LTE-Direct vs. WiFi-Direct for Machine-Type ... - IEEE Xplore

1 downloads 0 Views 463KB Size Report
Abstract—Current and future releases of 3GPP standards will include enhancements in LTE-Advanced for machine-type communications (MTC). The main ...
2015 IEEE 26th International Symposium on Personal, Indoor and Mobile Radio Communications - (PIMRC): Workshop on M2M Communications: Challenges, Solutions and Applications

LTE-Direct vs. WiFi-Direct for Machine-Type Communications over LTE-A Systems Massimo Condoluci†, Leonardo Militano† , Antonino Orsino † , Jesus Alonso-Zarate∗, and Giuseppe Araniti† † University

Mediterranea of Reggio Calabria, Italy, DIIES Department Technol`ogic de Telecomunicacions de Catalunya (CTTC), Barcelona, Spain Email: [massimo.condoluci|leonardo.militano|antonino.orsino|araniti]@unirc.it; [email protected] ∗ Centre

Abstract—Current and future releases of 3GPP standards will include enhancements in LTE-Advanced for machine-type communications (MTC). The main reason for this new trend is the fact that 5G wireless systems will need to enable the coexistence between MTC and human-type communications (HTC). Therefore, novel solutions are needed to efficiently exploit the radio resources for MTC, considering, among others, the need to optimize transmissions for small data packets. With this aim, this paper addresses the use of short-range device-to-device (D2D) communications as enabling technology to efficiently manage the radio spectrum and to reduce the energy consumption of MTC devices. We consider a scenario where MTC devices are grouped in a cluster; among the cluster members, one terminal acts as aggregator in charge for (i) receiving data from neighboring terminals via D2D links and (ii) relaying the aggregated data to the base station via macro-cellular link. The main contribution of this paper is to compare the performances of the two most popular D2D technologies, i.e., WiFi-Direct and LTE-Direct, used to transmit data toward the aggregator. The performance evaluation in terms of latency and energy efficiency has been conducted in a wide set of scenarios by varying the number of clustered devices and the data to upload per single MTC device. Index Terms—MTC; D2D; LTE-A; Energy Consumption.

I. I NTRODUCTION

T

HE support of machine-type communications (MTC) via cellular systems is foreseen to be a mandatory requirement for the fifth generation (5G) of wireless networks [1]. Indeed, MTC is expected to play a key role in the 5G scenario as testified by the exponential growth observed in the data traffic generated by heterogeneous devices (such as smart meters, remote sensors, etc.) which send their data to remote servers or to other machines without (or with minimal) human intervention. This opens unprecedented opportunities in different fields (e.g., transport and logistics, smart power grids) belonging to the Internet of Things (IoT) ecosystem [2]. The effective provisioning of MTC over 3rd Generation Partnership Project (3GPP) Long Term Evolution-Advanced (LTE-A) system represents one of the main challenges for cellular network providers [1]. Indeed, MTC puts constraints in terms of: (i) energy efficiency of battery-powered machines, (ii) computational efficiency of low-complexity embedded devices, (iii) low cost deployment to facilitate scaling, and (iv) low latency to support industry-compliant critical control applications [1]. A further issue is that data transmitted by This work has been partially funded by ADVANTAGE (FP7-607774) and the Catalan Government under grant (2014-SGR-1551).

978-1-4673-6782-0/15/$31.00 ©2015 IEEE

MTC devices are typically composed of few bytes. This is challenging because LTE-A systems can efficiently guarantee high data rates to human-related traffic but suffer in terms of capacity when huge amounts of MTC devices attempt to transmit few bytes in a very limited time interval [3]; this dictates for a paradigm shift on the packet scheduling as the minimum amount of radio resources that can be allocated to a single device in LTE-A could actually be too big for the actual needs of MTC traffic [4]. According to above mentioned concerns, an energy-efficient low-latency transmission mode able to efficiently manage MTC traffic composed of few bytes simultaneously transmitted by huge number of devices still needs to be properly designed. With this aim, in this paper we propose to handle MTC traffic through the use of network-assisted device-to-device (D2D) communications [5], [6] [7], whereby two or more devices in mutual proximity establish direct local short range links and bypass the base station (i.e., the eNodeB). This hypothesis is compliant with a large number of MTC applications which foresee that the machines/devices are placed closely to each other [3]. In the solution we propose, the expected benefits introduced by D2D paradigm for supporting MTC data delivery are in terms of lower data transmission time and better energy efficiency w.r.t. the case when MTC devices transmit data directly via macro-cellular links. The results will demonstrate that it is possible to obtain good results without modifying the LTE-A radio access standard for supporting MTC traffic. A comparison of two different D2D technologies for intra-cluster transmission is presented, based on either 3GPP LTE-Direct or WiFi-Direct [8] technology. In particular, the objective of this analysis is to evaluate the most suitable D2D solution for MTC intra-cluster link by considering their performance in terms of latency and energy efficiency in uploading the data from a cluster of MTC devices. The remainder of the paper is organized as follows. In Section II, the reference MTC scenario and system model are described. Section III defines and discusses upon the three different transmission modes considered for data uploading in this paper, i.e. cellular uplink and both in-band and out-ofband D2D schemes. The performance evaluation results are summarized in Section IV, whereas the last section concludes the paper.

2298

2015 IEEE 26th International Symposium on Personal, Indoor and Mobile Radio Communications - (PIMRC): Workshop on M2M Communications: Challenges, Solutions and Applications

Fig. 1.

Reference MTC scenario under LTE-A cellular coverage.

II. R EFERENCE S CENARIO

AND

S YSTEM M ODEL

We consider the reference scenario sketched in Fig. 1. MTC devices are grouped into clusters and send their data to an aggregator via D2D links. These links can either use LTE radio interface in D2D mode or a WiFi-Direct mode, as it will be further detailed below. Once the data from the neighboring terminals is collected, the aggregator relays the aggregated data in one single data packet to the eNodeB using a regular cellular link. As discussed in [9], such operation yields: (i) a better usage of the radio resources and (ii) a reduction in the power transmission both for intra-cluster communications over D2D links and over the cellular link transmissions from the aggregator to the eNodeB. For the cellular link between the aggregator and the eNodeB, a User Equipment (UE) in an LTE-A network typically communicates through a macro-cellular link by sending its own data to the eNodeB. The eNodeB executes the resource allocation every Transmission Time Interval (TTI, lasting 1ms) by assigning the adequate number of Resource Block (RB) pairs to each scheduled UE and by selecting the related Modulation and Coding Scheme (MCS). Scheduling decisions are based on the Channel Quality Indicator (CQI) that is associated to a maximum supported MCS. For the case of local communications, a UE may exploit short-range D2D links thus bypassing the eNodeB and exchanging data directly with the destination of the data. Two approaches exist: (i) in-band D2D (i.e., D2D communication exploiting cellular spectrum and radio interface), and (ii) outof-band D2D (i.e., D2D communication exploiting unlicensed spectrum) [10]. In this paper, we consider the two cases, and compare their performance. When considering the first option and focusing on 3GPP LTE-Direct D2D links, network assisted functionalities, such as the D2D session setup (e.g., bearer setup), the power control, and resource allocation procedures, are executed by the eNodeB. Data transmissions occur in the uplink cellular resources of a Time Division Duplex (TDD) frame. The reasons are [11]: (i) the TDD mode only requires devices to have a single radio chain, thus lowering the total cost and complexity of MTC devices w.r.t. Frequency Division

Duplex (FDD); (ii) the use of uplink slots, as a contrast to the downlink, because it is easier to manage the interference at the eNodeB. When considering out-of-band D2D links, several contributions have already demonstrated the feasibility of WiFi for D2D links. Although the idea of supporting direct links was already defined in the original IEEE 802.11 Standard specification through the ad-hoc mode, the lack of efficient power saving and enhanced QoS support has led to a limited market penetration of this functional mode [12]. For this reason, the Wi-Fi Alliance recently certified WiFi-Direct to support peer-to-peer (P2P) communications between 802.11 devices by exploiting an ad-hoc infrastructure mode. WiFiDirect allows devices to implement the role of either a client or an access point (AP), and hence to take advantage of all the enhanced Quality of Service (QoS), power saving, and security mechanisms typical of the infrastructure mode. Therefore, in this paper we consider that data communication is accomplished by creating a group of devices wherein a group owner (GO) handles service announcement through beacon transmission and is also in charge for the crossconnection of the devices belonging to its own group to an external network (e.g., an LTE-A network as addressed in this paper). Noteworthy, the role of the GO fits well to the role of the aggregator for a cluster of MTC devices, where the GO will also have the task of orchestrating the communications within the cluster to optimize the performance. For simplicity in the analysis, in the two considered D2D approaches, in-band LTE-direct and out-of-band with WiFi-Direct, we consider a single cluster of MTC devices with a single aggregator. In the LTE-direct case, all RB pairs in the system are available for the uplink MTC data and no background interfering traffic is considered. The intra-cluster D2D communications are managed according to an ideal Round Robin policy set by the aggregator, thus assuming a priori knowledge of the network composition. Then, the aggregator exploits a decodeand-forward (DF) relaying configuration operating in halfduplex TDD mode, i.e., cluster members send their data to the aggregator which will then forward all aggregated data to the eNodeB. The scheduling performed by the eNodeB guarantees that D2D transmissions (from MTC devices to the aggregator) and macro-cellular transmissions (from the aggregator to the eNodeB) will never occur in the same TTI. Therefore, in case of LTE-Direct D2D communications, the LTE-A uplink slots are alternatively used by D2D communications and transmissions toward the eNodeB. This solution avoids the intra-cell interference between D2D and macro-cellular connections in scenarios with multiple aggregators. The allowance of such simultaneous operation remains an interesting research area towards maximization of channel reuse and thus capacity [13]. III. U PLOADING T RANSMISSION S CHEMES In this section, we define and discuss the different approaches for transmission of MTC data to the enodeB. These are the direct transmission link to the eNodeB from an MTC, and both the in-band and out-of-band D2D schemes.

2299

2015 IEEE 26th International Symposium on Personal, Indoor and Mobile Radio Communications - (PIMRC): Workshop on M2M Communications: Challenges, Solutions and Applications

A. Transmission Over a Cellular Link Let us consider an LTE-A eNodeB that receives data from a set of devices. Data uploading in the traditional macro-cellular mode occurs through the establishment of independent radio links from each MTC device to the eNodeB. In this case, the eNodeB measures the CQI from each UE and decides the MCS and the number of allocated RB pairs in the uplink slots of the TDD transmission frame. Formally, let K be the set of K MTC terminals; each device has some data dk to upload in a TTI. Let C be the number of available CQI levels and let ck ∈ {1, 2, . . . , C} be the CQI relevant to device k ∈ K. With the macro-cellular mode, the energy consumption of data uploading is intrinsically determined by the amount of data to upload and the spectral efficiency bm for the MCS supported by each device. More in general, the energy efficiency can be expressed as the ratio between the amount D of data to upload (measured in bits) and the energy consumption E (measured in Joule), i.e., η = D/E. The energy consumption for user k over the LTE-A link can be computed as the product of the transmit power Pk per single RB, the number of allocated RBs rk and the transmission time (i.e., one TTI in our considered scenario): Ek = Pk · rk · T T I. Therefore, the overall energy efficiency for the data uploading from all K LTE-A equippeddevices in macro-cellular mode can be dk computed as: η = Pk ·rk ·T T I . k∈K

To promote energy efficiency, we exploit the solution proposed in [4] to guarantee the optimal MCS selection and thus the energy consumption reduction for the data forwarding performed by the aggregator. Being R the number of RBs allocated to the aggregator, the optimal MCS to be adopted ∗ for an energy efficient solution, i.e., M CSR , can be computed as follows:   D ∗ M CSR , = arg max T BS(M CS, R) MCS s.t. D ≤ T BS(M CS, R), M CS ≤ CQI (1) where T BS(M CS, R) is the Transport Block Size (TBS) determined by the MCS and the number of allocated resources R. This maximization process guarantees that the data is sent with the minimum TBS which fulfills the data transmission; energy saving can be obtained by the power decreasing due to the exploitation of a lower-order MCS. However, considering a fixed value for R means that the UE has to use all the R allocated RBs; this may involve inefficiencies in the spectrum allocation. To remove this constraint with the aim of better exploiting the available radio resources, the most energy efficient MCS can be selected as: ∗ ) · R) M CS ∗ = arg min (δmcs (M CSR ∗ MCSR

(2)

∗ where δmcs (M CSR ) is the power offset of each optimal MCS to the basic MCS. This minimization procedure finds the MCS with the smallest transmit power as this is equal to: P = Pbasic · δmcs · R, where Pbasic is the power per RB for the

basic MCS, δmcs is the power offset between MCS and the basic MCS. The adopted solution is to select for each RB number n = [1, R] (R is the number of RBs allocated to the aggregator) the MCS according to equation (1) and from all the resulting MCS values, select the one which minimizes the power transmission according to equation (2). B. D2D-Enhanced Solution based on LTE-Direct Let ci,j be the CQI value for each LTE-Direct D2D link between devices i, j ∈ K, i = j. Each CQI level is associated to a given supported MCS. For a given MCS value m, the bits per RB that can be sent depend on the spectral efficiency for the given MCS, bm , expressed in bit/s/Hz. Finally, let R be the set of available R RBs. Recall the Round Robin radio resource allocation scheme used to coordinate the local D2D transmissions as described in Section II. In case of LTE-Direct D2D links, the maximum uplink resources allocated to each UE is bounded by rk = R/K, ∀k ∈ K (achieved when all devices experience the same channel quality); the maximum data rate for UE k depends by the number of allocated resources rk and the spectral efficiency of the MCS related to the CQI level ck . Since D2D links are inherently in a local scope and thus cover short distances, good channel quality can be typically obtained even if the transmission power is small, thus implicitly yielding energy savings. The proposed solution exploiting LTE-Direct D2D links works as follows when the data collection from a cluster of MTC devices is triggered: • Aggregator selection: The MTC device with the highest macro-cellular CQI is elected as aggregator (by the eNobeB) with the aim of improving the spectral efficiency during data forwarding at the eNodeB. • MCS in the uplink from aggregator to eNodeB: The aggregator implements the energy efficient solution discussed in the previous subsection, to find the optimal MCS and the exact number of RBs among the R available ones to exploit for data forwarding at the eNodeB. • D2D link configuration: The R RBs are equally distributed among the cluster members so that the number of resources per cluster member is bounded by rd = R/|si |. Based on the number rd of RBs available on a single D2D link, an energy efficient configuration is performed for each D2D link within the cluster similarly to the macro-cellular uplink solution discussed earlier in this Section. Upon reception of all packets data from its neighbors, the aggregator transmits the aggregated data to the eNodeB. C. D2D-Enhanced Solution based on WiFi-Direct The WiFi-Direct system is modeled according to the guidelines in [8]. In particular, the influencing parameter is the distance di,j between devices i, j ∈ K, i = j as this directly influences the achievable throughput ti,j over the link between user i and user j. This throughput depends on the inter-node distance that is supposed to be known to the eNodeB. In

2300

2015 IEEE 26th International Symposium on Personal, Indoor and Mobile Radio Communications - (PIMRC): Workshop on M2M Communications: Challenges, Solutions and Applications

particular, the throughput on the WiFi-Direct link will comply with the measurements in [12] where also interference is considered. As this is the scenario of interest for our problem, we refer to the average throughput results in [12]. The average throughput reaches the highest value (about 20Mbps) at less than 1 meter inter-node distances, and decreases reaching zero at about 41 meters of inter-node distance. The WiFi-Direct power consumption values are set like in [14]. The power consumption for an MTC cluster member when transmitting data to the aggregator is: wif i = βwif i + αu Rwif i Ptx

(3)

with βwif i = 132.86mw/M bps, αd = 283.17 mw/M bps and Rwif i follows the values in [12] as a function of the inter-device distance. In listening mode, we assume the power consumption is equal to the power consumption in reception wif i Prx , which follows equation (3), but with αu being replaced by αd = 137.01mw/M bps. In particular, for the WiFi-Direct based solution, we still consider a network-assisted solution where the eNodeB implements the following steps: • Aggregator selection: The election of the aggregator in the cluster is based on the CQI feedbacks of MTC devices as it was described for the LTE-Direct algorithm. • WiFi-Direct links quality estimation: We assume that the eNodeB is aware of the position of the devices in the cluster and can learn the relative distances between the devices. This information gives a measure of the link quality for the WiFi-Direct communication. • WiFi-Direct communications scheduling: Based on the distances, the eNodeB schedules all the D2D communications from the devices in the cluster to perform consecutive unicast transmissions from the MTC devices to the aggregator. In particular, we assume higher priority is given to devices being closer to the aggregator as this means better channel quality. A device with lower priority will be active listening to the channel waiting for its turn to transmit. Once data is transmitted, the device switches in idle mode (i.e., turning-off the radio interface). The perfect scheduling of the WiFi-Direct transmission is an assumption for the proposed model which we believe gives good initial indications on the best case analysis (no collisions occur) for the communications in a cluster of MTC devices. In particular, the standard Medium Access Control (MAC) protocol of the WiFi channel would lead to a very high number of collisions in case of large number of devices with a consequent dropping of the performances. In future work we foresee to model the impact of collisions and errors over the channel as considered in as well as the usage of alternative MAC protocols which can lead to higher performance. IV. P ERFORMANCE E VALUATION A simulation of the two proposed D2D techniques has been R conducted by using Matlab . The objective is to compare the performance of the LTE-Direct and WiFi-Direct solutions. The key performance indicators are:

total uploading time, i.e., the overall time which occurs to upload data by all cluster members (this value accounts for waiting times in the WiFi-direct transmission mode); • average energy consumption, i.e., the mean value of the energy spent by cluster devices (this value accounts for the power consumption in transmission, reception and waiting modes); • energy efficiency, i.e., the ratio between the number of bits to be transmitted (the packet size) and the average energy consumption. Main simulation parameters are listed in Table I. The focus is on a single TTI; for data requiring multiple TTIs, the same solution is applied in consecutive TTIs. Channel conditions for the UEs have been evaluated in terms of SINR experienced over each sub-carrier when path loss and fading phenomena affect the signal reception. The maximum range for a D2D cellular link connection is set to 50m. The performance evaluation has been conducted by following the guidelines defined in [15] and for a number of available RBs R = 100 per cell, a varying data size per device to be uploaded in the [1 − 80] byte range (typical values for MTC) and a varying number of devices in the cluster in the range [10 − 100]. •

TABLE I M AIN S IMULATION PARAMETERS Parameter Carrier frequency Frame Structure Path loss (cell link) Path loss (D2D link) Shadowing std. LTE eNodeB Tx power Maximum UE Tx power Thermal Noise Power and rate control Frequency resources Signaling mode Wi-Fi RF equipment

Value 2 GHz Type 2 (TDD), configuration 1 128.1 + 37.6 log(d), d[km] 16.9 log(d) + 46.8, d[m], f[GHz] 10 dB (cellular); 12 dB (D2D) 46 dBm 23 dBm (D2D mode: 13 dBm) -174 dBm/Hz Open-loop SINR target at 25 dB 20 MHz TDMA Green-field, control rate 18 Mbps Noise fig. 7 dB, noise floor -95 dBm

The total uploading time is shown in Fig. 2(a). As we can notice, both the LTE-Direct and the WiFi-Direct solutions outperform the standard LTE data uploading1, with the LTEDirect solution performing the best. In particular, in the WiFiDirect case, the total time increases linearly with the number of devices and the packet size. This is due to the fact that when using WiFi-Direct, it is not possible to conduct simultaneous parallel transmissions, as with LTE-direct. This effect becomes more apparent with the increase with the number of devices. The average energy consumption for the devices required to upload the data is shown in Fig. 2(b). For this analysis, we considered the power consumption by taking into account waiting, receiving and transmitting period during data uploading; devices are assumed to switch in idle mode once data is uploaded. LTE-Direct outperforms WiFi-Direct, which has even worse performance than the conventional LTE uploading. 1 For standard LTE uploading, a weighted round-robin scheduling is considered where terminals are served in a descending order of uploading time

2301

2015 IEEE 26th International Symposium on Personal, Indoor and Mobile Radio Communications - (PIMRC): Workshop on M2M Communications: Challenges, Solutions and Applications

(a) Total uploading time.

(b) Average energy consumption. Fig. 2.

(c) Energy efficiency.

LTE-Direct vs. WiFi-Direct technology.

This result is related to the fact that over WiFi-Direct we have sequential transmissions in time, whereas over LTE-Direct it is possible to have parallel transmissions over the available RBs. In addition, the transmission power is lower in the case of LTE-Direct. In particular, in the case of WiFi-Direct, the last device uploading its data will be the one spending more energy as it needs to wait for its turn while consuming energy in the channel listening state. Finally, the energy efficiency is shown in Fig. 2(c). As we can observe, in general, the LTE-Direct solution is the best performing solution, followed by the WiFi-Direct solution and the standard LTE. When the amount of data is relatively small ([1 − 30] bytes), WiFi-Direct has even better performance than LTE-Direct. This is related to the combination of high data-rate achieved over WiFi-Direct and the small data. As a consequence, the effect of sequential transmissions in time when WiFi-Direct is used is mitigated by the small amount of time needed to forward the data. Nevertheless, when the packet size increases, devices using LTE-Direct solution are able to upload more efficiently their data. Moreover, it is possible to note that for high number of devices and larger data size, the energy efficiency for LTE-Direct becomes higher. This is because, in case of higher packet sizes, the energy efficiency of LTE-A increases with the TBS.

V. C ONCLUSION In this paper, we have proposed two different D2D solutions for efficient MTC data uploading over LTE-A systems. In particular, LTE-Direct and WiFi-Direct have been compared as enabling technologies to reduce the energy consumption of a set of clustered devices when uploading the data to the eNodeB through a single aggregator device. Simulation results show that both solutions outperform standard LTE uploading where each MTC devices transmits directly to the eNodeB. The LTE-Direct technology is able to provide the most energy efficient communication scheme when the number of user is relatively high. However, WiFi-Direct outperforms LTE-Direct in terms of energy efficiency in case of small amount of data.

R EFERENCES [1] K. Zheng, S. Ou, J. Alonso-Zarate, M. Dohler, F. Liu, and H. Zhu, “Challenges of massive access in highly dense lte-advanced networks with machine-to-machine communications,” Wireless Communications, IEEE, vol. 21, no. 3, pp. 12–18, June 2014. [2] M. Nitti, R. Girau, A. Floris, and L. Atzori, “On adding the social dimension to the internet of vehicles: Friendship and middleware,” in Communications and Networking (BlackSeaCom), 2014 IEEE International Black Sea Conference on. IEEE, 2014, pp. 134–138. [3] M. Condoluci, M. Dohler, G. Araniti, A. Molinaro, and K. Zheng, “Toward 5g densenets: architectural advances for effective machine-type communications over femtocells,” Communications Magazine, IEEE, vol. 53, no. 1, pp. 134–141, January 2015. [4] K. Wang, J. Alonso-Zarate, and M. Dohler, “Energy-efficiency of lte for small data machine-to-machine communications,” in Communications (ICC), 2013 IEEE International Conference on, June 2013. [5] S. Mumtaz, L.-L. Yang, C. Wang, F. Adachi, and N. Ali, “Smart-deviceto-smart-device communications,” Communications Magazine, IEEE, vol. 52, no. 6, pp. 88–89, 2014. [6] S. Andreev, A. Pyattaev, K. Johnsson, O. Galinina, and Y. Koucheryavy, “Cellular traffic offloading onto network-assisted device-to-device connections,” Communications Magazine, IEEE, vol. 52, no. 4, pp. 20–31, April 2014. [7] L. Wang and H. Wu, “Fast pairing of device-to-device link underlay for spectrum sharing with cellular users,” IEEE Communications Letters, vol. 18, no. 10, pp. 1803–1806, 2014. [8] “Wi-Fi Peer-to-Peer (P2P) Technical Specification v1.0, Wi-Fi Alliance, P2P Technical Group,” Tech. Rep., 2009. [9] A. Laya, K. Wang, A. Widaa, J. Alonso-Zarate, J. Markendahl, and L. Alonso, “Device-to-device communications and small cells: enabling spectrum reuse for dense networks,” Wireless Communications, IEEE, vol. 21, no. 4, pp. 98–105, August 2014. [10] A. Asadi, Q. Wang, and V. Mancuso, “A survey on device-to-device communication in cellular networks,” Communications Surveys Tutorials, IEEE, 2014. [11] K. Doppler, M. Rinne, P. Janis, C. Ribeiro, and K. Hugl, “Deviceto-device communications; functional prospects for lte-advanced networks,” in Workshops IEEE International Conference on Communications (ICC), June 2009, pp. 1–6. [12] A. Pyattaev, K. Johnsson, S. Andreev, and Y. Koucheryavy, “3gpp lte traffic offloading onto wifi direct,” in Wireless Communications and Networking Conference Workshops, 2013. IEEE, 2013. [13] A. Meloni and M. Murroni, “Interference calculation in asynchronous random access protocols using diversity,” Telecommunication Systems, pp. 1–9, 2015. [14] J. Huang, F. Qian, A. Gerber, Z. Mao, S. Sen, and O. Spatscheck, “A close examination of performance and power characteristics of 4G LTE networks,” 10th international conference on Mobile systems, applications, and services (MobySis), Jun. 2012. [15] 3GPP, “TS 36.213 Evolved Universal Terrestrial Radio Access (EUTRA): Physical layer procedures, Rel. 11,” Tech. Rep., Dec. 2012.

2302