Radwa Sultan

Dr. Radwa Sultan

Electrical and Computer Engineering Department, Manhattan College, New York City, NY 10471, USA

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 September 2020 in IEEE Wireless Communications Letters
Verified
Journal Article
Closed access
MEMohamed A. ElMossallamy
Mohamed A. ElMossallamy
HZHongliang Zhang
Hongliang Zhang
RSRadwa Sultan
Radwa Sultan
KSKarim Seddik
Karim Seddik
LSLingyang Song
Lingyang Song
ZHZhu Han
Zhu Han
We consider an uplink multi-user scenario and investigate the use of reconfigurable intelligent surfaces (RIS) to optimize spatial multiplexing performance when a linear receiver is used. We study two different formulations of the problem, namely maximizing the effective rank and maximizing the minimum singular value of the RIS-augmented channel. We employ gradient-based optimization to solve the two problems and compare the solutions in terms of the sum rate achievable when a linear receiver is used. Our results show that the proposed criteria can be used to optimize the RIS to obtain effective channels with favorable properties and drastically improve performance even when the propagation through the RIS contributes a small fraction of the received power.

8 June 2021 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
RSRadwa Sultan
Radwa Sultan
KSKarim Seddik
Karim Seddik
ZHZhu Han
Zhu Han
BABehnaam Aazhang
Behnaam Aazhang
In this paper, we study the effects of joint transmit and receive antennas’ selection on full-duplex (FD) multiple-input-multiple-output (MIMO) networks’ performance. The antennas’ selection problem is, in general, a combinatorial problem whose complexity grows exponentially with the number of antennas. To fully understand the effects of antennas’ selection, we study the sum rate maximization problem in a single-cell network with an FD-MIMO base-station (BS). First, we consider a system with a normal-scale MIMO full-duplex BS, i.e., a normal-scale FD MIMO system. The sum rate maximization problem is studied for two different scenarios; in the first scenario, we consider jointly optimizing the transmit and receive antennas’ selection with the precoder and the receiver weights. A Generalized Bender's Decomposition based algorithm is proposed to solve the mixed-integer nonlinear sum rate maximization problem. In the second scenario, we consider self-interference cancellation via zero-forcing (ZF) transmission. A heuristic algorithm is proposed to solve the sum rate maximization problem by optimizing the selection of the transmit antennas, receive antennas, and the receive antennas at which self-interference is nulled. Second, in a very-large scale, i.e., massive MIMO system, we derive lower bounds for the uplink and downlink rates with ZF receiver and precoder, respectively. The sum rate is maximized by jointly optimizing the transmit to receive antennas ratio and the ratio of the receive antennas at which self-interference is nulled. Finally, via numerical analysis, we evaluate the performance of the formulated sum rate maximization problems.

2 April 2024 in IEEE Systems Journal
Verified
Journal Article
Closed access
RSRadwa Sultan
Radwa Sultan
ASAhmed Shamseldeen
Ahmed Shamseldeen
We consider a single-cell full-duplex (FD) network. In FD transmission, the downlink rate is degraded by the cochannel interference (CCI) from the active uplink transmission. In this letter, we study how utilizing reconfigurable intelligent surfaces (RIS) can mitigate the effects of the CCI. In a multiple-input-multiple-output (MIMO) network, we consider two scenarios for CCI cancellation. In the first scenario, we assume that both the RIS-base station uplink and downlink direct links suffer from deep fading, i.e., they do not exist. In that case, the CCI minimization problem will be cast as an unconstrained manifold optimization problem. In the second scenario, we assume that the RIS-base station uplink and downlink direct links exist and the RIS controls all the network's links. In that case, the CCI minimization problem will be constrained by the uplink and the downlink channel gains, and accordingly, a penalty-based approach will be adopted to solve the CCI minimization problem. Our results show that utilizing the RIS can be very effective in minimizing the CCI and achieving a higher downlink rate than benchmark schemes.

9 January 2020 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
RSRadwa Sultan
Radwa Sultan
LSLingyang Song
Lingyang Song
KSKarim Seddik
Karim Seddik
ZHZhu Han
Zhu Han
Resource allocation problems in full-duplex orthogonal frequency division multiple access (FD-OFDMA) networks are challenging due to their combinatorial, non-convex nature. In this paper, user pairing, subcarrier and power allocation in a single cell FD-OFDMA network are considered. A joint optimization problem is formulated to maximize the network's sum rate while satisfying downlink (DL) and uplink (UL) transmission power' constraints. Due to the sheer complexity of the proposed formulation, mainly due to its combinatorial nature, an efficient, iterative two-step solution algorithm for the joint problem is proposed. In the first step, based on defining the DL user equipment (UE) signal to noise ratio (SNR) threshold, which is the least SNR that can be detected by the DL-UE, an algorithm is proposed for user pairing and subcarrier assignment. In the second step, the power allocation problem for the assigned users' pairs is formulated and solved using the Alternating Direction Method of Multipliers (ADMM) in the high signal-to-interference-noise ratio regime. Finally, numerical results are presented to validate the performance of the proposed algorithms. We show that the performance of our proposed computationally-efficient two-step algorithm is very close to the sum rate upper bound derived from solving the dual problem.