Browsing by Author "Ardakani, M."

Now showing 1 - 4 of 4

Metadata only
Joint lifetime-outage optimization in relay-enabled IoT networks—A deep reinforcement learning approach
(IEEE, 2023-01) Heidarpour, A. R.; Heidarpour, M. R.; Ardakani, M.; Tellambura, C.; Uysal, Murat; Electrical & Electronics Engineering; UYSAL, Murat
Network lifetime maximization in Internet of things (IoT) is of paramount importance to ensure uninterrupted data transmission and reduce the frequency of battery replacement. This letter deals with the joint lifetime-outage optimization in relay-enabled IoT networks employing a multiple relay selection (MRS) scheme. The considered MRS problem is essentially a general nonlinear 0-1 programming which is NP-hard. In this work, we use the application of the double deep Q network (DDQN) algorithm to solve the MRS problem. Our results reveal that the proposed DDQN-MRS scheme can achieve superior performance than the benchmark MRS schemes.
Metadata only
Network-coded cooperative systems in cognitive radio networks
(IEEE, 2022-12) Heidarpour, A. R.; Ardakani, M.; Tellambura, C.; Uysal, Murat; Electrical & Electronics Engineering; UYSAL, Murat
We study the performance of a network-coded cooperative (NCC) system in an underlay cognitive radio network (CRN). The primary network (PN) consists of a single transmitter-receiver pair, and the secondary network (SN) is an NCC system with $N$ users, $M$ relays, and a single destination. The relays employ decode-and-forward (DF) protocol and use network coding (NC). We study the performance of the SN under two types of power constraints: i) the combined peak interference power constraint on the PN and maximum transmit power constraint at the SN; and ii) the single peak interference power constraint on the PN. For the SN, an exact closed-form expression and an asymptotically tight end-to-end outage probability are derived, and the diversity order and coding gain are quantified. Compared to the existing literature, the proposed CRN NCC has four main distinguishable features: i) it applies to general CRN NCC network settings with an arbitrary number of users and relays; ii) it considers general relay selection mechanism and independent and non-identically distributed (i.n.i.d.) $Nakagami-m$ fading channels; iii) it assumes secondary-to-primary and primary-to-secondary interference links; and iv) it provides a generalization of previous work and includes existing results in the literature as special cases.
Metadata only
Network-coded cooperative systems with generalized user-relay selection
(IEEE, 2020-11) Heidarpour, A. R.; Ardakani, M.; Tellambura, C.; Di Renzo, M.; Uysal, Murat; Electrical & Electronics Engineering; UYSAL, Murat
We consider a network-coded cooperative (NCC) system that consists of N >= 2 sources, M >= 1 decodeand-forward (DF) relays, and a single destination. The relays perform network coding (NC) on the received sources' symbols using maximum distance separable (MDS) codes. For this system, we propose the most generalized user-relay selection (GURS) scheme in the literature that selects any arbitrary subsets of K users and any arbitrary subsets of L relays subject to practical constraints such as load balancing conditions and scheduling policy. Our analytical results and design guidelines generalize and subsume all existing results as special cases. To this end, we derive a new closed-form outage probability (OP) expression, assuming non-identically and independently distributed (n.i.i.d.) Rayleigh fading channels. The asymptotic outage expression at high signal-to-noise ratio (SNR) regime is further derived, based on which, the achievable diversity order and coding gain are quantified. The theoretical derivations are also validated through Monte-Carlo simulation.
Metadata only
Soft actor-critic-based computation offloading in multiuser MEC-enabled IoT-A lifetime maximization perspective
(IEEE, 2023-10-15) Heidarpour, A. R.; Heidarpour, M. R.; Ardakani, M.; Tellambura, C.; Uysal, Murat; Electrical & Electronics Engineering; UYSAL, Murat
This article studies the network lifetime optimization problem in a multiuser mobile-edge computing (MEC)-enabled Internet of Things (IoT) system comprising an access point (AP), a MEC server, and a set of K mobile devices (MDs) with limited battery capacity. Considering the residual battery energy at the MDs, stochastic task arrivals, and time-varying wireless fading channels, a soft actor-critic (SAC)-based deep reinforcement learning (DRL) lifetime maximization, called DeepLM, is proposed to jointly optimize the task splitting ratio, the local CPU-cycle frequencies at the MDs, the bandwidth allocation, and the CPU-cycle frequency allocation at the MEC server subject to the task queuing backlogs constraint, the bandwidth constraint, and maximum CPU-cycle frequency constraints at the MDs and the MEC server. Our results reveal that DeepLM enjoys a fast convergence rate and a small oscillation amplitude. We also compare the performance of DeepLM with three benchmark offloading schemes, namely, fully edge computing (FEC), fully local computing (FLC), and random computation offloading (RCO). DeepLM increases the network lifetime by 496% and 229% compared to the FLC and RCO schemes. Interestingly, it achieves such a colossal lifetime improvement when its nonbacklog probability is 0.99, while that of FEC, FLC, and RCO is 0.69, 0.53, and 0.25, respectively, showing a significant performance gain of 30%, 46%, and 74%.