Javad Hajipour

Javad Hajipour

Auto-generated profile, based on publicly available information

Javad Hajipour

Department of Computer Science, University of Tabriz, Tabriz, Iran

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
18 November 2024 in IEEE Communications Letters
Verified
Journal Article
Closed access
Amin MohajerAmin Mohajer
Amin Mohajer
JHJavad Hajipour
Javad Hajipour
Victor C. M. LeungVictor C. M. Leung
Victor C. M. Leung
Network slicing and computation offloading play a pivotal role in enabling edge service providers to handle dynamic service demands effectively. However, traffic fluctuations and resource diversity pose significant challenges, often constrained by static configurations lacking flexibility. To overcome these limitations, this paper presents FlexSlice, a dynamic offloading framework designed to optimize resource allocation in mobile edge networks. Our approach leverages a sparse multi-head graph attention mechanism for precise traffic prediction, capturing complex spatio-temporal dependencies to enhance network slicing decisions. Additionally, we present an adaptive offloading strategy based on the twin delayed deep deterministic policy gradient algorithm, which incorporates twin critics and prioritized experience replay to improve decision-making under dynamic conditions. Simulation results confirm FlexSlice’s outstanding performance and adaptability in diverse operational scenarios, achieving higher profits and reliable quality of service.

21 July 2017 in IEEE Transactions on Wireless Communications
Verified
Journal Article
Closed access
JHJavad Hajipour
Javad Hajipour
JNJavad Musevi Niya
Javad Musevi Niya
DNDerrick Wing Kwan Ng
Derrick Wing Kwan Ng
In this paper, we study energy-efficient resource allocation in the downlink of buffer-aided wireless relay networks. We aim at maximizing the system average energy efficiency while maintaining the queue stability at both the base station (BS) and relays. We formulate the resource allocation design as a novel stochastic network optimization problem and based on the well-known Lyapunov drift-plus-penalty policy and the system constraints, we transform it to an instantaneous non-convex optimization problem to be solved in each time slot. We analyze the instantaneous utility function and propose a novel algorithm to find its optimum point. Based on that, we present an effective distributed strategy to get the globally optimal solution for channel and power allocation. Furthermore, we show that the proposed algorithm can be used as a building block for energy-efficient resource allocation in conventional relay networks, where the relays do not have buffering capability, but the BS queues need to be stabilized. Using extensive simulations, we show that the proposed algorithm is able to provide higher energy efficiency compared with the existing algorithms, while keeping the system queues stable.

27 January 2020 in IEEE Communications Letters
Verified
Journal Article
Closed access
JHJavad Hajipour
Javad Hajipour

15 December 2015 in EURASIP Journal on Wireless Communications and Networking
Verified
Journal Article
Open access
PDF file
JHJavad Hajipour
Javad Hajipour
RRRukhsana Ruby
Rukhsana Ruby
Amr MohamedAmr Mohamed
Amr Mohamed
VLVictor C. M. Leung
Victor C. M. Leung
Buffer-aided relaying has recently attracted a lot of attention due to the improvement in the system throughput. However, a side effect usually deemed is that buffering at relay nodes results in the increase of packet delays. In this paper, we study the effect of buffering at relays on the end-to-end delay of users’ data, from the time they arrive at the source until delivery to the destination. We use simple discussions to provide an insight on the overall waiting time of the packets in the system, taking into account the queue dynamics both in the source and relay. We analyze the end-to-end delay in the relay networks with Bernoulli data arrivals and channel conditions and prove that the data packets experience lower average end-to-end delay in the buffer-aided relaying system compared with the conventional one. Moreover, using intuitive generalizations, we conclude that the use of buffers at relays improves not only throughput but ironically the average end-to-end packet delay. Through extensive simulations, we validate our analytical results for the system when the data arrival and channel condition processes follow Bernoulli distribution. Furthermore, via the simulations under the settings of practical systems, we confirm our intuition for the general scenarios.

7 September 2016 in IEEE Communications Letters
Verified
Journal Article
Closed access
JHJavad Hajipour
Javad Hajipour
CLCyril Leung
Cyril Leung
JNJavad Musevi Niya
Javad Musevi Niya
In this letter, we study relay selection strategies for file transfer in a two-hop buffer-aided relay network when the source and relay nodes have limited energy. We consider the fact that the constrained energies of the relays limit their transmission capabilities. Using the concept of a token as an approximate sign of the possibility for packet reception and forwarding by a relay, we propose a context (specifically, channel, energy, and queue state)-aware relay selection strategy, which dynamically defines a scheduling rule at each time slot, based on the channel states, residual energies, and queue sizes. At each time slot, the scheduling rule specifies which of the source or relay nodes should use the channel for packet transmission. Numerical results confirm that the proposed strategy generally provides a higher number of packet transfers and a lower file transfer time compared with existing algorithms.

3 May 2022 in IEEE Wireless Communications Letters
Verified
Journal Article
Closed access
JHJavad Hajipour
Javad Hajipour
This letter studies hierarchical Mobile Edge Computing (MEC) with multiple sources, multiple buffer-aided relays, and one Higher-level computing Node (HN). The relays may or may not have computation resources and therefore, they randomly decide to process the received tasks or to offload them to the HN. We consider the delay in the transmission and computation buffers, and propose a novel Fairness and Efficiency Cost Function (FECF) as the weighted sum of the maximum Average Response Time (ART) and the overall system ART. We formulate a novel problem to minimize the FECF while keeping the system queues stable, by jointly optimizing the offloading probabilities at the relays and the capacities of the virtual machines at the HN. We prove that the problem is a multi-convex optimization problem and propose an effective solution method by alternating optimization. Simulation results demonstrate the effectiveness of the different variants of the proposed scheme in the scenarios with server-less and server-empowered relays and provide important insights regarding the fairness and efficiency in terms of the ART.