K.P. Subbalakshmi | Stevens Institute of Technology (original) (raw)

Papers by K.P. Subbalakshmi

Handbook of Cognitive Radio, 2019

We propose a distributed spectrum decision protocol resilient to primary user emulation attacks (... more We propose a distributed spectrum decision protocol resilient to primary user emulation attacks (PUEA) in dynamic spectrum access (DSA) networks. PUEA is a type of denialof-service attack that can result in unreliable and/or disconnected DSA networks by depriving legitimate secondary user s of spectrum access. We first propose an individual detection mechanism for secondary users to achieve preliminary sensi ng results. For this, we characterize the received power at a go od secondary user, using a flexible log-normal sum approximati on. We then develop a distributed spectrum decision protocol in which secondary users exchange individual sensing results with their one-hop neighbors to increase resilience to PUEA. We call this protocol NEAT: NEighbor AssisTed spectrum decisi on protocol. We provide mathematical analysis of this protocol in terms of both the probability of successful PUEA as well as the probability of missing the primary, under Byzantine attacks – when the malicious u...

The well documented growth in mobile traffic is mainly driven by increasingly sophisticated smart... more The well documented growth in mobile traffic is mainly driven by increasingly sophisticated smart phone applications. Simultaneously, user preference for lighter phones has resulted in more battery power constrained hand-helds that offload computations to resource intensive cloud. This second trend exacerbates the bandwidth crunch often experienced over wireless networks. Our idea (joint cognitive offloading and scheduling) is to use dynamic spectrum access and management concepts from wireless networking to effect computation offloading and scheduling solutions that achieves near optimal trade-offs between the mobile device and wireless resources. We use all radio available interfaces (e.g. WiFi, LTE) in multi-RAT enabled devices to schedule appropriate components of the application to run either on the mobile device or on the resource-rich cloud, while staying adaptive to the conditions of the wireless network.

The exponential growth of the Internet (WWW in particular) has opened-up several avenues for cove... more The exponential growth of the Internet (WWW in particular) has opened-up several avenues for covert channel communication. Steganographic communication is one such avenue. Hiding secret messages in digital data such as images using steganographic software tools is becoming easier. These digital images posted in public Web sites can then be downloaded at the receiver and the hidden messages may be extracted securely. To thwart covert channels on the Internet new types of search engines that can identify, detect and track these channels are necessary. Traditional search algorithms will fail to identify these channels. In this paper, we discuss various key issues involved in developing a stego (forensic) Web search engine. We also propose approaches to address some of these issues. Finally, we discuss a prototype forensic search engine that we developed called STEALTH and discuss in detail its architecture. Some experimental results are also reported.

IEEE Transactions on Cognitive Communications and Networking, 2015

Multimedia Forensics and Security

Proceedings of the 2011 SIAM International Conference on Data Mining, 2011

We study the dynamics of social networks in terms of population growth and control of user behavi... more We study the dynamics of social networks in terms of population growth and control of user behavior. Most of the current research in social networks focus on static analysis through graph theoretic models to represent the networks or focus on modeling the traffic. Here, we study the cost of collaborative vs individualistic behavior of users in order to grow their network size in a social network. Each user incurs a cost (monetary or emotional) for collaboration. We formulate the behavior of the users as a non-linear optimization problem with a cost. The objective function of the optimization problem is obtained using a stochastic analysis of population growth in social networks, based on the first-passage time of a birth-death process. The stochastic model is validated by comparison with real data obtained from Twitter Results indicate that a homogeneous social network (in which users have similar characteristics) will be individualistic. However, heterogeneous social networks (users with different characteristics) exhibit a threshold effect, i.e., there is a minimum cost, below which the network is as collaborative as desired and a maximum cost above which the network is individualistic as required. To the best of our knowledge, this is one of the first analysis of dynamics of user behavior and temporal population growth in social networks.

Lecture Notes in Computer Science, 2011

ABSTRACT

Proceedings of the 9th workshop on Multimedia & security - MM&Sec '07, 2007

This paper presents a novel steganalysis technique to attack quantization index modulation (QIM) ... more This paper presents a novel steganalysis technique to attack quantization index modulation (QIM) steganography. Our method is based on the observation that QIM embedding disturbs neighborhood correlation in the transform domain. We estimate the probability density function (pdf) of this statistical change in a systematic manner using a kernel density estimate (KDE) method. The estimated parametric density model is then used for stego message detection. The impact of the choice of kernels on the estimated density is investigated experimentally. Simulation results evaluated on a large dataset of 6000 quantized images indicate that the proposed method is reliable. The impact of the choice of message embedding parameters on the accuracy of the steganalysis detection is also evaluated. Simulation results show that the proposed method can distinguish between the quantized-cover and the QIM-stego with low false alarm rates (i.e. P f n ≤ 0.03 and P f p ≤ 0.19). We demonstrate that the proposed steganalysis scheme can successfully attack steganographic tools like Jsteg and JP Hide and Seek as well.

This paper combines data and mathematical mod-eling to present evidence for several dynamic hiera... more This paper combines data and mathematical mod-eling to present evidence for several dynamic hierarchical structures in Twitterverse. A model is presented to formalize the hierarchical connections between the number of tweets, population of cities, areas of cities, etc. The parameters of the model reveal several insights, such as, a topic has a broader appeal across several geographically distributed cities or has a limited influence concentrated in a few cities. Using a combination of spatial clustering and partitioning, the geographical circles of influences of a Twitter topic and the source cities that influence the topic are identified. A number of Twitter data examples are provided to validate the approach.

Machine Learning and Data Mining in Pattern Recognition, 2011

It is easy to hide the true identity of the author of an email. The author's actual name, email a... more It is easy to hide the true identity of the author of an email. The author's actual name, email address, etc. can be changed arbitrarily to deceive an email receiver. For example, a sender can change his/her identity in the email header to send different emails to various recipients. Therefore, in this paper, we investigate techniques for authorship similarity detection from the text content of a short length, topic-free email. 150 stylistic cues are identified for this problem. A frequent pattern and machine learning based method is proposed. Extensive experiment results are also presented for the Enron email data set.

EURASIP Journal on Information Security, 2007

Recommended by E. Magli We propose a joint encryption and compression (JEC) scheme with emphasis ... more Recommended by E. Magli We propose a joint encryption and compression (JEC) scheme with emphasis on application to video data. The proposed JEC scheme uses the philosophy of distributed source coding with side information to reduce the complexity of the compression process and at the same time uses cryptographic principles to ensure that security is built into the scheme. The joint distributed compression and encryption is achieved using a special class of codes called high-diffusion (HD) codes that were proposed recently in the context of joint error correction and encryption. By using the duality between channel codes and Slepian-Wolf coding, we construct a joint compression and encryption scheme that uses these codes in the diffusion layer. We adapt this cipher to MJPEG2000 with the inclusion of minimal amount of joint processing of video frames at the encoder.

2007 4th IEEE Consumer Communications and Networking Conference, 2007

Over the past few years there has been a growing demand for radio resources and at the same time ... more Over the past few years there has been a growing demand for radio resources and at the same time these resources are under utilized due to static spectrum allocation techniques. Dynamic spectrum access (DSA) has been thought of as a solution that would satisfy both the growing demand for radio resources and to efficiently utilize the spectrum. The radio devices that have the capability to dynamically sense the spectrum and access the under utilized bands are called cognitive radios (CR). There are two broad classes of users in CR, the primary user is a licensed user of a particular radio frequency band and the secondary users are unlicensed users who cognitively operate without causing harmful interference to the primary user. In this paper we consider a denial attack on centralized DSA networks where a malicious secondary user masquerades as a primary user and effectively shuts off access to all other secondary users. Note that this problem is unique to CR due to the distinction between primary and secondary users. We propose a public key cryptography based primary user identification mechanism that prevents malicious secondary users from masquerading as primary users. We show that the proposed identification mechanism and the associated key management are computationally light weight. We also discuss some advantages and limitations of the proposed identification mechanism.

2011 IEEE Global Telecommunications Conference - GLOBECOM 2011, 2011

In this paper, we investigate the parochialism in dynamic spectrum access networks and its effect... more In this paper, we investigate the parochialism in dynamic spectrum access networks and its effect on spectrum resource competition and self-coexistence among cognitive radio secondary users. We assume that some greedy secondary users in the network form a parochial community in private and try to maximize their own utilities without concern for the interests of other secondary users outside the community. A noncooperative game is formulated first to analyze the equilibrium strategy of secondary users without any parochial community in the network. In the scenario where there is a parochial community, we analyze the expected payoff for the insider secondary users who deviate from the Nash equilibrium and derive the optimal strategy for them. Through numerical analysis, we see how the optimal strategies for the secondary users inside the parochial community in different circumstances deviate from Nash equilibrium. Simulations results demonstrate that being a part of the parochial community, the secondary user can obtain more utility than the outsider secondary users and the system will change from an equilibrium to an unfair situation.

2010 IEEE International Conference on Communications, 2010

2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698), 2003

Multiple description coding (MDC) schemes have become popular as robustness tools in packet switc... more Multiple description coding (MDC) schemes have become popular as robustness tools in packet switched networks where multiple paths with varying packet loss rates are available. In this paper we propose a new MDC scheme for video using 3D wavelet decomposition. The coefficient stream obtained from the spatio-temporal decomposition of a group of frames (GOF) is divided into groups using a uniform, random selection scheme. The descriptions are then constructed by placing a subset of those groups in the primary and secondary (redundant) positions. The bit-allocation problem between the primary and secondary groups is formulated and solved using the generalized BFOS [4] algorithm. Experiments using revesible 5/3 wavelet for the spatial and Haar wavelet for the temporal decomposition, along with embedded block coding with optimal truncation for the different groups have shown significant performance for test video sequences. Our scheme performs up to 7dB better than a hierarchical selection scheme at the same source rate. We are currently working on comparing our scheme with FEC protected MDC scheme and we will include the results in the final version of the paper.

2010 IEEE Global Telecommunications Conference GLOBECOM 2010, 2010

Primary user emulation attack (PUEA) is a denial of service (DoS) attack unique to dynamic spectr... more Primary user emulation attack (PUEA) is a denial of service (DoS) attack unique to dynamic spectrum access (DSA) networks. While there have been studies in the literature to detect and mitigate PUEA, the impact of PUEA on the call blocking and call dropping probabilities of users in secondary networks has not been studied. We present the first analysis to study the impact of PUEA on the secondary users in terms of call blocking and call dropping. We propose a three dimensional continuous time Markov chain (3D-CTMC) to model the channel occupancy in DSA networks. We use the 3D-CTMC to determine the call blocking and call dropping probabilities for secondary users. We validate our analysis with simulations. Results indicate that while PUEA does not affect the call blocking probability, it can increase the call dropping probability of secondary users by more than one order of magnitude. We also evaluate the call dropping performance of some of our previously proposed protocols to mitigate PUEA. We show that our protocols can improve the call dropping performance by about 40% for high traffic loads of malicious users and can provide almost the same performance as that of a system with no PUEA, when the malicious user traffic is low.

Data Compression Conference

Most current distributed joint source channel coding (DJSCC) work focus on sources without memory... more Most current distributed joint source channel coding (DJSCC) work focus on sources without memory, like i.i.d sources. In this paper we consider a more general scenario where the sources are not only correlated with each other, but also have memory. In particular, we design a constrained distributed joint source-channel decoder (C-DJSCD) for the optimal variable length multiple access source code (MASC)

2013 IEEE Global Communications Conference (GLOBECOM), 2013

This paper considers multiple self-interested cognitive radios and multiple self-interested serve... more This paper considers multiple self-interested cognitive radios and multiple self-interested servers with multiple interfaces and solve the optimal flow management problem arising in this network. We formulate a novel flow management problem in which cognitive nodes compete to minimize the total delay over all interfaces while at the same time the servers compete to maximize their individual profit. We model this problem as a leader-follower game and propose a dynamic linear pricing scheme designed to achieve an optimal flow allocation. We propose an iterative algorithm to solve the game and analyze the criteria for convergence to the unique Nash Equilibrium. The messaging required to implement this algorithm is minimal thereby making it suitable for distributed implementations. Numerical simulations demonstrate significant improvement in terms of average total delay for cognitive nodes in comparison with alternative algorithms. Simulation results show that when quality of service in terms of average total delay is fixed, our algorithm improves the capacity of the network by 40% for the maximum allowable throughput demand.

Lecture Notes in Computer Science, 2006

In this paper we combine the error correction and encryption functionality into one block cipher,... more In this paper we combine the error correction and encryption functionality into one block cipher, which we call High Diffusion (HD) cipher. The error correcting property of this cipher is due to the novel error correction code which we call High Diffusion code used in its diffusion layer. Theoretical bounds on the performance of the HD cipher in terms of security and error correction are derived. We show that the proposed HD cipher provides security equivalent to Rijndael cipher against linear and differential cryptanalysis. Experiments based on a four round HD cipher reveal that traditional concatenated systems using the Rijndael cipher followed by Reed Solomon codes require 89% more expansion to match the performance of HD cipher.