Shankhanaad Mallick

Robust Resource Optimization for Cooperative Cognitive Radio Networks with Imperfect CSI

We develop robust resource-allocation schemes for a cognitive radio network (CRN), where the secondary users (SUs) try to communicate with each other from different small cell primary user (PU) networks. User cooperation technique is considered for communication among the SUs since PUs are in close proximity and there are tight interference constraints on the PU bands. Power allocation and relay selection schemes are optimized with the provision of quality of service to each SU considering different channel uncertainty models. We incorporate the channel outage events that have resulted from the imperfect channel state information under slow-fading channels in our resource optimization algorithms. We maximize the system goodput of the CRN while satisfying the interference constraints of the PU bands both probabilistically and for the worst case scenario. The original probabilistic optimization problem is approximated and transformed into a convex deterministic form, and a closed-form analytical solution for power allocation is derived. The closed-form power allocation solution helps us to develop an efficient relay selection scheme based on Hungarian algorithm. Simulation results reveal the effectiveness of our proposed schemes and the implications of ignoring the imperfectness among different channels when developing resource-allocation algorithms for CRNs.

Joint Relay Selection and Power Allocation for Decode-and-Forward Cellular Relay Network with Channel Uncertainty

In this paper, we propose joint relay selection and power allocation algorithms that can work robustly under imperfect channel knowledge for a decode-and-forward (DF) cellular relay network. The objective is to minimize the uplink transmit power of the network taking each users target data rate as the quality of service (QoS) constraint in the presence of imperfect channel state information (CSI). We consider the worst-case optimization approach, in which QoS constraint is satisfied for all users assuming both probabilistic and deterministic channel estimation error models. In this optimization framework, equivalent convex formulations are derived for the nonlinear optimization problems that are often combinatorially hard to solve in their original forms. After relay selection, efficient centralized as well as distributed power allocation algorithms scalable with respect to the size of the networks are developed. We also consider the case of power constrained networks where the objective is to provide QoS in the presence of limited power budgets on source and relays. The robust optimization problem is reformulated accordingly and efficient solution is provided. Numerical results show the effectiveness of the proposed algorithms and demonstrate the implications of ignoring channel estimation errors while developing relay selection and power allocation algorithms.

Resource Allocation for Selective Relaying Based Cellular Wireless System with Imperfect CSI

In this paper, we propose robust power allocation and admission control schemes for providing probabilistically constrained quality of service (QoS) in selective relaying based decode-and-forward (DF) cooperative cellular systems. The proposed schemes are robust against imperfect channel state information (CSI) in slow fading while optimizing the total uplink transmit power in these cooperative wireless networks. At first, we derive novel closed-form solutions for the optimization problem, where the objective is to minimize total uplink transmit power while meeting the probabilistic QoS guarantees for a given number of admitted users. This is achieved by approximating the probabilistic optimization problem into a convex deterministic form and then by deriving closed form analytical solutions for power allocation using Karush-Kuhn-Tucker (KKT) conditions. The closed-form property of these solutions allows us later to develop a very low-complexity suboptimal algorithm for joint admission control and power allocation in presence of imperfect CSI and selective relaying. We also conduct comprehensive simulation experiments to demonstrate the effectiveness of our proposed schemes and to highlight the benefits gained from considering channel estimation errors in resource allocation for cooperative cellular systems.

Robust power allocation designs for cognitive radio networks with cooperative relays

In this paper, we develop robust power allocation schemes for cognitive radio networks (CRNs) that can operate in multiple bands with cooperative relays considering uncertainty among the channels of secondary user (SU) network and among the channels of SU transmitters to primary user (PU) receivers. Our objective is to formulate the robust design optimization problems taking into account the interference threshold in the PU band specified by the regulatory guidelines. To optimally allocate power with channel uncertainty, two robust algorithms are developed: (i) the worst-case optimization, where the interference constraints are satisfied for all channels contained in some bounded uncertainty regions, and (ii) the probabilistically constrained optimization, where interference constraints are satisfied with certain probabilities. We show that the formulated problems are convex, which can be efficiently solved. Numerical results show the effectiveness of the proposed schemes and the implications of ignoring the uncertainties among different channels when designing power allocation schemes for CRNs.

Power Allocation for Decode-and-Forward Cellular Relay Network with Channel Uncertainty

In this paper, we propose centralized and distributed power allocation algorithms for a multi-user, multi relay cellular network using decode-and-forward (DF) cooperation strategy taking channel uncertainty into account. The objective is to minimize the total uplink power of the network taking each users target data rate as the quality of service (QoS) constraint under imperfect channel state information (CSI). We consider the worst-case optimization approach, in which QoS constraint is satisfied for all channels contained in some uncertainty region. First, a centralized power allocation scheme is developed to optimally allocate the power among the users and the relay nodes. Then, a suboptimal distributed algorithm is proposed based on a standard primal decomposition approach, where each relay can independently and separately minimize its own power. The proposed solutions are based on second order cone programming (SOCP), which is computationally efficient. Simulation results show that the performance of the suboptimal distributed solution is near-optimal and reveals the fact that DF cooperation strategy is more efficient in total power reduction and more robust under channel uncertainty over non-cooperative systems.

Joint resource optimization for OFDMA cellular networks with user cooperation and QoS provisioning

In this paper, a joint resource optimization scheme is designed for orthogonal frequency division multiple access (OFDMA) cellular wireless networks with multi-user cooperation. Joint relay selection, subcarrier allocation and pairing and power allocation algorithms are developed with the objective of maximizing the total capacity of the system considering the quality of service (QoS) requirements of the users. The optimization problem is a mixed integer nonlinear program (MINLP), which is often very difficult to solve in its original form. We provide a novel optimization framework to solve such non-linear optimization problems. The joint relay selection and subcarrier allocation problem is modified to a linear assignment problem and an efficient algorithm is developed to obtain the optimal assignment solution based on the Hungarian method. We propose computationally efficient solution to the joint resource optimization problem via dual decomposition method. Numerical results demonstrate the effectiveness of our proposed scheme.

QoS Provisioning Based Resource Allocation for Energy Harvesting Systems

In this paper, we propose quality-of-service (QoS) based resource allocation (RA) schemes for energy harvesting (EH) systems. We consider a system model with a single source and multiple destination nodes or users, in which the source node harvests energy from the environment. Our goal is to develop efficient RA policies for EH systems when the harvested energy at the source node is uncertain and insufficient to satisfy the QoS of the users completely. We develop two different schemes and RA policies to address this problem. In the first scheme, we minimize the total dissatisfaction of the users over a finite period of time through goal programming approach. In the second scheme, we maximize the number of admitted users and provide guaranteed QoS to them. For both schemes, we first develop offline algorithms assuming the availability of perfect and complete information about the harvested energy and channel state information (CSI). Next, we devise online algorithms based on dynamic programming (DP) considering the availability of causal information about the harvested energy and CSI. Numerical results demonstrate the effectiveness of our proposed algorithms and the importance of QoS provisioning based RA schemes in EH systems.

Joint Relay Selection and Power Allocation for Decode-and-Forward Cellular Relay Network with Imperfect CSI

In this paper, we propose a joint relay selection and power allocation algorithm for a multi-user, multi relay cellular network using decode-and-forward cooperation strategy with imperfect knowledge of channels. The objective is to minimize the total uplink power of the network taking each users target data rate as the quality of service constraint under imperfect channel state information (CSI), assuming some statistical knowledge is known about the channel estimation error. Joint relay selection and power allocation is a mixed integer programming problem which is combinatorially hard. For that an efficient sub-optimal solution with low complexity using convex relaxation approach is proposed. After relay selection, a centralized power allocation scheme is developed to optimally allocate the power among the users and the selected relay nodes. Then a low-complexity distributed algorithm is proposed based on a standard primal decomposition approach, where each selected relay can independently minimize its own power. Numerical results demonstrate the effectiveness of the proposed algorithms.

Resource allocation with QoS provisioning for energy harvesting systems: A goal programming approach

In this paper, we address the problem of resource allocation with quality-of-service (QoS) provisioning for an energy harvesting system with a single source node and multiple users (destination nodes). We consider that the source node harvests energy from the environment, which is insufficient to satisfy the QoS of the users due to its random nature. We develop resource allocation schemes to minimize the dissatisfaction of the users through goal programming approach. First, we develop an offline scheme assuming that complete information about the harvested energy and channel state information (CSI) is available. The optimization problem is formulated as convex and efficient solution is provided based on dual decomposition algorithm. Next, we devise an online scheme considering the availability of causal and incomplete information about the harvested energy and CSI. We propose an algorithm based on dynamic programming to obtain the online optimal solution. Numerical results demonstrate the effectiveness of our proposed schemes in terms of fairness and the importance of QoS provisioning in energy harvesting systems.

Distributed subcarrier pairing and relay selection for OFDM based cooperative relay networks

In this paper, we propose distributed subcarrier pairing and relay selection for OFDM based cooperative systems. Our objective is to maximize the throughput of the network using minimal signaling overhead when uniform power allocation is performed at the subcarriers. Since the optimal centralized solution for the formulated problem requires high signaling overhead and processing complexity, we propose two suboptimal distributed algorithms scalable and suitable for practical implementation. The first proposed algorithm is based on Hungarian method and is efficient in terms of signaling overhead reduction. The performance of this algorithm is extremely close to the optimal solution for large number of subcarriers. The second proposed distributed algorithm has lower complexity and it significantly reduces the signaling overhead at the expense of a small degradation in performance. Numerical results reveal the effectiveness of the proposed algorithms and demonstrate the superiority over other existing algorithms in the literature.