DOI:10.32604/iasc.2023.026523
| Intelligent Automation & Soft Computing DOI:10.32604/iasc.2023.026523 | |
| Article |
Base Station Energy Management in 5G Networks Using Wide Range Control Optimization
Department of Electronics and Communication Engineering, Sathyabama Institue of Science and Technology, Chennai, 600119, India
*Corresponding Author: J. Premalatha. Email: premalathajeyaraman@gmail.com
Received: 29 December 2021; Accepted: 13 March 2022
Abstract: The traffic activity of fifth generation (5G) networks demand for new energy management techniques that is dynamic deep and longer duration of sleep as compared to the fourth generation (4G) network technologies that demand always for varied control and data signalling based on control base station (CBS) and data base station (DBS). Hence, this paper discusses the energy management in wireless cellular networks using wide range of control for twice the reduction in energy conservation in non-standalone deployment of 5G network. As the new radio (NR) based 5G network is configured to transmit signal blocks for every 20 ms, the proposed algorithm implements withstanding capacity of on or off based energy switching, which in-turn operates in wide range control by carrying out reduced computational complexity. The proposed Wide range of control for base station in green cellular network using sleep mode for switch (WGCNS) algorithm toon and off the base station will work in heavy load with neighbouring base station. For reducing the overhead duration in air, heuristic versions of the algorithm are proposed at the base station. The algorithm operates based on the specification with suggested protocol-level to give best amount of energy savings. The proposed algorithm reduces 40% to 83% of residual energy based on the traffic pattern of the urban scenario.
Keywords: 5G base station; energy management; energy saving; traffic pattern; sleep mode
The demand for high data rate in 5G networks with the possibility of internet access and flexible applications endure value to the new proposal. Increased data rate in Gbps will produce more energy consumption and increased waste collection to mobile communications [1]. Hence the global carbon dioxide (CO2) of 2% emission becomes a big question as stated by Mingayin in [2] where the author discusses about the year 2007 which made environmental sustainability asset backed as “Green IT,” an big discussion which have been drew out till 2009 is significance because of climate changes. This made organizations involved in spending more cost or else money saving by avoiding excess energy spent for increased action repeatedly. During off peak time the base station used is not that much efficient [3]. Thus, the energy saving and reduction depends on the optimal resource spotted in state of art, environmental sustainability maintenance, enterprise operations, supply chain for reply and returns, products and services, resources utilization and life cycle of each stations used. Malmodin et al. [4] gave wide view on the above mentioned energy management issues and infer that CO2 emission is controlled as a factor of 0.2 and 0.4.The authors endeavour is to provide environmental impact of information Communication Technology (ICT) that focus in electricity use and greenhouse gas emission [5] that simplifies and digitize at all the levels.
With new wisdom and vision given by Fehske et al. [6] this article specially focus on the global carbon footprint of mobile communication systems rolled out for ecological and economic hassles. Measuring the survival of assessment models, the author predicts increased CO2 equivalent emissions handhold from the year 2007 to 2020. Thus, when rate of emission increases due to data transfer then raise of mobile traffic volume will be at post disbursal. On enhancing improved spectral efficiency, the system leveraging the three orders of magnitude will analyze more traffic with the same energy consumption.
Compared to prior work, ICT foot print archives less defects in this proposal even when the carbon footprint goes higher than earlier. Taking CO2 emission as exhaustion this work pioneered the design and implementation as easy for necessity. This proposal minimize the electricity bill (EB) paid in account of all action taken well prior [7]. As a concequence, in 2013 the EB may raise to $22 billion. Energy saving can be optimized by assuming that fraction of cells can be switched off [8]. This proposals primarily concentrate to diverse use of power consumed by base station which may consume high energy from 60–80% of the total energy in wide range of cellular networks.
The Author in [9] discusses about lowering the energy consumption of mobile radio systems in an economical way. The author mainly focuses in deployment of base stations based on numbers of micro sites streamlining to already existing macro sites. These systems aggressively pursue to manipulate output for full load conditions. The adverse impact of sleep mode for the base station to save energy, which makes possibility of active base station having low traffic loads and decreased noise decibels to get off permanently [10]. Total energy consumed by the network can be minimized by taking care of spectral efficiency [11]. It’s well known that day time traffic will be more compared to night time which gives same effects when weekdays and weekend traffics which are compared for an output. To make sure that all base station runs in good and effective way during night time and in weekends it’s a compulsion to make unnecessary base station to sleep for a while. It is also taken care of the point that whatever is the work load may be the power consumption consumed by the base station will be same which is clearly given by author [12,13]. This main disadvantage with prior introduction gives a way to this proposal, which in-turn makes wide range of control to base station a compulsion for making it to sleep mode.
The traffic is exemplary with least usage when it covers spacious area of base station which are used less. To make the proposed system more effective emphasis is made on questions like when and which base station is in sleep mode? And pointing the parameters for determining sleep decision for concerned base station?
This work concentrates toward the wide range of control to base station in Green Cellular Networking using Sleep mode for Switch (WGCNS) are as follows:
■ Pseudo Followed: Base station energy saving is a time consuming process which is concentrated here. To reduce overheads there is a need for a common controller. In the proposed work, the effects given to the network while making a base station to sleep mode is calculated. These parameters concentrate further more with internal and external deepening ownership of traffic and loads named as threshold respectively. Taking all the parameters in consideration the linear complexity is calculated which may arise due to wide range of control to base station during sleep condition. All these are completely modified for best result.
■ Practical Results Received: For making this base station sleep mode to work better three heuristic algorithms are used, which operates either partially or without feedback. Also the performance gap of regular used algorithm and optimal search algorithm which reduce consumed energy to 10% in a real traffic is calculated. Besides above said condition the other execution is also calacuted like: i) a crash avoidance based on cross checking whether two base station try to go in sleep mode at a time during controlling besides traffic maintenance ii) a time to measure is repeated to ensure sleep mode occurrence for particular base station because of higher workloads.
Saving energy in wireless network seems to be unpredicted as internal and external load introduction was given brief by later works as mentioned in the reference by author [14]. Energy savings can be achieved in similar area with two operators with energy aware cooperative management [15]. Relating to previous work only one state implementation of dynamic base station may be possible on calculating load but this gives un-assurance of result [16]. Also in [17] energy savings is carried out by switching off the base station in night time. This system is carried with an analytical model. But all prior works is carried out only in hexagonal and Manhattan model network. There also exists a computational and energy trade off between wireless and fixed networks [18]. This work is carried with a feature implementation having sleep mode possibility for dynamic works in base station. In [19,20] the authors stated that the energy saving is implemented as multi hop system. In [21] the author carried cell zooming activity which resulted in implementation as it is highly impossible for better result. In the proposed system mentioned with its protocol suites many general problems are detailed and resolved. Whereas in Section 3 sleep mode algorithm is used to show its better results. In Section 4 first-order analysis is computed whereas Section 5 gives the result for real traffic profiles. Finally, conclusion and next state of implementation is suggested. Even after enlarging the area coverage for working base station that moves to off condition using a power controller in previous works secured a 50% energy saving alone [22]. Implementation of energy saving methods for uncountable base station going with awake or to sleep will be a difficult problem and also, they enforce high computation complexity use on high signal overheads [23]. All previous works have discussed about green cellular operation issue which lead a way for better understanding of what’s happening till now and what can be modified from existing to mere future. One extension needs to consider is more practical model for base station which in turn depends on its total path usage, instead of fixed energy accumulation model used in utmost all previous work. As described by author in [24,25], a model that utilize both energy in demand and fixed energy accumulation model is designed. Considering the network to be heterogeneous, having varied configuration of base stations like macro, micro, femto and WiFi with application support, which may accumulate more transmission power combined with operational power and varied frequency bands [26].
The existing system works into downlink capacity and overhead resolving methods, in future same steps can be used along with uplink capacity calculation. Here the specified system suggest trying to work with both points together to make this system suitable for any protocols received on working with vital grounding. From most of the literature it is evident that at the most concentration is focused for a friendly, reliable and efficient way as to how energy can be saved in wireless netwok with growing demand [27]. Energy harvesting is concentrated along with quality off service (QoS) in 5G technology [28]. Also, uniform energy can be consumed in wirelees sensor network using mobile sinks which also provide load as balanced one [29]. In recent times research is going on, on how to reduce the cost of operating the base station. Now a days, base station are powered by sun and wind power rather than the conventional diesel generators [30].
The rest of the paper is structured as follows: The proposed technique and traffic pattern is detailed in Section 2. The sleep mode algorithm is given in Section 3. Result and discussion are detailed in Section 4 and Section 5 concludes the major findings of the paper.
3 Proposed Technique and Working Methods
■ Protocol Suites: This network model put a huge impact in dimensional area called downlink communication for primary stage in mobile internet i.e., connection between the base station and equipment.
■ TCH Model: Datagram switching is applied for simulation of traffic channel model (TCH). An assumption that the used equipment x is working in time t that range with independent Poisson distribution with mean arrival rate λ(j, t). The result may be an exponentially distributed random variable with mean
In Eq. (1) there may be some intervention to the traffic data arrived due to targeted QoS as displayed in Tab. 1.
■ Grounding Rule: Equipment or network located with x ∈ A is associated with served base station which gives the best signal strength,
In Eq. (2)
■ Channel Model: Service rate of equipment at location j from base station e2 at time t is calculated as
BW in Eq. (3) is the system bandwidth and SINRe2(j, t) is received indicator with obstruction. Noise ratio is given as SINR at location j from base station e2 at time t that is given by
where σ2 refers to the noise power in Eq. (4).
Loads Organization: System load refers to the time required to specify the covered transport volume. QOS is well maintained when user traffic load and service rates are starting steps giving way to end result. The system load of base station e2 at time t will inch their way upwards to total traffic load which is formulated in Eq. (5),
where E1e2 represents base station e2 these notations are given in Tab. 2.
The proclaimed strength in wide range of base station control algorithm will be minimizing the energy accumulated for cellular networks.
In Eq. (6) E2S Base station–energy in demand
e1–signal received for all base station during t time.
Thus, sleep mode time is calculated as follows to make it omnipotent during transmission.
Threshold pth( ≤ 1) is for balancing trade-offs between the system resolution/dependability and effective working is given in problem formulation. For low threshold the system work load will also be less. This silent waiting state with less call drop probability ensures base stations are robust to any traffic. High threshold value results in more energy saving, but with reduced performance. Tab. 1 discusses the notation used in proposed work [1].
E1 ⊂ R2 Region represents over a period of time, e2 ∈ E2 says the index of base station, E1e2 ⊂ E1 base station (e2) gives the Coverage, x ∈ E1 are Area in consistent space,
4 Proposed Sleep Mode Algorithm
Sleep mode algorithm looks forward as more important as the peak traffic volume that balance and reduce traffic load overheads, it seems to be extreme pleasure in formalizing the sleep mode switch during peak hours as displayed in Tab. 2. The working level of WGCNS is displayed in Fig. 1. This section briefs out the use of having sleep mode option for making impale base stations to be more active. Here the newly proposed idea of having sequential algorithm, called sleep mode, in which base stations get turned on/off one by one for ensuring the QoS.
Figure 1: Working level
On making a single base station to sleep will fairly increase overload to neighbouring system [29,30]. Considering this elevator situation of a sleep mode to the allotted base station and less pay for se2(x) to increased interval for equipment used along with base stations undergo with examination. Because of this activity the service rate se2(x) will be increasing. Let us take a consideration which station can be made to sleep first. Set of neighbouring to base station e2 is represented as Nt1, so that n Nt1 which provides the best signal strength to base stations located in x ∈ E1e2 as mentioned below in Eq. (9):
There may be interruption between base stations due to heavy traffic after getting base station e2 to sleep for a while. This e2 can go to sleep mode when it satisfies below condition:
where E1e2 → n is the coverage of equipment who supports working of base station n to base station e2 in Eq. (10) which might go to sleeping condition. The unchanged travelling attenuation pn is defined as the middle travelling attenuation for base station (n). The increased travelling attenuation is represented as pb→n which gives the foreign loads e2 to n base station. For example, in total of 4, either base station 2 or base station 3 must go to sleep mode which in turn enhance travelling attenuation of base station 1 to be in greater scope for threshold. Thus, the network decides by self to make base station 1 or base station 4 to go for sleep mode. As base station 1 accumulates more space for traffic maintenance compared to base station 4 then chooses 1 as finally sleeping. Thus, all these type of additional work loads are represented as pb→n addition to original. Hence WGCNS (1, 1):
In Eq. (11) Base stations n ∈ Nt1e2- maximum neighbouring
This in turn occupies less spare room for traffic management on high demand. For average system load of neighbouring base station1
4.2 Parameterisation of WGCNS Algorithm
On matching to equation criteria in Eq. (10) which depends on location of base stations and its neighbour localization of sleep mode going base station can be fixed. This is emphasised from system information like signal strength, system load shared among base stations and equipment’s at work. Thus, each and every base station long time go for a self-check whether it is to go for sleep mode or not as displayed in Fig. 2. The sleep mode protocol summary of WGCNS is depicted in Fig. 3. This completely avoid the concept of having centralized controller. This algorithm follows below three steps
■ Initialization: Periodical feedback information for received signal is considered mainly for initializing the base station to go in sleep conditions. When a base station goes to sleep mode then remaining users in coverage area will select the second-best base station by mentioning its identity back to equipment in use.
■ Decision Making: Network impact for users and neighbouring base stations are given more important than any other for deciding to go for sleep mode or not. If Ge2 < ρth, then go to sleep mode for Nt1e2. To avoid such conflict, first raise broadcasting request to switching-off request to switching-off signal (RTGS). Then after only switches to sleep mode on receiving clear to switching-off (CTGA) from all its neighbouring base station. Thus, an alert from all neighbours on confirmed such as confirmation of switching-off (CRGS) is received on final.
■ Finalized Result: Thus, base station e2 goes to sleep mode on receiving the signal CTGA from all neighbouring. Then the next second-best signal strength for a base station will be noted and reported as right now equipment, which looks similar like conventional hand-over where all base station makes a same decision of giving rights to second best signal received base station. This type of group hand-over is majorly concentrated for many network processes which are ongoing in research that may be used in mass transportation and this is established by controlling signal.
Figure 2: Sleep mode working protocol of WGCNS
Figure 3: Sleep mode protocol summary of WGCNS
4.2.2 Awake from Sleep Algorithm
To awake a base station from sleep mode all its process is a reverse step of above conditions. This made when system load reaches the same value as like base station be foregoing to sleep condition. But this cannot be implemented by itself due to lack of current system load, which in turn make necessity of having neighbouring base station reply a compulsory one. On what basis a node gets to sleep will be affected with neighbouring base station.
■ Initialization: Initial steps for sleep algorithm is processed in well co-ordnance with final steps to be received as result. After receiving CTGA from all neighbouring base station, e2 base station will definitely go to sleep mode. Then e2 records overall system load along with hand-over traffic from e2 to others.
Before making a base station to go for sleep mode the neighbouring base stations will share about the sleep mode status, i.e., RTGS which is same as like classical hidden terminal problem.
■ Decision Making: On making e2 to go for sleep mode and then when it reaches system load matching to it, its result will thus make its neighbouring to e2 go to sleep mode in turn making the prior to awake by making condition as switching-on (RTGA). Thus, last sleep node to be switched on first. This is enriched as follows
If pe2 > pe2b + ∈, then forward the request to awake neighbouring base station e2 will be in sleep mode only.
Hence the sleep condition follows three parts as mentioned below:
■ Finalized Result: On receiving RTGA the base station which has gone to sleep state will get back to awake. In same manner as in previous, the equipment will re-select the next base station as which has best signal strength. When the traffic pattern occupies equivalent space then automatically the sleep mode protocol will work in undo for above operation namely awake condition. The overheads raised due to feedback that overcome in following ways to avoid slowdown of transactions.
Revision in feedback received: Because of the concept adaptive modulation, equipment sends feedback for received signal strength from worked base station. After this the additional feedback like next choice of good and strong indicator associated base station identity were used to calculate the lay down for travelling attenuation which may rise some overheads. This can be reduce as mentioned below:
In Eq. (13) k–Represent the benefit for deployed base station and its neighbouring base station. On working with hexagonal cellular network this benefit is estimated as 1. Thus, sleep condition is redefined as follow
WGCNS(1, 0):
Balancing work load with neighbours: On exchange with feedback message the system work load may increase such that it will further reduce predicting system load of neighbouring. This type of execution can be done as follows:
Based on approximation Eq. (15) on homogeneous distribution this gets to hold when the user traffic path is changing continuously. Thus, the network impact is redefined as Eq. (16):
WGCNS (0, 1):
Adding Eqs. (12) and (14) here received a network effect similar to information without feedback like below Eq. (17):
WGCNS (0, 0):
Algorithm: Network Impact External Internal From Np
■ Other Issues: For making a node to go in sleep condition, exchange message like RTGS, CTGA and CRGS is necessary [1]. These message exchanges prevent multiple base stations switched off at the same time ensuring QoS. System goes inefficient during message exchange on same time. For example, E1 and E2 base stations send an RTGS to same neighbouring E3 simultaneously then, E3 responses CTGA to E2. But E2 doesn’t go to sleep mode by the other neighbouring base station which is not connected with E1. Hence E1 and E2 cannot go to sleep condition, hence the conclusion is made that the network is asynchronous. To avoid this overhead, the system need to go for synchronous operation that prevents collision of message exchange during RTGS in waiting or multi-step for CTGA. Because of high variation to system load repeated sleep condition working may happen. To avoid such things happening hysteresis margin is used. Then threshold is rewritten as follows:
When amount of energy saved decreases by the lower threshold as in Eq. (18) there may be reduced Sleep and Awake conditions that in-turn considers the trade-off made.
The process of calculating energy saved using above mentioned parameters may go challenging as base station deployment are dynamical during sleep condition declaration.
mean
Thus, energy saving ratio based on Eq. (19) is given by the equation,
where |E2| is the total number of base station t. Now consider that the average base sleep condition to account with its duration. Energy saving ratio is given in equation Eq. (18) say how long sleep condition works for a base station. Thus, it is represented as ton for e2 and toff for e2 for sleep time and awake time. Considering sinusoidal work in Eq. (20) for given traffic the result received is,
δ- Time gap with (1 − k/|Ne2|)
If there were more base station then this value will decrease. Hence, the peak traffic load is at time ton in Eq. (21), e2 will be in sleep mode. Thus, to ensure QOS the rules are followed as like below Eq. (22):
For cosine inverse, t on e2 is calculated as t on
where
Thus based on Eq. (23) energy saving ratio will be like
On redefining Eq. (24) the Taylor series will become
Eq. (25), gives the result for maximum energy saved even when traffic parameters are low values but the number of neighbouring base station is high. In urban commercial areas during night time that too in weekend the traffic will be less. Thus, δ and X are dynamical parameters while sleep or awake process.
On doing the stimulation with real 5G network topology holding 20 base stations in the area of 5 × 5 km, to avoid edge effects, Wrap-around technology is followed. A traffic load will be homogeneous but with varied traffic arrival rate. Increased traffic arrival rate, for base station with pth will make the system load = 1 with threshold value pth as 0. Making sleep condition to start with value 0.5 with system reliability for real travelling attenuation the resultant will be equal to 1 when traffic gets to peak. Considering the transmitted volume and service life of traffic structure for selected base station is given as Pi = 18W and EBS = 860W other parameters like channel propagation model are followed as per 3GPP release 15 in urban macro model. Fig. 4 depicts the traffic loads shared to neighbouring grounding during sleep or live mode for three 3 base station intially. Fig. 5 represents the graph between frequency and power spcrtral density of the system. Fig. 6 represents the energy in demand for varied traffic path having travelling attenuation ranging from 0 to 1. Here WGCNS (1, 1) with full feedback information consumes utmost 9% more energy for the entire system load. Thus, a distributed protocol with linear complexity is in result.
Figure 4: Traffic loads shared to neighbouring grounding during sleep or awake mode
Figure 5: Impact of traffic management and implementation of sleep algorithm
Figure 6: Energy saved ratio impact of traffic management and sleep algorithm
When feedback is like partial system load is smaller than the system functions too which is comparatively good, due to performance gap the system load will be 1. There is 3−6% of performance difference with WGCNS (0, 1), WGCNS (1, 0) and WGCNS (0, 0) as displayed in Fig. 6. Traffic benefit k is used in WGCNS (1, 0) and WGCNS (0, 0), which is 1 with our simulation to show strong signal during traffic load moving from base station to neighbouring. Thus 55% energy is saved in weekdays and 80% during weekend with performance gap less than 10%. On assuming the traffic loads will be 12% in peak and 35% in weekdays and 45% in weekends thus declared the above result. In Fig. 7 When Δh increases from 0.01 to 0.30 going for blocking of base station reaches sleep or awake point because ther is variation in system load based on time but may result in small loss of energy saving.
Figure 7: Balancing system overloads
Energy saving with working in sleep, considering the travelling attenuation that depends on the inner and outer work along with distance between base station and its neighbour are displayed in Fig. 8. When the system load is 0.04 the working base station is turned off one by one by the SLEEP algorithm. Average system load of base station going to sleep mode will have values less with neighbouring base stations. Thus, base station with low internal and external system load is getting to sleep first. Thus, proved was the internal factor of the sleep base station is having more network impact than the external factor of neighbouring base station with high system loads. When the density of active base station decreases then the distance increases between base stations. So, the average values of sleep station will be less compared to its neighbour with following interruption: making a base station to sleep-in high-density area will result in least collision occurrence for the network compared to base station with less density as displayed in Fig. 9.
Figure 8: Distance coverage of adjoining base stations
Figure 9: Number of switching off base station to total number of base station
5.1 Number of Switching off Base Station
Anyhow average number of neighbouring base stations remains same through sleep algorithm. Due to coverage increase with base station the density is reduced either way which may get reduced when there is more overheads and interference in the network.
In this system, the focus is for base station going to sleep to enforce traffic clearance and energy saving in 5G wireless cellular networks. The proposed algorithm using WGCNS provides more valuables for network-impact in all its sleep scheduling. Sleep mode for base station is considered for three constraints, namely, ease of implementation, computational complexity and signalling between the base stations, which deal in distributed way so that the network operator can apply this methodology in non-standalone implementation of 5G network. The main contribution is first-order analysis where energy in demand dependent on the transmission range mean (M) and variance (υ) for all the active base station. The proposed algorithm using WGCNS states results in energy conservation up to 80%.
Funding Statement: The authors received no specific funding for this study.
Conflicts of Interest: The authors declare that they have no conflicts of interest to report regarding the present study.
1. E. Oh, K. Son and B. Krishnamachari, “Dynamic base station switching-on/off strategies for green cellular networks,”IEEE Transactions on Wireless Communications, vol. 12, no. 5, pp. 2126–2136, 2013. [Google Scholar]
2. S. Mingay, “Green IT: TF new industry shock wave,”Gartner RAS Research Note G, vol. 153703, no. 7, pp. 1–7, 2007. [Google Scholar]
3. E. Oh and B. Krishnamachari, “Energy savings through dynamic base station switching in cellular wireless access networks,” in Proceedings of Global communications Conference, GLOBECOM, Miami, Florida, USA, pp. 1–5, 2010. [Google Scholar]
4. J. Malmodin, Å. Moberg, D. Lundén, G. Finnveden and N. Lövehagen, “Greenhouse gas emissions and operational electricity use in the ICT and entertainment & media sectors,” Journal of Industrial Ecology, vol. 14, pp. 770–790, 2010. [Google Scholar]
5. J. Lloret, S. Sendra and E. Macias-Lopez, “Advances in green communications and networking,” Mobile Networks and Applications, vol. 24, no. 2, pp. 653–656, 2019. [Google Scholar]
6. A. Fehske, G. Fettweis, J. Malmodin and G. Biczok, “The global footprint of mobile communications: The ecological and economic perspective,” IEEE Communications Magazine, vol. 49, no. 8, pp. 55–62, 2011. [Google Scholar]
7. K. Son, H. Kim, Y. Yi, B. Krishnamachari, J. Wu et al., “Toward energy-efficient operation of base stations in cellular wireless networks,” in Green Communications: Theoretical Fundamentals, Algorithms, and Applications, 1st ed., Boca Raton: CRC Press, pp. 435–474, 2012. [Google Scholar]
8. M. A. Marsan, L. Chiaraviglio, D. Ciullo and M. Meo, “Optimal energy savings in cellular access networks,” in Proc. IEEE Int. Conf. on Communications Workshops, Dresden, Germany, pp. 1–5, 2009. [Google Scholar]
9. A. J. Fehske, F. Richter and G. P. Fettweis, “Energy efficiency improvements through micro sites in cellular mobile radio networks,” in Proc. IEEE Globecom Workshops, Honolulu, Hawaii, USA, pp. 1–5, 2009. [Google Scholar]
10. P. Rost and G. Fettweis, “Green communications in cellular networks with fixed relay nodes,” in Cooperative Cellular Wireless Networks, Cambridge: Cambridge University Press, pp. 300–323, 2011. [Google Scholar]
11. K. Son, E. Oh and B. Krishnamachari, “Energy-aware hierarchical cell configuration: From deployment to operation,” in Proc. IEEE Conf. on Computer Communications Workshops (INFOCOMWKSHPS), Shanghai, China, pp. 289–294, 2011. [Google Scholar]
12. D. Willkomm, S. Machiraju, J. Bolot and A. Wolisz, “Primary users in cellular networks: A large-scale measurement study,” in Proc. 3rd IEEE Symposium on New Frontiers in Dynamic Spectrum Access Networks, Chicago, USA, pp. 1–11, 2008. [Google Scholar]
13. L. M. Correia, D. Zeller, O. Blume, D. Ferling, Y. Jading et al., “Challenges and enabling technologies for energy aware mobile radio networks,” IEEE Communications Magazine, vol. 48, no. 11, pp. 66–72, 2010. [Google Scholar]
14. L. Chiaraviglio, D. Ciullo, M. Meo, M. A. Marsan and I. Torino, “Energy-aware UMTS access networks,” in First International Workshop on Green Wireless (W-GREEN 2008), Lapland, Finland, 2008. [Google Scholar]
15. M. A. Marsan and M. Meo, “Energy efficient management of two cellular access networks,” ACM SIGMETRICS Performance Evaluation Review, vol. 37, no. 4, pp. 69–73, 2010. [Google Scholar]
16. J. Kwak, K. Son, Y. Yi and S. Chong, “Impact of spatio-temporal power sharing policies on cellular network greening,” in Proc. Int. Symposium of Modeling and Optimization of Mobile, Ad Hoc, and Wireless Networks, Princeton, New Jersey, USA, pp. 167–174, 2011. [Google Scholar]
17. E. Oh, B. Krishnamachari, X. Liu and Z. Niu, “Toward dynamic energy-efficient operation of cellular network infrastructure,” IEEE Communications Magazine, vol. 49, no. 6, pp. 56–61, 2011. [Google Scholar]
18. P. Rost and G. Fettweis, “On the transmission-computation-energy tradeoff in wireless and fixed networks,” in Proc. IEEE Globecom Workshop, Miami, Florida, USA, pp. 1394–1399, 2010. [Google Scholar]
19. W. Yang, L. Li, Y. Wang and W. Sun, “Energy-efficient transmission schemes in cooperative cellular systems,” in Proc. of IEEE GreenComm, Miami, Florida, USA, 2010. [Google Scholar]
20. F. Han, Z. Safar, W. S. Lin, Y. Chen and K. J. R. Liu, “Energy-efficient cellular network operation via base station cooperation,” in Proc. IEEE Int. Conf. on Communications (ICC), Ottawa, Canada, pp. 4374–4378, 2012. [Google Scholar]
21. D. Cao, S. Zhou, C. Zhang and Z. Niu, “Energy saving performance comparison of coordinated multi-point transmission and wireless relaying,” in Proc. IEEE Global Telecommunications Conf. GLOBECOM, Miami, Florida, USA, pp. 1–5, 2010. [Google Scholar]
22. Z. Niu, Y. Wu, J. Gong and Z. Yang, “Cell zooming for cost-efficient green cellular networks,” IEEE Communications Magazine, vol. 48, no. 11, pp. 74–79, 2010. [Google Scholar]
23. S. R. Danve et al., “Energy Efficient Cellular Network Base Station: A Survey,” 2019 IEEE Pune Section International Conference (PuneCon) MIT World Peace University, Pune, India, pp. 1–5, 2019. [Google Scholar]
24. A. K. Sadek, W. Yu and K. R. Liu, “On the energy efficiency of cooperative communications in wireless sensor networks,” ACM Transactions on Sensor Networks (TOSN), vol. 6, no. 1, pp. 5, 2009. [Google Scholar]
25. S. R. Mamidala and S. R. Nalapatla, “Literature review on energy efficiency of base stations and improving energy efficiency of a network through cognitive radio,” Ph.D. dissertation, Blekinge Institute of Technology, Sweden, 2012. [Google Scholar]
26. M. Vinay and Y. R. Rudresh, “A review on green communications,” International Journal of Engineering Research & Technology, vol. 6, no. 13, pp. 1–3, 2018. [Google Scholar]
27. P. O. Oviroh and T. C. Jen, “The energy cost analysis of hybrid systems and diesel generators in powering selected base transceiver station locations in Nigeria,” Energies, vol. 11, no. 3, pp. 687, 2018. [Google Scholar]
28. N. Piovesan, A. F. Gambin, M. Miozzo, M. Rossi and P. Dini, “Energy sustainable paradigms and methods for future mobile networks: A survey,” Computer Communications, vol. 119, pp. 101–117, 2018. [Google Scholar]
29. S. Tabibi and A. Ghaffari, “Energy-efficient routing mechanism for mobile sink in wireless sensor networks using particle swarm optimization algorithm,” Wireless Personal Communications, vol. 104, no. 1, pp. 199–216, 2019. [Google Scholar]
30. M. H. Alsharif, J. Kim and J. H. Kim, “Green and sustainable cellular base stations: An overview and future research directions,” Energies, vol. 10, no. 5, pp. 1–27, 2017. [Google Scholar]
 | This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. |