Sumon Chowdhury

Bio: Sumon Chowdhury is an academic researcher from Tata Consultancy Services. The author has contributed to research in topics: Graph (abstract data type) & Connectivity. The author has an hindex of 2, co-authored 3 publications receiving 17 citations. Previous affiliations of Sumon Chowdhury include University of Calcutta.

Algorithms for generating all possible spanning trees of a simple undirected connected graph: an extensive review

Abstract: Generation of all possible spanning trees of a graph is a major area of research in graph theory as the number of spanning trees of a graph increases exponentially with graph size. Several algorithms of varying efficiency have been developed since early 1960s by researchers around the globe. This article is an exhaustive literature survey on these algorithms, assuming the input to be a simple undirected connected graph of finite order, and contains detailed analysis and comparisons in both theoretical and experimental behavior of these algorithms.

Generation of Simple, Connected, Non-isomorphic Random Graphs

Abstract: In graph theory, generation of random graphs finds a wide range of applications in different scheduling problems, approximation algorithms, problems involving modeling and simulation, different database applications, and obviously to test the performance of any algorithm. The algorithm, which has been devised in this paper, is mainly for the purpose of providing test bed for checking performance of other algorithms. It generates different non-isomorphic graph instances of a given order and having unique number of edges. The number of such instances possible for a graph of given order has also been subsequently formulated. Different such graph instances of different orders, generated in a uniform computing environment, and the computing time required for such generations have also been included in this paper. The simplicity and efficiency of the algorithm, subsequently proved in the paper, give us a new insight in the area of random graph generation and have called for further research scope in the domain.

Two Algorithms for Computing All Spanning Trees of a Simple, Undirected, and Connected Graph: Once Assuming a Complete Graph

Abstract: In this paper, we have proposed altogether different and new approaches for the computation of all possible spanning trees of a simple, undirected, and connected graph. Our proposed algorithms have the capability to solve the major bottlenecks in this area, namely, generation of duplicate trees and circuit checking. In the first algorithm, the given graph has been converted to its corresponding weighted complete graph, which proposes to have novelties and limitations of its own. In addition, we have also proposed another related algorithm, and as a result, we have been able to come up with new ideas in this research domain of graph theory.

Graph theory with applications to engineering and computer science, by Narsingh Deo. Pp xvii, 478. 1974. SBN 0 13 363473 6 (Prentice-Hall)

Abstract: Graph Theory with Applications to Engineering and Computer ... This outstanding introductory treatment of graph theory and its applications has had a long life in the instruction of advanced undergraduates and graduate students in all areas that require knowledge of this subject. The first nine chapters constitute an excellent overall introduction, requiring only some knowledge of set theory and matrix algebra.

Improving the Software-Defined Wireless Sensor Networks Routing Performance Using Reinforcement Learning

Abstract: Software-defined networking (SDN) is an emerging architecture used in many applications because of its flexible architecture. It is expected to become an essential enabler for the Internet of Things (IoTs). It decouples the control plane from the data plane, and the controller manages the whole underlying network. SDN has been used in wireless sensor networks (WSNs) for routing. The SDN controller uses some algorithms to calculate the routing path; however, none of these algorithms have enough ability to obtain the optimized routing path. Therefore, reinforcement learning (RL) is a helpful technique to select the best routing path. In this article, we optimize the routing path of SDWSN through RL. A reward function is proposed that includes all required metrics regarding energy efficiency and network Quality-of-Service (QoS). The agent gets the reward and takes the next action based on the reward received, while the SDWSN controller improves the routing path based on the previous experience. However, the whole network is also controlled remotely through the Web. The performance of the RL-based SDWSN is compared with SDN-based techniques, including traditional SDN and energy-aware SDN (EASDN), QR-SDN, TIDE and non SDN-based techniques, such as $Q$ -learning and RL-based routing (RLBR). The proposed RL-based SDWSN outperforms in terms of lifetime from 8% to 33% and packet delivery ratio (PDR) from 2% to 24%. It is envisioned that this work will help the engineers for achieving the desired WSN performance through efficient routing.

Efficient Detection Scheme for Physical-Layer Network Coding in Multiway Relay Channels

Abstract: In this paper, we propose a novel scheme for physical-layer network coding (PNC) in multiway relay channels (MWRC) from the perspective of sequential multiuser detection. We consider an uplink MWRC scenario where $N$ users, each equipped with a single antenna, simultaneously transmit their signal to a relay equipped with $K$ antennas ( $K < N$ ). Extraction of the network codes from the superimposed user signals at the relay node is formulated as an under-determined linear system. To solve this problem with low decoding complexity, the proposed method combines successive interference cancellation (SIC) with Babai estimation for regularized integer least squares (ILS). Specifically, SIC decoding is first employed to detect a selected subset of stronger user signals and remove their interfering effects. Babai estimation is then applied to extract the remaining user signals, which is formulated as an ILS problem with reduced dimension. We develop a power allocation scheme to enhance the performance of both SIC and ILS steps, and discuss an optimal user pairing strategy based on the average decoding error probability. Numerical results demonstrate the performance improvement of the proposed method in extracting network codes from multiple superimposed user signals.

Cliques and a new measure of clustering: With application to U.S. domestic airlines

Abstract: We propose a higher-order generalization of the well-known overall clustering coefficient for triples C ( 3 ) to any number of nodes. We give analytic formulae for the special cases of three, four, and five nodes and show that they have very fast runtime performance for small graphs. We discuss some theoretical properties and limitations of the new measure, and use it to provide insight into dynamic changes in the structure of U.S. airline networks.

A Pivot Gray Code Listing for the Spanning Trees of the Fan Graph

Abstract: We use a greedy strategy to list the spanning trees of the fan graph, \(F_n\), such that successive trees differ by pivoting a single edge around a vertex. It is the first greedy algorithm for exhaustively generating spanning trees using such a minimal change operation. The resulting listing is then studied to find a recursive algorithm that produces the same listing in O(1)-amortized time using O(n) space. Additionally, we present O(n)-time algorithms for ranking and unranking the spanning trees for our listing; an improvement over the generic \(O(n^3)\)-time algorithm for ranking and unranking spanning trees of an arbitrary graph.