion culling algorithms
[18, 9] to compute PVS for efficient rendering. We use this PVS
data to determine visibility between two players. In contrast to pre-
vious methods that overlay simple 2D squares or hexagons on the
3D maps [10] to compute distance or visibility between players,
we propose using the 3D data directly from the renderer. Hence
our approach is efficient (minimum computations are required), ac-
curate (we handle the 3D world as it is without overlaying any 2D
structures) and easy to integrate with modern game engines (PVS
is already implemented in most of these modern game engines).
In addition, we use VPL, to filter players based on their distance
to improve scalability. For VPL, we compute Maximum Percepti-
ble Distance (MPD) which is the distance up to which a player can
view other players or bots and interact. As human behaviour is hard
to predict, we use a feedback based error controlling mechanism to
keep the errors within a given threshold.
To test our algorithms, we used the commercially successful
Quake III [14] First Person Shooting (FPS) game and its Android
port (Kwaak3 [11] ). We used Quake III because it is open-sourced.
As a fast paced FPS game, Quake III brings maximum challenges
to our approach by demanding lower processing time and network
latency.
2. RELATED WORK
Power Conservation A large body of prior work has also looked
at saving wireless interface power. Meyer et. al [17] show that
power can be saved through better link utilisation while Anastasi
et. al [4] propose a proxy assisted power saving mechanism. Var-
ious authors have also proposed saving power by observing and
exploiting the statistical correlations between applications and the
1Kwaak3 is an Android port of the game Quake III [14]
2. network traffic [6, 22, 23]. These statistical methods have also been
applied to games [12, 13, 1].
Interest Management There are two widely used categories of
techniques for Area of Interest (AoI) or Interest Management: Dis-
tance and Visibility based techniques. There are various algorithms
for determining the AoI [10] such as Euclidean distance, square
tile distance, hexagonal tile distance [19, 8, 2, 15] and more com-
plex visibility and orientation-based algorithms [7]. However, in all
these works the authors overlay a 2D grid over a 3D game map han-
dling the 3D environment inaccurately. Such basic approaches, re-
sulted in more errors as the number of players increased. Moreover,
the pure visibility or AoV (Area-of-Visibility) based algorithms fail
to work for huge open maps where the players are always visible to
each other. In addition, pure AoV based approaches [19] are not
scalable. If there are n players, for each player pi, we always need
to check the remaining n−1 players for visibility. Hence the com-
putational cost will be O(n4) for checking and dynamically predict-
ing visibilities between each other.
There are two main differences in our work. Firstly, we consider
distance and visibility in 3D for improved accuracy. Secondly, we
use the 3D rendering engine’s data directly to avoid additional com-
puting overhead and to increase the efficiency of our algorithms.
3. POWER MANAGEMENT ALGORITHMS
Our power management algorithms are run in two phases, namely
macro scanning and micro scanning. Macro scanning filters the
players by distance and passes only relevant players for micro scan-
ning. Micro scanning queries the PVS from the current area of a
player and puts his wireless interface to sleep mode for a period of
time.
The pseudo code presented in Algorithm 1 outlines the main al-
gorithm.
Algorithm 1 Main
for each server frame do
for each active client do
if client is sleeping then
continue
else
do t = Macro_scan()
if t > 0 then
sleep for t msec
else
do t = Micro_scan()
if t > 0 then
sleep for t msec
end if
end if
end if
end for
end for
Area, Cluster and Portal in a Game World. Area is a space,
partially enclosed by geometric surfaces. Adjacent areas that share
same surfaces are grouped together to form a cluster [21]. Clusters
are separated by a portal. For example, door is a portal, because
player can travel from one cluster to another through it. Thus, a
map is logically divided into clusters through portals. This graph
structure together with areas forms a hierarchical network.
3.1 Macro Scanning Algorithm
We use path-distance to measure the distance from the position
of the current player to the clusters containing the MPD end point
as shown in the Figure 1, so only necessary clusters are added to
the queue for processing. Breadth-First Search (BFS) is used to
traverse clusters in nearest-first order and stopped at the MPD of
the current player. Current player is the player or client who is cur-
rently under consideration for power management. Detailed pseudo
Figure 1: BFS follows the Path Distance of each Cluster
code for macro scanning is presented in Algorithm 2. The algorithm
initialises sleep time to maximum sleep time. Maximum sleep time
is game specific. Setting the wireless interface to sleep mode be-
yond this duration produces noticeable latency in game play. It is
about 180ms [5] for fast paced FPS games and 400ms for slower
RPG games.
Algorithm 2 Macro_scan
clusterQueue: initialize to contain the current player’s (playeri) cluster
playerQueue: initialize to empty, will pass to micro scan
sleepTime: initialize to maximum sleepTime (game dependent value)
while clusterQueue is not empty do
cluster = pop clusterQueue
for each playerj in cluster; j = i do
if playerj’s distance to playeri > MPD then
calculate time playerj takes to MPD of playeri using MS
if time < sleepTime then
sleepTime = time
end if
else
put playerj in playerQueue
sleepTime = 0
end if
end for
for each portalk of cluster that has path-distance to playeri < MPD do
clusterOther = portalk− >otherCluster
add clusterOther to clusterQueue
end for
for each portalq of cluster that has path-distance to playeri > MPD do
calculate the time taken from the portalq to MPD
if time < sleepTime then
sleepTime = time
end if
end for
end while
return sleepTime
The distance from any player to the current player’s MPD can
be easily calculated from the positions of the players. We compute
the sleep time for the current player by dividing this distance by the
players’ relative movement speed as shown in Equation (1). As hu-
man behaviour in a game play is complex, it is hard to predict when
and at what speed (and direction) the user will move his character.
Hence, computing relative movement speed for each pair of play-
ers (n2 combinations in worst case) at every frame is complex and
would incur too much overhead. To keep it practically feasible,
we use estimated average movement speed (here after we simply
refer it as movement speed). If this estimated movement speed is
lower than actual movement speed of the players, there are more
possibilities for errors. Hence, the estimated movement speed is
dynamically tweaked based on error rate, which we describe in the
section 3.3. Error rate is a ratio of total sleep time with error to
total sleep time in a given period.
3. We stop traversing the clusters after reaching the MPD. Once
the MPD is reached, we conservatively assume that there could
be a player in the next (unexplored) cluster, just behind the por-
tal (door), and he could enter the current cluster (and the MPD of
current player) in no time. To cater for such incidents, the algo-
rithm takes the distance between the MPD of the current player to
the nearest portal (excluding the portal through which the BFS en-
ters the cluster) in the current cluster and computes the sleep time
based on this distance. The shortest of all the sleep times is returned
by the algorithm. Note, if the BFS finds another player within the
MPD distance of current player, it will return zero as sleep time.
Sleep Time =
MPDdistance
2×MS
(1)
where, MPDdistance - is distance from a player to the MPD of the
current player; MS - is estimated movement speed of players.
As BFS scans the clusters in incremental order and stops at MPD,
our macro scanning is naturally scalable. In huge maps (as in
MMOG games), it is highly efficient.
3.2 Micro Scanning Algorithm
As our spatial subdivision scheme (eg. BSP tree in Quake III
game, Octree in Ogre3D engine) is pre-computed by 3D renderer
and stored together with the data structure of the game map, mi-
cro scanning becomes simple and efficient. For the current player
under consideration we get his PVS from the renderer and check
whether any of the players in the player queue (created by macro
scan) is present in the PVS. The PVS data is usually stored in the
form of bit-array. For example, in Quake III, Area x is visible to
area y if (1 << y%8) bit of PVS[x∗no. of areas+y/8] is set. As-
suming the bit-array structure for PVS, the pseudo code for micro
scanning is described in Algorithm 3.
Algorithm 3 Micro_scan
playerQueue: passed as parameter
currentPlayer: index no. of current player
for each player inside playerQueue do
if (1<<player%8) & PVS[currentPlayer ∗ no. of areas + player/8]==1 then
return NO_SLEEP
end if
end for
return FIXED_SLEEP_TIME
In micro scanning, the time to sleep is always fixed. We call it
Fixed Sleep Time (FST). When there are no other players in the
current player’s PVS the client is put into sleep for a FST time. If
FST is too high, there are more possibilities for errors. The FST
value is dynamically tweaked based on the error rate, which we
describe in Section 3.3.
3.3 Feedback based Error Controller
There is a trade off between the error rate and amount of power
saved. The error rate always grows together with the total power
saved. The error threshold (maximum acceptable error rate) really
depends on how much an user can bear with the affected game ex-
perience. At the time of low battery level, he may want to sacrifice
a bit to save more power.
To control the errors according to the Error Threshold (ET) set
by the user, we need to tweak two complex parameters - Move-
ment Speed (MS) described in Section 3.1 and Fixed Sleep Time
(FST) described in Section 3.2 to optimal levels for maximum en-
ergy saving. In our current implementation we have used Additive
Increase and Multiplicative Decrease (AIMD) method to make our
system more conservative on quality. Once the error threshold is
passed we enforce large changes (multiplicative) to FST and MS
to bring down the error as fast as possible. The key advantage of
multiplicative decrease is that it enables all other clients to get the
correct state (eg. position) information about the current player
much faster, resulting in rapid global game state synchronisation
among all clients.
As the game activities are very random and human behaviours
are hard to predict, such feedback based control systems are very
robust and adaptive to different requirements. The pseudo code
presented in Algorithm 4 outlines the AIMD based error control
procedure. Average Error Rate (AER) in the algorithm is the run-
ning average error rate for the past ’t’ time period. Running average
is used to avoid any abrupt changes to FST and MS. Note: After
several sets of experiments we have arrived at the values 2 and 1/4
of the current value for increment or decrement. These values gave
reasonably fast convergence and a well controlled error level.
Algorithm 4 AIMD Based Run-Time Error Controller
- Initialise: FST = 200ms, MS = 500 game unit/ms. Game unit is pixels in most
games. Movement speed here is the normal average movement speed in game. 500
pixels/ms is obtained from multiple game play experiments.
- Let the user set an ET between 1%, 3% and 5%. Higher threshold results in lower
quality and higher power saving. It determines the trade-off between quality and
power.
- Assumption : Main algorithm is running (macro and micro scanning).
while game is not end do
update AER
if AER < ET then
FST = FST + 2
MS = MS − 2
else
FST = FST + 1
4 ∗FST
MS = MS − 1
4 ∗MS
end if
end while
4. IMPLEMENTATION
The power management algorithms were implemented and tested
in a community maintained copy of Quake III source called io-
quake3 [14].
We implemented our algorithms in the Quake III server as the
server knows the entire state information of the clients. We used
Ubuntu 10.10 with Eclipse as our development platform. The server
estimated the client’s state and the time period that the client could
sleep without affecting the quality of game play adversely.
In our implementation, we added a new sleep command on top of
the existing client/server communication in the game. The server
used this sleep command to tell specific clients how long to sleep.
Upon receiving this command, the client sent a signal to the wire-
less hardware to make it sleep for the specified time period.
To put the wireless card into sleep mode and wake up fast we
used Madwifi driver [16], which is an advanced Linux based WiFi
driver for WiFi cards with Atheros chipset. MadWifi comes with
Atheros Hardware Access Layer (HAL). The HAL provides hard-
ware support for wireless network adapters. More details on this
can be obtained through the ubuntu man pages [20].
5. EVALUATION METHODOLOGY
Figure 2 shows the testbed used for our experiments. The game
server was implemented and ran on a Dell Precision T7500 com-
puter which came with Intel(R) Xeon(R) E5520 2.26GHz/8MB L3
Cache/12 GB RAM. The game clients were connected through a
wired network and different wireless networks (WiFi, 3G) to the
game server. We used a variety of Lenovo Thinkpad laptops (Model
T61/T60/W500) and Android phones (HTC Magic/Dream/Nexus
One/HTC Desire HD) as game clients. To obtain accurate power
measurements, we used a high-speed multifunction Data Acquisi-
4. Figure 2: Testbed: Game Play and Power Measurement
tion Equipment (DAQ) USB-6251 and a Signal Conditioning Equip-
ment (SC-2345) from National Instruments (NI).
5.1 Defining Evaluation Metrics
In this section we describe our evaluation metrics.
Definition of Error. We define a miss or an error as the case
when a client, on waking up from sleep, finds that there is some
other client in its PVS and MPD.
Sources of Errors. There are 3 scenarios for an error to occur:
1) Other players teleport near to the current player (instantly ap-
pear). 2) Other players run fast towards the MPD region during the
sleep period (Estimated average movement speed in Equation (1)
is too low). 3) The boundary area in PVS test is too small and the
other player is just near to that area and moving towards the current
player (fixed sleep time in micro-scan is too long).
The error rate and quality of the game are computed as,
ErrorRate =
Total sleep time with errors
Total sleep time
(2)
Accuracy = 1−ErrorRate (3)
Open and Closed map. We define Open Map as the map which
consists of only a few loosely distributed buildings and most of the
areas in map are flat planes. (Eg. - Simpson’s map of Quake III).
Closed Map is defined as the map that is mainly constructed by
architectural structures, such as the interiors of a castle. It consists
of many rooms separated by door-ways. (Eg. - Q3DM7).
5.2 Small Scale User Study - Methodology
We did a small scale user study with 21 non-author students from
our school to study the effects of the power saving algorithm on
human players’ game play experience. There were 16 male and 5
female students in the age group 22-30.
The study procedure was as follows: First, the participants filled
up basic demographic information in the survey form. Then, they
played the original Quake III game for 10 minutes to get some train-
ing. After that, they played six randomly chosen games from eight
versions (including the original and our modified versions with ET
values 1%, 2%, 3%, 4%, 5%, 6%, 7%). The users compared the
game plays and gave their observation on the quality of the game
play by filling up the questions in the survey form. All the evalua-
tions were done with Quake III game in a networked environment
with up to 10 concurrent players.
5.3 Experiments
For each of our experiments, we ran the game client and played
for around five minutes to capture the data. We varied the maps,
number of players (2, 4, 8, 16...) and networks (WiFi, 3G-WCDMA)
in each of these experiments. We used up to 10 concurrent human
players in each run. To ensure high interactivity and interesting
game play we added a few moderately intelligent bots with the
real human players. We used three different maps developed by
the Quake III Arena team (q3dm1, q3dm7, Simpsons). q3dm1 and
q3dm7 are highly occluded closed maps and Simpsons is a open
map with fewer occlusions.
Though we varied the number of clients in each run, to have a
good mix, only some of them ran in power saving mode while the
rest ran in normal mode. Such a combination is realistic and it
enabled us to study the effect of power saving clients on normal
clients. We measured the WiFi interface power consumption of the
mobile clients that were connected to the DAQ.
6. EVALUATION RESULTS
We measured the power (WiFi interface) and error convergence
rate for various map types, player density and error thresholds in
Quake III game.
6.1 Effects of Map Type
Our first experiment was to measure the effects of map type on
amount of energy saved. We set all other variables to fixed values
except the map type. The settings for the experiment are shown in
the Table 1.
Table 1: Experiment Variables Setup for Section 6.1
Open Map Closed Map
Map: Simpsons Q3DM7
Player no.: 4 4
Error Threshold: 3% 3%
MPD: 2500 game units 2500 game units
Fixed Sleep Time: 200 200
Movement Speed: 500 500
Game Length: 15 minutes 15 minutes
The graph in Figure 3a shows the test results. The values rep-
resent the total energy (in percentage) saved per client. For closed
maps we saved 34.39% energy with more than 97% accuracy. It is
much higher than 20% energy saving (with same level of accuracy
and player density) reported in our previous work for closed maps
[19].
The contributions of Macro Scan and Micro Scan were logged at
runtime. Overall, in closed maps we saved 4% more energy than
in open maps. This shows our algorithm can function well on both
open maps and closed maps. Most of the energy saved with the
open map came from macro scan while most of the energy saved
with the closed map came from micro scan. This is expected, be-
cause the PVS approach doesn’t gain much on an open map as there
are very few geometric occlusions. Visible distance between play-
ers in an open map is more likely to be greater than the MPD, thus
macro scan can save more energy. But in a closed map, the players
are within the MPD most of time, thus, it is not possible for macro
scan to return a valid sleep time. This experiment shows the impor-
tance of using both macro scan and micro scan to adapt to different
map types and to save the maximum possible energy.
6.2 Effects of Energy Threshold
From the previous experiments we can see, the amount of en-
ergy saved with different map types differs only with respect to
the contribution of macro scan and micro scan. Since they are
equivalently important, in the following measurements we are con-
cerned only about the total amount of energy saved. We chose to
5. 34.39%
1.91%
32.48%
30.30%
27.95%
2.35%
Total
Macro
Micro
Open Map Closed Map
16.86%
27.20%
49.02%
1
3
5
ET
%
Energy
Saved
with
8
players
27.20%
34.39%
57.43%
8
4
2
No.
of
Players
Energy
Saved
at
3%
ET
a. MapType vs. b. Error Threshold c. Player Density
Energy Saved vs. Player Density vs. Energy Saved
Figure 3: Effect of map type, error threshold and player density on energy saving
use the closed map for the remaining experiments thus eliminat-
ing the macro scan. With other parameters unchanged, we had two
variables, namely number of players and Error Threshold (ET) to
consider.
Firstly we measured the effects of different error thresholds on
the amount of energy saved. We fixed the number of players to 8,
and ran the experiments with ET of 1%, 3% and 5%. The result in
Figure 3b shows the percentage of energy saved for each value of
ET. The general trend is that for higher ET, more energy is saved.
The rationale behind the increase is that the higher ET results in
higher Fixed Sleep Time. From ET 1% to 3%, there is a 10.34%
increase in the amount of energy saved. From ET 3% to 5%, there
is a 21.82% increase. The non-linearity of increase is explained by
unpredictable fluctuations in players’ interactions.
6.3 Effects of Player Density
To measure the effects of the number of players (player density)
on the amount of energy saved, we fixed ET to 3%, and changed the
number of players to 2, 4 and 8. Note, for FPS games with highly
occluded maps (eg. Quake III), the average number of concurrent
players is around 6, though the game can accept up to a maximum
of 16 players for certain maps [10].
As shown in Figure 3c, fewer players resulted in more energy
savings. For a 2-player game, the current player is in fact alone
for most of the time. Intuitively, the client should sleep up to 80%
of time given that the map Q3DM7.pk3 is quite big. Due to the
overhead imposed by the wireless interface card, only 57.43% of
energy is saved. As the card is switched on/off often, the heat in the
hardware increases, resulting in slow response. It takes about 50-
200 milliseconds to wake-up from sleep mode. This is higher than
the normal 10-50 milliseconds lag. During these mode transition
(sleep-to-wakeup mode) period it is not possible to save energy.
We used off-the-shelf WiFi hardware in our experiments. With
better WiFi cards (with 10-50 milliseconds ON/OFF lag) , we have
estimated that it is possible to achieve close to 80% power saving
for 2 players and up to 50% for 8 players.
6.4 Optimisation of Fixed Sleep Time
The Error Controller is supposed to continuously adjust the Fixed
Sleep Time (FST) and Movement Speed (MS) to optimal values to
save maximum energy while meeting the quality requirement (Er-
ror Rate). The graph in Figure 4a shows how the FST is adjusted to
optimal values over time when playing with 8 players and 5% ET.
(This data is from the same game play used to plot Figure 3c for 8
players). Again , from the same game play data, we have also plot-
ted the graph in Figure 4b which shows the corresponding Average
Error Rate (AER) over the time.
From the Figure 4a we can see, the FST increases linearly for
the first 20 samples (at every tenth frame we took a sample). Then
it jumps down from about 550ms to about 300ms. This is because
the AER at this instant is higher than the threshold 5%; the FST
needs to be shortened in order to keep the AER within the ET. We
shorten FST to 75% of its current value. The AIMD method allows
AER to reduce quickly. The flat portion of the graph means the
AER is close to the ET. At this moment, FST stays relatively sta-
ble. Whenever another error occurs, the FST is reduced again until
the AER is lower than the ET. The graph in Figure 4b shows the
effect of this adjustment. The AER starts from 0. It then increases
until slightly higher than 5%, where FST is forced to be multiplica-
tively reduced. The AER is well controlled within the 5% threshold
which proves our AIMD method is effective.
The graphs in Figure 4c and Figure 4d show the convergence of
AER to ET level for game plays with 3% and 1% ET respectively.
The average convergence time is less than 6 seconds. This can be
shortened further by optimising the sample size and using weighted
average for AER.
6.5 Energy Overheads
There was a linear increase in processor load and memory re-
quirement with number of clients in power save mode. However,
with up to 8 players in power save mode, there was less than 3%
increase in processor load and memory. It resulted in 4% additional
energy consumption. This additional energy consumption is negli-
gible as the processor and memory consume only 4-15% of overall
system energy whereas wireless interface in mobile devices con-
sume up to 40% of system energy. We obtained such efficiency by
relying on the existing PVS data rather than recomputing.
6.6 Small Scale User Study
After training, the users played a set of six randomly chosen
games from eight versions as described in Section 5.2. They were
not informed about the meaning of each version. For each game
play, we asked them to rate how noticeable, if at all, were any net-
work related artefacts in the game, compared to the unmodified
version, on a 5-point Likert scale. Figure 5 shows the user study
results. The results show that the artefacts are almost unnotice-
able up to ET=3% and ET=3% is good enough to save significant
amount of energy as shown in Figure 3c.
4.8
4.6
4.2
3.7
3.6
3.2
3.1
1
2
3
4
5
1%
2%
3%
4%
5%
6%
7%
Unno$ceable
Barely
No$ceable
No$ceable
Highly
No$ceable
Error
Threshold
Player’s
Score
(Average)
No
Difference
Figure 5: User Study - ET Varies (1% - 7%)
6. 0
100
200
300
400
500
600
700
800
900
1 11 21 31 41 51 61 71 81 91
Samples
(Frames)
over
the
game
play
Time
Fixed
Sleep
Time
ET
=
5%
8
players
0%
1%
2%
3%
4%
5%
6%
7%
1 11 21 31 41 51 61 71 81 91
ET
=
5%
8
players
Error
Rate
Samples
(Frames)
over
the
game
play
Time
0%
1%
2%
3%
4%
5%
6%
7%
Error
Rate
Samples
(Frames)
over
the
game
play
Time
ET
=
3%
8
players
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
Samples
(Frames)
over
the
game
play
Time
Error
Rate
ET
=
1%
8
players
FST over Error Threshold 5% Error Threshold 3% Error Threshold 1%
Game Play Time vs. Energy Saved vs. Energy Saved vs. Energy Saved
Figure 4: Effect of feedback based control on FST and Error Rate
Objective Study. Offline analysis of game play data revealed
that there was no significant difference in performance (number of
opponents killed/fragged) of the users between original and power
efficient versions. It varied only by 1.8%. Experts always killed
more opponents than normal users in both normal and power effi-
cient versions.
7. CONCLUSION AND FUTURE WORK
We have presented a set of techniques which used novel and scal-
able algorithms to save wireless client’s energy. The techniques can
also be applied to reduce game server’s bandwidth requirement. We
can save up to 57% of Wireless Interface Energy (which is, about
22.8% overall system energy) per mobile client without affecting
the quality of the game play adversely using off-the-shelf network
hardware. Our future work will include, extensive studies on differ-
ent feedback based controlling mechanisms, large scale user studies
and integration into commercial game engines.
8. REFERENCES
[1] Anand, B., Ananda, A. L., Chan, M. C., Long, L. T., and
Balan, R. K. Game action based power management for
multiplayer online games. Proceedings of the 1st ACM
SIGCOMM Workshop on Networking, Systems, and
Applications on Mobile Handhelds (MobiHeld), Barcelona,
Spain, Aug. 2009.
[2] Anand, B., Thirugnanam, K., Long, L. T., Pham, D. D.,
Ananda, A. L., Balan, R. K., and Chan, M. C. Arivu:
Power-aware middleware for multiplayer mobile games.
Proceedings of the Ninth IEEE NetGames, Teipei, Taiwan,
Nov. 2010.
[3] Anand, B., Thirugnanam, K., Sebastian, J., Kannan, P. G.,
Ananda, A. L., Chan, M. C., and Balan, R. K. Adaptive
display power management for mobile games. ACM
Mobisys, 2011.
[4] Anastasi, G., Passarella, A., Conti, M., Gregori, E., and
Pelusi, L. A power-aware multimedia streaming protocol for
mobile users. Proceedings of the International Conference
on Pervasive Services, Santorini, Greece, July 2005.
[5] Armitage, G. An experimental estimation of latency
sensitivity in multiplayer quake 3. Networks, 2003.
ICON2003. The 11th IEEE International Conference on,
pages 137 – 141, sept.-1 oct. 2003.
[6] Bertozzi, D., Benini, L., and Ricco, B. Power aware network
interface management for streaming multimedia.
Proceedings of the IEEE Wireless Communications and
Networking Conference, 2002.
[7] Bharambe, A., Douceur, J., Lorch, J. R., Moscibroda, T.,
Pang, J., Seshan, S., and Zhuang, X. Donnybrook: Enabling
large-scale, high-speed, peer-to-peer games. Proceedings of
ACM Conference on Applications, technologies,
architectures, and protocols for computer communications
(SIGCOMM), Seattle, WA, USA, Aug. 2008.
[8] Bhojan, A., Akhihebbal, A., Chan, M., and Balan, R. Arivu:
Making networked mobile games green. Mobile Networks
and Applications, pages 1–8, may 2011.
10.1007/s11036-011-0312-8.
[9] Bittner, J., Mattausch, O., Wonka, P., Havran, V., and
Wimmer, M. Adaptive global visibility sampling. ACM
SIGGRAPH 2009 papers, SIGGRAPH ’09, pages
94:1–94:10, New York, NY, USA, 2009. ACM.
[10] Boulanger, J.-S., Kienzle, J., and Verbrugge, C. Comparing
interest management algorithms for massively multiplayer
games. NetGames ’06: Proceedings of 5th ACM SIGCOMM
workshop on Network and system support for games, page 6,
New York, NY, USA, 2006. ACM.
[11] Google Open Source Project. A port of Quake3 to Android.
http://code.google.com/p/kwaak3/, 2010.
[12] Gu, Y. and Chakraborty, S. A Hybrid DVS Scheme for
Interactive 3D Games. IEEE Real-Time and Embedded
Technology and Applications Symposium, 2008.
[13] Gu, Y. and Chakraborty, S. Power Management of Interactive
3D Games Using Frame Structures. VLSI Design, 2008.
[14] Id Software. Quake 3 Arena Source Code.
http://ioquake3.org/, July 2010. (Version 3.21).
[15] Kazem, I., Ahmed, D. T., and Shirmohammadi, S. A
visibility-driven approach to managing interest in distributed
simulations with dynamic load balancing. DS-RT, 2007.
[16] MadWifi Organisation. MAdWiFi Wireless Driver for Linux,
Release v0.9.4, Feb. 2008.
http://madwifi-project.org/.
[17] Meyer, M., Sachs, J., and Holzke, M. Performance
evaluation of a tcp proxy in wcdma networks. Proceedings of
the Eighth Annual ACM/IEEE International Conference on
Mobile Computing and Networking (MOBICOM), 2002.
[18] Teller, S. J. Visibility Computations in Densely Occluded
Polyhedral Environments. PhD thesis, Technical Report
UCB/CSD-92-708, EECS Department, University of
California, Berkeley, Oct 1992.
[19] Thirugnanam, K., Anand, B., Sebastian, J., Kannan, P. G.,
Ananda, A. L., Balan, R. K., and Chan, M. C. Dynamic
lookahead mechanism for conserving power in multi-player
mobile games. Proceedings of the 31st IEEE INFOCOM
2012, Orlando, USA, Mar. 2012.
[20] Ubuntu. Atheros Hardware Access Layer.
http://manpages.ubuntu.com/manpages/maverick/
en/man4/ath_hal.4freebsd.html/.
[21] Waveren, J. P. V. The quakeiii arena bot.
http://www.kbs.twi.tudelft.nl/docs/MSc/2001/Waveren_Jean-
Paul_van/thesis.pdf. University of Technology Delft, Faculty
of ITS, 2001.
[22] Wei, Y., Chandra, S., and Bhandarkar, S. A statistical
prediction-based scheme for energy-aware multimedia data
streaming. Proceedings of the Wireless Communications and
Networking Conference (WCNC), Atlanta, GA, Mar. 2004.
[23] Yang, S.-R. Dynamic power saving mechanism for 3g umts
system. ACM Mobile Networks and Applications,
12(1):5–14, 2007.