RAPTAP: a socio-inspired approach to resource allocation and interference management in dense small cells

Sathya, Vanlin and Kala, Srikant Manas and Bhupeshraj, S and Tamma, Bheemarjuna Reddy (2021) RAPTAP: a socio-inspired approach to resource allocation and interference management in dense small cells. Wireless Networks, 27 (1). pp. 441-464. ISSN 1022-0038

Full text not available from this repository. (Request a copy)


Femto Cells offer higher data rates to users within closed spaces. Dense deployment of small cells is a characteristic of pre-5G/LTE-Advanced Pro (LTE-A Pro) networks and is a precursor to the future 5G cellular networks. Combined with a Frequency Reuse Factor (FRF) of one, the dense small cell systems result in high co-tier interference which is undesirable. High interference Optimized scheduling decisions and interference management techniques are required to guarantee certain minimum data rates to the users at the cell-edge and increase the overall throughput of the system. This work presents a centralized scheduling approach that mitigates the detrimental impact of interference, thereby maximizing the overall throughput of the system. In doing so, we make use of concepts from Social theory and incorporate these ideas in the solution design. The centralized approach uses optimized scheduling algorithm (RAPTA) which takes feedback from the indoor users of all Femtos as the input and formulates a Mixed Integer Non-Linear Programming (MINLP) problem. Thereafter, the MINLP problem is relaxed and its solution carries out the Resource Block allocation and ensures optimal power transmission for every allocated resource block in all Femtos. We implement the proposed OPT algorithm on top of Proportional Fair (PF) in the Vienna Simulator. Since the RAPTA is NP-Hard and takes a considerably long time to solve the MINLP problem, we derive from it a polynomial time Heuristic algorithm (RAPTAP) which performs Resource Block allocation and sub-optimal Power transmission quite close to the RAPTA algorithm in terms of performance. RAPTAP is a two-stage approach each of which is inspired by ideas from two different strands of Sociological theory. We demonstrate through experiments that the proposed RAPTA + PF achieves 60.67% improvement in the services offered to the mobile users when compared to the classic PF algorithm with Soft Fractional Frequency Reuse (SFFR) interference management technique in the built environment. The Socio-inspired RAPTAP performs almost as well as the RAPTA + PF algorithm, with only a marginal 4% drop in the overall system throughput as compared to the latter. Further, we evaluate the RAPTAP heuristic in scenarios involving mobile users and demonstrate 14.52% improvement when compared to classic PF algorithm with Fractional Frequency Reuse-Full Isolation (FFR-FI) interference management. Finally, we compare the proposed Socio-inspired RAPTAP + PF with two state-of-the-art algorithms, viz., a Genetic Algorithm approach to resource allocation (NSGA-UDN) and a recent work on LTE-Unlicensed (LTE-U) + PF. Proposed Socio-inspired solution outperforms the LTE-U + PF and NSGA-UDN by 28.26% and 31.41%, respectively, in terms of throughput.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Tamma, Bheemarjuna ReddyUNSPECIFIED
Item Type: Article
Uncontrolled Keywords: 5G mobile communication systems; Cells; Cytology; Femtocell; Genetic algorithms; Heuristic algorithms; Industrial management; Integer programming; Mobile ad hoc networks; Mobile telecommunication systems; Nonlinear programming; NP-hard; Polynomial approximation; Power transmission; Resource allocation; Scheduling
Subjects: Computer science
Divisions: Department of Computer Science & Engineering
Depositing User: . LibTrainee 2021
Date Deposited: 20 Jul 2021 04:36
Last Modified: 20 Jul 2021 04:36
URI: http://raiith.iith.ac.in/id/eprint/8444
Publisher URL: http://doi.org/10.1007/s11276-020-02460-7
OA policy: https://v2.sherpa.ac.uk/id/publication/14920
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 8444 Statistics for this ePrint Item