Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

M2M Potentials in logistics and transportation industry

M2M Potentials in logistics and transportation industry Logist. Res. (2016) 9:15 DOI 10.1007/s12159-016-0142-y ORIGINAL PAPER 1 2 1 • • • Yasir Mehmood Safdar Nawaz Khan Marwat Koojana Kuladinithi 1 3 1 4 • • • Anna Fo ¨ rster Yasir Zaki Carmelita Go ¨ rg Andreas Timm-Giel Received: 15 December 2015 / Accepted: 16 July 2016 / Published online: 28 July 2016 The Author(s) 2016. This article is published with open access at Springerlink.com Abstract All over the world, road congestion is among the scheme by using Long-Term Evolution Advanced (LTE- most prevalent transport challenges usually in urban envi- Advanced) Relay Nodes, which aggregates M2M traffic to ronments which not only increases fuel consumption and maximize radio resource utilization. Extensive system- emission of harmful gases, but also causes stress for the level simulations are performed using an LTE-Advanced- drivers. Intelligent Transportation System (ITS) enables a based model developed in the RIVERBED modeler to better use of the infrastructure by connecting vehicles to evaluate the performance of the proposed data multiplexing other vehicles as well as infrastructure and thus delivers a scheme. Simulation results show that approximately 40 % faster communication opportunity to ensure safe and secure more smart M2M devices used in ITS and logistics are driving. Machine-to-machine (M2M) communication is served per LTE-Advanced cell under the present system one of the latest information and communication tech- settings. nologies which offers ubiquitous connectivity among sev- eral smart devices. The use of mobile (cellular) M2M Keywords Road congestion  Intelligent transportation communications has emerged due to the wide range, high systems  Information and communication technologies reliability, increased data rates, decreased costs as well as Machine to machine  Relay nodes  LTE-Advanced easy and short-term deployment opportunities. Since the radio spectrum is a scarce resource, M2M traffic can potentially degrade the performance of mobile networks 1 Introduction due to the large number of devices sending small-sized data. This paper presents an efficient data multiplexing The dramatic use of communication technologies (wired and wireless), embedded systems as well as increasing penetration of the Internet has not only revolutionalized This article is part of a focus collection on ‘‘Dynamics in Logistics: human lives, but also reshaped almost all types of business Digital Technologies and Related Management Methods.’’ & Yasir Mehmood Andreas Timm-Giel ym@comnets.uni-bremen.de timm-giel@tuhh.de Safdar Nawaz Khan Marwat Communication Networks, University of Bremen, Bremen, safdar@uetpeshawar.edu.pk Germany Koojana Kuladinithi Department of Computer Systems Engineering, University of koo@comnets.uni-bremen.de Engineering and Technology, Peshawar, Peshawar, Pakistan Anna Fo¨rster Computer Science Department, New York University Abu anna.foerster@comnets.uni-bremen.de Dhabi (NYUAD), Abu Dhabi, UAE Yasir Zaki Institute of Communication Networks, University of yasir.zaki@nyu.edu Technology, Hamburg, Germany Carmelita Go¨rg cg@comnets.uni-bremen.de 123 15 Page 2 of 11 Logist. Res. (2016) 9:15 Energy and Ulies models and processes [1]. M2M communication is one of Adopon of M2M technology in industry – 2015 Automove the emerging technologies which offers ubiquitous con- Retail nectivity among intelligent devices, hence is one of the Consumer electronics 17% 19% 37% major enablers of the Internet-of-Things (IoTs) vision [2]. e-healthcare 28% 32% IoT is an innovative concept which offers to connect smart Logiscs 29% 32% Manufacturing devices often called things endowed with several sensing, automation as well as computing capabilities, with the Internet [3]. Resultantly, the connected devices are revo- Fig. 1 An illustration of leading industries adoption M2M technol- lutionalizing the future cyber physical systems, yielding ogy worldwide, based on the Vodafone report 2015 [8] several applications. Moreover, the mobile network oper- ators are partnering with industrial organizations in order to Besides, it also exhibits unique characteristics such as an bring forth innovative IoT services to facilitate end con- increasing device volume, sending small payloads, the sumers. For instance, M2M applications include intelligent demand for various mobility profiles, time-controlled, and transportations, logistics and supply chain management, mainly delay tolerant. Since the spectrum for mobile net- e-health, smart metering, surveillance and security, smart works will remain a scarce resource, efficient utilization of cities, and home automation [4–6]. Thus, M2M commu- radio spectrum is one of its major requirements. Therefore, nication is foreseen to reshape the business of operators, the objective of this paper is to exploit LTE-Advanced RNs service providers, M2M enterprises, and M2M enablers [7]. to multiplex small-sized M2M data packets in order to Vodafone revealed that M2M communication is ensure efficient LTE-Advanced radio resource utilization becoming one of the driving forces for businesses which and thus to support a large number of devices. inspires to bring forth innovative solutions almost in every The rest of the paper is structured as follows. We firstly sector such as logistics, automotive industry, cities, homes, present an overview of the leading communication tech- schools, and workplaces [8]. Approximately 90 % of the nologies used in ITS and logistics, followed by a generic companies worldwide have adopted M2M technology and overview of the two latest technologies, i.e., the Institute of imparted it as one of the most favorable technologies for Electrical and Electronics Engineers (IEEE) 802.11p and achieving noticeable outcomes. Automotive industry is one 3GPP LTE-Advanced in Sect. 2. An overview of major of the top sectors for adopting M2M technology. Approx- M2M services in ITS and logistics are discussed in Sect. 2. imately 32 and 17 % increasing growth rates for adopting Then, we discuss problem definitions by highlighting how M2M technology have been noticed in automotive and LTE-Advanced resources can be used inefficiently by logistic sectors, respectively, as shown in Fig. 1. Thus, ITS smart M2M devices in Sect. 3. Section 4 presents the and logistics are considered as one of the potential M2M proposed uplink M2M data multiplexing scheme [13]. The users worldwide [9]. In addition, NOKIA forecasted that simulation environment and parameter settings are pre- the use of M2M technology in automotive industry and sented in Sect. 5. We discuss our simulation results in logistical processes will dominate other applications in the Sect. 6. In the end, conclusions are drawn in Sect. 7.In future [10]. One of the major motivations is to deliver a addition, a list of most frequently used acronyms is pre- fully managed infrastructure which primarily guarantees, sented in Table 1. e.g., safe and secure driving, in time delivery, smart monitoring, and tracking of assets. Resultantly, this can revolutionize the existing methods of transportation and 2 Mobile M2M communications freight movements. In addition, optimum system perfor- mance can be achieved by reducing factor of costs, pol- This section presents an overview of ETSI (European lution and emission of harmful gases. Telecommunications Standards Institute) M2M architec- Mobile M2M communication greatly differs from tra- ture followed by the major M2M use cases and services in ditional human-to-human (H2H) communication in terms transportation and logistics. of traffic density, data packet size, and quality of service (QoS) requirements [11]. For instance, an experimental 2.1 ETSI M2M architectural overview study done in [12] shows that M2M traffic exhibits a sig- nificantly different behavior than the traditional smart- The high-level ETSI M2M network architecture is shown phone traffic in various aspects. For example, unlike in Fig. 2 [14]. The major components of mobile M2M traditional mobile traffic, M2M is an uplink dominant communication architecture include the device, communi- traffic which particularly generates bursty traffic volumes. cation as well as server domains. The primary functionality of the device domain is to collect and send sensor data such Sending information from M2M devices to base station. 123 Logist. Res. (2016) 9:15 Page 3 of 11 15 Table 1 List of used abbreviations technology for Vehicular Ad hoc Networks (VANET). IEEE 802.11p wireless access introduces minimum delay Abbreviation Acronym in ITS. However, its employment is limited due to its MAC Medium Access Control decentralized nature. The recent research has revealed that MME Mobility Management Entity maximum operating efficiency of IEEE 802.11p can be PHY Physical Layer achieved by supporting it with the LTE-Advanced mobile PRB Physical Resource Block networks. The main features of LTE-Advanced technology QCI QoS Class Identifiers include wide availability of modules/devices, system RLC Radio Link Control existence, decreased costs as well as easy deployment. RRC Radio Resource Control Since the cellular modules and sensors are easily available, e.g., in vehicles, the applications of LTE-Advanced mobile SAE System Architecture Evolution networks are dramatically increasing in automotive sector. S-GWs Serving Gateways The authors in [16–18] compared the performance of SGSN Serving GPRS Support Node two different technologies under varying channel and TTI Transmission Time Interval traffic load conditions. Moreover, several issues due to WLAN Wireless Local Area Network decentralized nature of IEEE 802.11p are highlighted in WPAN Wireless Personal Area Network [16]. For instance, the main shortcomings are less relia- WSN Wireless Sensor Network bility due to uncoordinated Medium Access Control (MAC) procedures, risk of network congestion, higher as the internal temperature and humidity level of a con- vehicle mobility, and low scalability. Moreover, the num- tainer, position and speed of a vehicle, and fuel con- ber of network-connected vehicles adversely affect the sumption. The role of the communication network is to performance of the standard system. On the other hand, create a communication path between the devices and LTE-Advanced networks are more reliable, scalable, and servers through either wired or wireless networks such as capable of supporting higher density of vehicles. In addi- Digital Subscriber Line (DSL) and cellular networks (e.g., tion, the authors analyzed the performance of the above LTE-Advanced), respectively. Finally, the server domain technologies in terms of delay, reliability, scalability, and consists of a middleware layer where the collected packets mobility support. The authors concluded that 3GPP LTE- go through several application services and later are used Advanced systems are more efficient in terms of high by related agencies. Thus, M2M technology employs scalability and mobility support than IEEE 802.11p stan- wired, wireless, and hybrid communication opportunities dard. However, it is further concluded that the performance among devices to ensure a fully automatic acquisition, of IEEE 802.11p is more sensitive during increased den- processing, and transmission of data. Thus, M2M repre- sity, traffic load, and higher mobility. sents a broad next-generation technology that is primarily incorporated in modern automobile industries to improve 2.2 M2M use cases the ease, safety, and quality of human life. The wired access technologies provide less delay, high throughput Mobile M2M communication offers manifold applications and are more reliable. Despite the aforementioned capa- and services in modern transport and logistical processes bilities, wired systems cannot be used in transport and such as onboard security, traffic and infrastructure man- logistical systems due to several limitations such as lack of agement, fleet management, and route planning [19]. In scalability, cost efficiency, and mobility. However, above- case of an emergency, the collected data are sent to other mentioned limitations can be overcome by incorporating vehicles as well as infrastructure to gain immediate atten- wireless (cellular) technologies to achieve maximum sys- tion. To avoid further incidents, communication between tem efficiency and reliability. the infrastructure and the vehicles must be very fast to Recently, the key emerging wireless technologies used detect emergency messages and deliver warning messages in modern transport and logistical systems include IEEE immediately. Similarly, traffic and infrastructure manage- 802.11p [15] (standardized to support vehicular commu- ment play a prominent role in handling the problem of road nication) and 3GPP LTE-Advanced mobile networks [16]. congestion. It tackles the problem by providing two-way IEEE 802.11p is achieved by making few advancements in communication opportunities between vehicles and well-known 802.11 Wireless Local Area Network (WLAN) infrastructure. Vehicles can send status updates about the technology. It is also considered as the most feasible position, speed, fuel consumption, and delivery status reports to the infrastructure and can also receive relevant instructions about road accidents and emergency braking 3rd Generation Partnership Project—3GPP—is the leading stan- system [20]. Moreover, M2M communications support dardization organization for mobile networks. 123 15 Page 4 of 11 Logist. Res. (2016) 9:15 Backend server (e.g., transportaon control center) Gateway DeNB: Donor eNB Downlink: Mobile communicaon Uplink: domain (e.g., LTE- Advanced) DeNB DeNB DeNB DeNB Applicaon domain Smart sensors deployed to (e.g., ITS and monitor temperature, humidity, pressure, power, water, etc. logiscs) Inventory Downlink informaon Vehicles cross talk for on-board Warehouse management Accidental/emergency security, status update, fleet (monitoring all sensors deployed unicast/broadcast to uplink informaon management, etc. for various metering operaons) all devices (vehicles) Fig. 2 ETSI mobile M2M communication architecture along with an overview of major M2M applications in intelligent transportation and logistics, based on [14] several operations such as tracking of a stolen vehicle, In addition to the above applications, M2M communi- traffic reports, and route planning as well as infotainment cation provides additional services in logistics such as services [6]. For instance, to recover a stolen vehicle, SVT decreased operational cost, high inventory flexibility, (Stolen Vehicle Tracking) service providers request data increased supply chain visibility, and reduced loss of about the location from Telematic Control Unit (TCU) vehicles and containers [21]. In supply chains, M2M located inside the vehicle. In addition, drivers are also technology enables tracking the status of goods in real-time updated by sending reports regarding traffic in a particular via M2M devices. This increasing visibility allows for region so that they can change or plan new routes in case of significant reduction in uncertainties in supply chain [22]. traffic jam or an emergency. Furthermore, infotainment Similarly, in a warehouse, M2M devices can be deployed services aim to provide news/information to drivers and to track the inventory so that stockholders and enterprises passengers through mobile TV, web-browsing, etc. can respond to the market dynamics and to decide when to Fleet management is also one of the major M2M refill and when to go on sale. Moreover, cross talk among applications in logistics [9]. The movements of vehicles, vehicles can also be effective to get immediate assistance. containers, buses, and cars are being tracked regularly Additionally, direct delivery of inventory from one vehicle through devices which collect data of the location, vehicle to another without storing it in a warehouse can also be speed, temperature, distribution progress, fuel consumption accomplished through mutual information sharing. Con- and send this information to monitoring servers. Through sequently, it can significantly reduce the required space of regular monitoring, several activities of the system can be warehouse, customer’s waiting time as well as the opera- performed in an efficient way. For instance, the goods tional costs for business entities. which are transported from one place to another are mon- itored regularly in order to accomplish in time delivery and to handle any undesirable situation during shipment pro- 3 Problem descriptions cesses. The cargo moves across several regions; therefore, it must be monitored in order to stay updated. Moreover, Mobile systems such as LTE-Advanced are mainly reporting gives the exact location of the freight, and thus designed to support the increasing traditional mobile traffic the conditions of the objects can be easily monitored. In by enabling improved broadband services [23]. For applications such as warehouse management, fleet man- instance, Ericsson in its mobility report [24] anticipates agement, robotics, and control systems, alarms are also approximately 5.6 billion cellular-based active smart- used to detect critical or emergency situations. phones by 2019. Therefore, in order to support such a 123 Logist. Res. (2016) 9:15 Page 5 of 11 15 100.00 Device payload (%) values, e.g., 20–26 representing favorable channel condi- 90.00 Zero padding (%) tions, the capacity of the PRB increases. However, small 80.00 packets with increasing percentage of padded zeros are 70.00 transmitted. This shows that more packets can be accom- 60.00 modated per PRB under favorable channel conditions in 50.00 40.00 the given exemplary scenarios. Since mobile radio 30.00 resources are valuable assets and scarcely available, it is 20.00 required to ensure efficient utilization of radio resources for 10.00 M2M communications. 0.00 0 2 4 6 8 101214161820222426 MCS Index 4 Proposed M2M data multiplexing Fig. 3 An example illustration of inefficient PRB utilization for M2M communication In the proposed data multiplexing scheme, the small-sized M2M data packets are multiplexed at the Packet Data massive mobile traffic, several enhancements in existing Convergence Protocol (PDCP) layer of the RN. A high- LTE systems [25] such as Relay Nodes (RNs) [26], mas- level illustration of E-UTRAN (Evolved UMTS Terrestrial sive Multiple Input–Multiple Output (MIMO) systems Radio Access Network) architecture with the functionali- [27], and femto cells [28] were introduced to fulfill the ties of LTE-Advanced relaying is depicted in Fig. 4. The IMT-Advanced (International Mobile Telecommunications RN communicates with devices using access link (Uu), Advanced) demands of maximum throughput of upto 1 whereas it communicates with DeNB using backhaul link Gbps in downlink and 500 Mbps in uplink [29]. On the (Un). Thus, it behaves like an eNB toward the M2M other hand, mobile standards are foreseen as an attractive devices, whereas it acts as a UE toward the DeNB. option to support future M2M applications such as in The RN PDCP is selected to aggregate M2M packets in transportation, logistics, smart city, and living [2]. Several order to maximize the multiplexing gain. This increasing forecasts have also reported a considerable market growth multiplexing gain is achieved without aggregating the for M2M device volume, making quality of service pro- additional overheads such as those from the PDCP, Radio visioning a huge challenge [10]. Consequently, mobile Link Control (RLC), and Medium Access Control (MAC). networks will not be able to support the increasing number A high-level illustration of the proposed scheme is given in of M2M devices worldwide [30]. Fig. 5. Generally, the PDCP layer is a part of LTE air According to 3GPP, the smallest unit of the radio interface control and user plane protocols. It exists in the spectrum allocatable to a single device is 1 PRB which is UE, Donor eNB (DeNB), and in the RN. The major capable of transmitting several hundred bits under favor- functionalities of PDCP layer for the user plane include able channel conditions. However, allocating 1 PRB to a header compression and decompression, user data transfer, single M2M device used in transportation/logistics could delivery of upper layer Packet Data Units (PDUs) in significantly degrade radio spectrum utilization sequence, and retransmission of PDCP Service Data Units [13, 19, 31]. This is due to the fact that the capacity of a (SDUs). Moreover, the control plane services include PRB can be much higher than the actual size of the device ciphering and integrity protection and transfer of control payload under favorable channel conditions. Figure 3 plane data. The proposed multiplexing scheme can also be illustrates how inefficiently a PRB can be utilized when a applied at the lower layers of the RN such as RLC and device sends small-sized packets when experiencing good MAC on the Uu interface or in the PDCP of Un interface. channel conditions. An exemplary device payload of 4 However, in these cases, the multiplexing gain will be bytes is considered in order to evaluate the utilization of a reduced due to the multiplexing of additional headers along PRB against all the possible values of the Modulation and with the original small-sized data packets [13]. Coding Scheme (MCS). A PRB utilization efficiency is A multiplexing buffer is created at the RN PDCP layer evaluated in terms of percentage of the actual payload and which multiplexes the incoming packets according to the the padded zeros with respect to MCS index. It can be seen available service rate which is equal to Transport Block from Fig. 3 that with the lower MCS values depicting Size (TBS)—RN Un protocol overheads in terms of bits unfavorable channel conditions, the PRB capacity is per TTI, as shown in Fig. 5. The Un protocol overhead smaller, and therefore, it is fully utilized to transmit the includes 12, 28, and 4 bytes for GTP (GPRS Tunneling given payload. On the other hand, with the higher MCS Protocol), UDP (User Datagram Protocol)/IP, and layer 2, respectively. The multiplexed packet is sent to the RN GTP over the Un link. The additional overheads such as from Physical Resource Block. Ulizaon/PRB in % 15 Page 6 of 11 Logist. Res. (2016) 9:15 Applicaon server Serving Overcoming gateway shadowing problems (S-GW) (DeNB) (RN) (M2M) Un Uu Un Uu Un Uu Hotspot coverage for ITS and logiscal operaons Coverage extensions Fig. 4 An illustration of LTE-Advanced relaying technology. Re-drawn based on [26]) IP Applicaon GTP GTP GTP GTP MUX TCP/UDP UDP UDP UDP UDP IP IP IP IP λ λ λ 1 2 λ IP PDCP PDCP PDCP PDCP RLC RLC RLC RLC RLC Layer 2 MAC MAC MAC MAC MAC PHY PHY PHY PHY PHY PHY M2M device RN DeNB aGW Fig. 5 A high-level illustration of the proposed multiplexing scheme along with protocols stack of the M2M device, RN, and DeNB the GTP, UDP/IP, PDCP, and RLC are added. The block low loaded scenario, there is a comparatively longer delay diagram of the data differentiating algorithm at the DeNB due to the low arrival rate. Consequently, the performance is given in Fig. 6. From the physical layer of the RN, the of delay sensitive M2M applications can be degraded. To multiplexed packets are sent to the DeNB over the Un tackle this issue, an expiry timer, T is introduced. The max interface, as depicted in Fig. 5. timer is set with a fixed value such as 10 ms in the current The multiplexing approach significantly improves PRB implementation. This means that the buffer serves the utilization. However, there are certain constraints regarding multiplexed packet after 10 ms at the latest. The value of latency requirements of high priority M2M traffic such as the timer could also be adaptive, i.e., it can change its value accidental information, emergency alerting, and e-health. according to priorities of the incoming packets. For this This is due to the fact that each packet waits until the size purpose, the algorithm must be fully aware of various of the buffer B is equal to the available TBS—RN Un priorities of M2M applications. protocol overheads. In a highly loaded scenario, the wait- In this work, a fixed RN is implemented and used to ing time is not long due to a high arrival rate. However, in a multiplex the uplink traffic from M2M devices deployed 123 Logist. Res. (2016) 9:15 Page 7 of 11 15 Start Router Packet arrival at the PDCP layer DeNB of the DeNB Remote Access Backhaul link Access link server gateway RN Identify the destination IP of the X2 received packet If False The received packet belongs (incoming packet IP == to the standalone user DeNB IP) M2M devices True Send the packet to GTP-U Fig. 7 Project editor of the LTE-Advanced-based model developed in the RIVERBED Modeler simulator The received packet belongs to the RN • The overall service rate is l ¼ l  5ðPRBsÞ b/s. Send to GTP-U via IP and PRB UDP • So, the system utilization q can be given as Arrival rate k Utilization; q ¼ ¼ ð1Þ De-tunnel to extract the Service rate l original transmitted standalone packets N  656 Utilization; q ¼ ð2Þ • Moreover, the maximum number of devices served can Tunnel each packet again be given as follows, and route towards aGW q  321600 N ¼ ð3Þ max end Fig. 6 Block diagram of the data differentiating algorithm at the DeNB 5 Simulation model and parameters in ITS and logistics (i.e., for onboard security and The Optimized Network Engineering Tool (OPNET) metering purposes). The position of the RN corresponds Modeler which is newly named as RIVERBED Modeler is to an MCS of 16, whereas 5 PRBs are allocated to RN to used as a primary modeling, simulation, and analysis tool serve M2M traffic. The given values of MCS and PRBs for this research. It provides a simulation environment for correspond to TB size of 1608 bits per TTI. Since a TTI the performance measurements of communication net- is of 1-ms duration, the TB size can also be given as works [32]. The project editor of the LTE-Advanced-based 1608000 bits/s. According to the 3GPP standardizations, model developed in the simulator with several nodes along the capacity of a single PRB varies according to the with LTE-Advanced functionalities and protocols is shown MCS. Therefore, under the given values of MCS and in Fig. 7. Additionally, an RN is implemented and placed PRBs, 321 bits per TTI or 321000 bits/s can be sent within the coverage of DeNB to relay and multiplex M2M within a single PRB. In general, the system utilization, q data traffic. The DeNB is responsible to connect UEs/M2M can be determined in terms of the arrival rate, k and the devices with the transport network, thus includes both radio service rate, l. Additionally, the maximum number of interface and transport protocols to communicate with served devices, N can be determined according to the max UEs/M2M devices and the core network, respectively. The following procedure, radio protocols traffic coming from aGW toward the RN. The remote server and the aGW (access gateway) are • The size of the each transmitted M2M packet per interconnected with an Ethernet link with an average delay second is P ¼ 656 bits. size of 20 ms. Moreover, aGW acts as the tunneling point for • The arrival rate k due to N number of devices is the downlink and uplink traffic coming from PDN-GW and ðN  656Þb=s. DeNB, respectively. The aGW node protocols include the • The service rate per PRB is l ¼ 321000 b/s. PRB 123 15 Page 8 of 11 Logist. Res. (2016) 9:15 Table 2 Simulation parameters work. Later, we discuss our simulation results in detail with varying load conditions. In this work, we consider M2M Parameters Values devices used in ITS and logistics in order to generate Simulation length 1000 s uplink traffic. eNB coverage radius 350 m Min. eNB UE distance 35 m 6.1 Simulation scenarios description Max terminal power 23 dBm Terminal speed 120 km/h The scenarios are simulated according to three major cat- Mobility model Random way point (RWP) egories, i.e., no multiplexing, multiplexing without timer, Frequency reuse factor 1 and multiplexing with timer. In the first category, M2M data packets are relayed in uplink without multiplexing. In Transmission bandwidth 5 MHz the second category, the data packets from all the active No. of PRBs 25 devices are multiplexed at the RN before being sent to the MCS QPSK, 16QAM, 64QAM, DeNB. However, in this category, no expiry timer is con- Channel models Pathloss, slow fading, fast fading sidered to control the multiplexing process, and the mul- Path loss 128:1 þ 37:6 log ðRÞ, R in km tiplexed packet is served when its size is large enough to Slow fading Log-normal shadowing, correlation 1, deviation 8 dB utilize full capacity of the available spectrum. In the third Fast fading Jakes-like method [33] category, an expiry timer is introduced in order to limit the RN parameters multiplexing delay especially in the low loaded scenarios. In this case, the multiplexed packet is served after T at PRBs for RN 5 max the latest. Each category is further divided into several Corresponding MCS 16 scenarios. The scenarios are simulated for both load con- TBS 1608 bits ditions (i.e., low and high). In the low loads, the number of Simulated scenarios No multiplexing, multiplexing without timer, and multiplexing devices, N, is kept 200 in the first scenario. The value of with timer N is incremented by 200 in the subsequent scenarios till the Timer expiry values 10 ms (Sect. 6.2) limiting case, i.e., when all 5 PRBs are fully utilized. In Timer expiry values 5, 10 and 20 ms (Sect. 6.3) high loads, 2800 devices are placed in the first scenario, Type of RN Fixed and the number is also incremented by 200 in the subse- M2M Traffic model [34] quent scenarios. Message size 38 bytes (constant) at the device PDCP Inter-send time 1 s (constant) 6.2 Simulation results The simulation results for the mean number of used PRBs Internet Protocol (IP) and Ethernet. The aGW and DeNB with 95 % confidence interval (CI) are illustrated in Figs. 8 nodes (names as eNB1..) communicate through IP routers and 9 for the low and high load conditions, respectively. (R1..). QoS parameters at the Transport Network (TN) The values of the upper and lower bound of CI are very ensure QoS parameterization and traffic differentiation. small in most of the scenarios as shown in Figs. 8 and 9. The user movement in a cell is emulated by the mobility These PRBs are used by RN to transmit the multiplexed model by periodically updating the location of the user. data toward the DeNB. The user mobility information is stored in the global user The simulation results clearly show the efficient uti- database (Global-UE-List). The channel model parameters lization of PRBs in uplink with the proposed multiplexing scheme. For instance, in the no multiplexing scenario with for the air interface include path loss, slow fading, and fast fading models. In this paper, the simulation modeling 400 devices, the arrival rate at the PHY layer of the RN is mainly focuses on the user plane to perform E2E perfor- 262.4 bit/TTI which almost utilize 1 PRB. However, in the mance evaluations. Furthermore, Table 2 presents a case of multiplexing, only half of the PRBs are used to detailed description of simulation parameters and settings. serve the 400 devices, see Fig. 8. Similarly, without mul- tiplexing, the RN serves nearly 2400 devices with 5 PRBs in uplink, which is actually the limiting case as shown in 6 Results and discussion 5 PRBs are allocated to RN in order to serve M2M devices. These This section investigates the performance of the proposed PRBs are fully utilized when the number of devices is raised to 2600 multiplexing scheme in the low and high loaded scenarios. without multiplexing. So the high load scenarios start with 2800 We firstly describe all scenarios which are simulated in this devices. 123 Logist. Res. (2016) 9:15 Page 9 of 11 15 5.00 0.03 No mulplexing No mulplexing 4.50 Mulplexing without mer Mulplexing without mer 4.00 Mulplexing with mer Mulplexing with mer 3.50 0.02 3.00 2.50 2.00 0.01 1.50 1.00 0.50 0.00 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 0.00 M2M devices 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 M2M devices Fig. 10 A comparison of packet mean E2E delay in low loaded scenarios. T ¼ 10 ms max Fig. 8 A comparison of mean PRBs used with and without data multiplexing in low loaded scenarios. T ¼ 10 ms max 0.03 5.00 4.50 No mulplexing Mulplexing without mer 4.00 0.02 Mulplexing with mer 3.50 3.00 No mulplexing 0.01 2.50 Mulplexing without mer 2.00 Mulplexing with mer 1.50 0.00 1.00 2800 3000 3200 3400 3600 3800 4000 4200 4400 0.50 M2M devices 0.00 2800 3000 3200 3400 3600 3800 4000 4200 4400 Fig. 11 Comparing packet mean E2E delay in high loaded scenarios M2M devices with timer expiry, T ¼ 10 ms max Fig. 9 Comparing simulation results of mean PRBs used with and without data traffic multiplexing in high loaded scenarios. delay is higher in the case of multiplexing for low loads as T ¼ 10 ms max shown in Fig. 10. This increase in delay time is due to less arriving packets especially in the low loaded scenarios. The buffer aggregates packets until its size plus the additional Fig. 8. However, in case of multiplexing, the number of devices served by the RN nearly doubles, i.e., approxi- Un overhead is equal to the available TBS. However, the use of an expiry timer limits the delay by serving the mately 4200 devices are served as depicted in Fig. 9.This is due to the fact that in the case of no multiplexing, each aggregated packets at the latest after 10 ms. On the other hand, in high loaded scenarios, the E2E delay is slightly data packet contains an additional Un air interface over- head of GTP, UDP, IP, and layer 2. The additional over- higher compared to the case of no multiplexing, see Fig. 11. This is due to high arrival rate and the buffer head causes an extra PRB usage and overall reduces the PRB utilization efficiency. Moreover, in low loaded sce- aggregates the incoming packets to make a large aggre- narios such as with 200 devices (see Fig. 8), the average gated packet within less time. Moreover, the value of the number of PRBs used is slightly higher in the case of E2E delay is very large in fully loaded scenarios as multiplexing with timer. This is due to the fact that the RN depicted in Fig. 10, when the RN utilizes all 5 PRBs with the maximum capacity. serves the traffic at the latest after 10 ms, and thus, the PRB is not necessarily used with its maximum capacity due to 6.3 Impact of timer expiry on data multiplexing the low arrival rate. However, in high load scenarios, the timer has almost no impact, and a nearly equal number of In this section, we investigate the impact of timer expiry, PRBs is used with and without timer. Figures 10 and 11 show the simulation results of the T on mean PRB utilization and E2E delay by varying max input traffic. However, we only consider low loaded sce- mean packet E2E delay in low and high load scenarios, respectively. The results show that the value of packet E2E narios due to the fact that the timer expiry has almost no Mean PRBs used Mean PRBs used Mean E2E delay [s] Mean E2E delay [s] 15 Page 10 of 11 Logist. Res. (2016) 9:15 1.40 average, as it allows more packets to be multiplexed before Tmax = 05 ms timer expiry. Resultantly, it increases the multiplexing Tmax = 10 ms 1.20 Tmax = 15 ms gain. Additionally, it is further noted that larger values of 1.00 T has almost no affect on PRBs used in the case of high max 0.80 loads. For instance, when N ¼ 1000, the average values of PRBs used are 1.219 and 1.213 for the maximum waiting 0.60 time of 10 and 15 ms, respectively. 0.40 Similarly, Fig. 13 compares the simulation results of the 0.20 mean E2E delay for the given timer expiry values. As discussed in the beginning, the larger values of waiting 0.00 100 200 300 400 500 600 700 800 900 1000 time, T , increase the mean packet E2E delay, particu- max M2M devices larly in the low loaded scenarios. However, the effect reduces with the larger values of N . For instance, in the Fig. 12 A comparison of packets mean E2E delay low loaded scenario when N ¼ 100, the given values of maximum waiting time introduce a mean E2E delay of 0.010 0.0045, 0.0070, and 0.0095 ms, respectively. Since the Tmax = 05 ms 0.009 effect of timer on the multiplexing process decreases due to Tmax = 10 ms 0.008 Tmax = 15 ms the increasing arrival rate, the mean E2E delay values are 0.007 reduced to 0.00368, 0.00416, and 0.00419, respectively, for 0.006 N ¼ 1000. Furthermore, it is noted that the values of mean 0.005 E2E delay for T of 10 and 15 ms are almost similar. max 0.004 Moreover, the impact of timer expiry completely vanishes 0.003 when input traffic load is increased beyond 1000, as dis- 0.002 cussed earlier in Figs. 10 and 11. 0.001 0.000 100 200 300 400 500 600 700 800 900 1000 M2M devices 7 Conclusion Fig. 13 A comparison of packets mean E2E delay The latest 3GPP LTE-Advanced networks have primarily focused on dominating non-cellular technologies to support affect on PRB utilization and mean E2E delay in high M2M applications also in the future. Since automotive loaded scenarios in the multiplexing process, as discussed applications and logistical processes demand wide cover- earlier in Figs. 10 and 11. In general, the mean value of ages, increased data rates, high reliability, and low costs, used PRBs decreases and E2E delay increases for larger LTE-Advanced is considered as the ready-to-use technol- values of the timer expiry. Additionally, this trade-off is ogy to fulfill the future M2M service requirements. Nev- heavily dependent on the arrival rate, as discussed earlier. ertheless, the major challenge is to support the myriad of In this work, 10 scenarios are simulated in order to devices sending small-sized data and thus inefficiently evaluate the impact of timer expiry on the multiplexing utilizing radio resources (PRBs). Since the spectrum for process. In the first scenario, 100 M2M devices are con- mobile networks will remain a scarce resource, new con- sidered to generate input traffic. The number of devices is cepts for new traffic type demand careful planning and incremented by 100 in the subsequent scenarios. The above evaluation. In this paper, the data multiplexing scheme is scenarios are simulated for the timer expiry values of 5, 10, proposed to overcome the risk of inefficient PRB utiliza- and 15 ms. The simulation results of mean number of used tion for mobile M2M traffic. Our results show that PRBs and E2E delay are compared for the given values of approximately 40 % more M2M devices in ITS and maximum waiting time. Figures 12 and 13 compare the logistics are served under the present system settings as simulation results of mean PRB utilization and E2E delay, compared to the state of the art without the use of multi- respectively. Figure 12 shows that less PRBs are used in plexing approach. Additionally, the QoS provisioning of the case of larger values of T , when the arrival rate is the M2M traffic is ensured by considering an expiry timer max kept constant. For instance, in the first scenario with which minimizes the multiplexing delays especially in low N ¼ 100, average numbers of PRBs used are 0.48, 0.27, loaded scenarios. More importantly, approximately 60 % and 0.20 for the timer expiry values of 5, 10, and 15 ms, of PRB utilization is improved for a maximum timer expiry respectively. It shows that for the larger values of maxi- value of 15 ms especially in low loaded scenarios. Thus, mum waiting time, less number of PRBs is used on the proposed scheme can dramatically reduce the network Mean PRBs used Mean E2E delay [s] Logist. Res. (2016) 9:15 Page 11 of 11 15 16. Mir ZH, Filali F (2014) Lte and ieee 802.11 p for vehicular load by sharing PRBs among several devices and can be a networking: a performance evaluation. EURASIP J Wirel Com- very useful scheme for network operators as well as service mun Netw 2014(1):1–15 providers. 17. Chou CM, Li CY, Chien WM, Lan KC (2009) A feasibility study on vehicle-to-infrastructure communication: Wifi vs. wimax. In: Compliance with ethical standards Tenth international conference on mobile data management: systems, services and middleware, 2009. MDM’09. IEEE, Conflict of interest The authors have no conflict of interest in this pp 397–398 manuscript. 18. Trichias K, Berg VDJ, Heijenk G, Jongh DJ, Litjens R (2012) Modeling and evaluation of LTE in intelligent transportation systems. In: Joint ERCIM eMobility and MobiSense Work- Open Access This article is distributed under the terms of the shop, Santorini, Greece, pp 48–59 Creative Commons Attribution 4.0 International License (http://crea 19. Mehmood Y, Goerg C, Muehleisen M, Timm-Giel A (2015) tivecommons.org/licenses/by/4.0/), which permits unrestricted use, Mobile M2M communication architectures, upcoming chal- distribution, and reproduction in any medium, provided you give lenges, applications, and future directions. EURASIP J Wirel appropriate credit to the original author(s) and the source, provide a Commun Netw 2015:1–37 link to the Creative Commons license, and indicate if changes were 20. Greenwood DA, Dannegger C, Dorer K, Calisti M (2009) made. Dynamic dispatching and transport optimization-real-world experience with perspectives on pervasive technology integra- tion. In: HICSS, pp 1–9 References 21. Machine to Machine (M2M)—Deutsche Telekom; Transport and Logistics. https://m2m.telekom.com/vertical-industries/transport- 1. Zhang Y, Yu R, Nekovee M, Liu Y, Xie S, Gjessing S (2012) and-logistics. Accessed 1 Sept 2015 Cognitive machine-to-machine communications: visions and 22. Lee HL, Whang S (2004) E-business and supply chain integra- potentials for the smart grid. IEEE Netw 26(3):6–13 tion. In: The practice of supply chain management: where theory 2. Taleb T, Kunz A (2012) Machine type communications in 3GPP and application converge. Springer, pp 123–138 networks: potential, challenges, and solutions. IEEE Commun 23. Chen Y, Wang W (2010) Machine-to-machine communication in Mag 50(3):178–184 LTE-A. In: IEEE 72nd vehicular technology conference fall 3. Atzori L, Iera A, Morabito G (2010) The internet of things: a (VTC 2010-Fall). IEEE, pp 1–4 survey. Comput Netw 54(15):2787–2805 24. ERICSSON Mobility Report on the pulse of the networked 4. Exalted: Expanding LTE for Devices. http://www.ict-exalted.eu/ society. Technical report, ERICSSON (2014) fileadmin/documents/EXALTED_WP2_D2.1.pdf. Accessed 30 25. 3GPP LTE-Advanced. http://www.3gpp.org/technologies/key Sept 2013 words-acronyms/97-lte-advanced. Accessed 28 May 2015 5. Towards 50 billion connected devices. Technical report, 26. 3GPP: Technical specification group radio access network; ERICSSON (2010) evolved universal terrestrial radio access (E-UTRA); relay 6. ETSI: Machine-to-Machine communications (M2M); Use Cases architectures for E-UTRA (LTE-Advanced). Technical report, of automotive applications. In: M2M capable networks. Techni- TR 36.806 V9.0.0 (2010) cal report, TR 102 898 V1.1.1 (2013) 27. Mehmood Y, Afzal W, Ahmad F, Younas U, Rashid I, Mehmood 7. Slicing up the M2M revenue pie; How to get your share and boost I (2013) Large scaled multi-user mimo system so called massive your business. Technical report, COMARCH (white paper) mimo systems for future wireless communication networks. In: 8. Vodafone M2M Adoption Barometer. Technical report, Voda- 2013 19th international conference on automation and computing fone (2015) (ICAC). IEEE, pp 1–4 9. The state of M2M in transport and logistics; Key findings from 28. Chandrasekhar V, Andrews JG, Gatherer A (2008) Femtocell the 2013 Vodafone M2M Adoption Barometer. Technical report, networks: a survey. IEEE Commun Mag 46(9):59–67 Vodafone (2014) 29. Cox C (2012) An introduction to LTE: LTE, LTE-advanced, SAE 10. NOKIA: LTE M2M: optimizing LTE for the internet of things. and 4G mobile communications. Wiley, New york Technical report, Nokia white paper (2014) 30. METIS: Novel radio link concepts and state of the art analysis. 11. Laner M, Svoboda P, Nikaein N, Rupp M (2013) Traffic models Technical report, Deliverable Number ICT-317669-METIS/D2.2 for machine type communications. In: Proceedings of the tenth (2013) international symposium on wireless communication systems 31. Mehmood Y, Marwat SNK, Po¨tsch T, Ahmad F, Go¨rg C, Rashid I (ISWCS). VDE, pp 1–5 (2014) Impact of M2M traffic on LTE data traffic performance. 12. Shafiq MZ, Ji L, Liu AX, Pang J, Wang J (2012) A first look at In: LogDynamics international conference. Springer (in-press) cellular machine-to-machine traffic: large scale measurement and 32. Opnet Modeler Technologies. https://www.opnet.com/services/ characterization. In: ACM SIGMETRICS performance evalua- modeling-simulation-service.html. Accessed 10 Jul 2013 tion review, vol 40. ACM, pp 65–76 33. Cavers JK (2000) Mobile channel characteristics. Springer, 13. Mehmood Y, Nawaz Khan Marwat S, Grg C, Zaki Y, Timm-Giel Berlin A (2015) Evaluation of M2M data traffic aggregation in LTE-A 34. IEEE 802.16p-11/0014: IEEE 802.16p Machine to Machine uplink. In: ITG/VDE mobile communication conference. VDE, (M2M) Evaluation Methodology Document (EMD). Technical pp 24–29 report, IEEE 802.16 Broadband Wireless Access Working Group 14. ETSI: Machine-to-Machine communications (M2M), Functional (2010) architecture. Technical report, ETSI TS 102 690 V2.1.1 (2013) 15. IEEE, Wireless Access in Vehicular Environments. http://stan dards.ieee.org/findstds/standard/802.11p-2010.html. Page Acces- sed 21 Nov 2015 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Logistics Research Springer Journals

Loading next page...
 
/lp/springer-journals/m2m-potentials-in-logistics-and-transportation-industry-qxS5D6q0sy
Publisher
Springer Journals
Copyright
Copyright © 2016 by The Author(s)
Subject
Engineering; Engineering Economics, Organization, Logistics, Marketing; Logistics; Industrial and Production Engineering; Simulation and Modeling; Operation Research/Decision Theory
ISSN
1865-035X
eISSN
1865-0368
DOI
10.1007/s12159-016-0142-y
Publisher site
See Article on Publisher Site

Abstract

Logist. Res. (2016) 9:15 DOI 10.1007/s12159-016-0142-y ORIGINAL PAPER 1 2 1 • • • Yasir Mehmood Safdar Nawaz Khan Marwat Koojana Kuladinithi 1 3 1 4 • • • Anna Fo ¨ rster Yasir Zaki Carmelita Go ¨ rg Andreas Timm-Giel Received: 15 December 2015 / Accepted: 16 July 2016 / Published online: 28 July 2016 The Author(s) 2016. This article is published with open access at Springerlink.com Abstract All over the world, road congestion is among the scheme by using Long-Term Evolution Advanced (LTE- most prevalent transport challenges usually in urban envi- Advanced) Relay Nodes, which aggregates M2M traffic to ronments which not only increases fuel consumption and maximize radio resource utilization. Extensive system- emission of harmful gases, but also causes stress for the level simulations are performed using an LTE-Advanced- drivers. Intelligent Transportation System (ITS) enables a based model developed in the RIVERBED modeler to better use of the infrastructure by connecting vehicles to evaluate the performance of the proposed data multiplexing other vehicles as well as infrastructure and thus delivers a scheme. Simulation results show that approximately 40 % faster communication opportunity to ensure safe and secure more smart M2M devices used in ITS and logistics are driving. Machine-to-machine (M2M) communication is served per LTE-Advanced cell under the present system one of the latest information and communication tech- settings. nologies which offers ubiquitous connectivity among sev- eral smart devices. The use of mobile (cellular) M2M Keywords Road congestion  Intelligent transportation communications has emerged due to the wide range, high systems  Information and communication technologies reliability, increased data rates, decreased costs as well as Machine to machine  Relay nodes  LTE-Advanced easy and short-term deployment opportunities. Since the radio spectrum is a scarce resource, M2M traffic can potentially degrade the performance of mobile networks 1 Introduction due to the large number of devices sending small-sized data. This paper presents an efficient data multiplexing The dramatic use of communication technologies (wired and wireless), embedded systems as well as increasing penetration of the Internet has not only revolutionalized This article is part of a focus collection on ‘‘Dynamics in Logistics: human lives, but also reshaped almost all types of business Digital Technologies and Related Management Methods.’’ & Yasir Mehmood Andreas Timm-Giel ym@comnets.uni-bremen.de timm-giel@tuhh.de Safdar Nawaz Khan Marwat Communication Networks, University of Bremen, Bremen, safdar@uetpeshawar.edu.pk Germany Koojana Kuladinithi Department of Computer Systems Engineering, University of koo@comnets.uni-bremen.de Engineering and Technology, Peshawar, Peshawar, Pakistan Anna Fo¨rster Computer Science Department, New York University Abu anna.foerster@comnets.uni-bremen.de Dhabi (NYUAD), Abu Dhabi, UAE Yasir Zaki Institute of Communication Networks, University of yasir.zaki@nyu.edu Technology, Hamburg, Germany Carmelita Go¨rg cg@comnets.uni-bremen.de 123 15 Page 2 of 11 Logist. Res. (2016) 9:15 Energy and Ulies models and processes [1]. M2M communication is one of Adopon of M2M technology in industry – 2015 Automove the emerging technologies which offers ubiquitous con- Retail nectivity among intelligent devices, hence is one of the Consumer electronics 17% 19% 37% major enablers of the Internet-of-Things (IoTs) vision [2]. e-healthcare 28% 32% IoT is an innovative concept which offers to connect smart Logiscs 29% 32% Manufacturing devices often called things endowed with several sensing, automation as well as computing capabilities, with the Internet [3]. Resultantly, the connected devices are revo- Fig. 1 An illustration of leading industries adoption M2M technol- lutionalizing the future cyber physical systems, yielding ogy worldwide, based on the Vodafone report 2015 [8] several applications. Moreover, the mobile network oper- ators are partnering with industrial organizations in order to Besides, it also exhibits unique characteristics such as an bring forth innovative IoT services to facilitate end con- increasing device volume, sending small payloads, the sumers. For instance, M2M applications include intelligent demand for various mobility profiles, time-controlled, and transportations, logistics and supply chain management, mainly delay tolerant. Since the spectrum for mobile net- e-health, smart metering, surveillance and security, smart works will remain a scarce resource, efficient utilization of cities, and home automation [4–6]. Thus, M2M commu- radio spectrum is one of its major requirements. Therefore, nication is foreseen to reshape the business of operators, the objective of this paper is to exploit LTE-Advanced RNs service providers, M2M enterprises, and M2M enablers [7]. to multiplex small-sized M2M data packets in order to Vodafone revealed that M2M communication is ensure efficient LTE-Advanced radio resource utilization becoming one of the driving forces for businesses which and thus to support a large number of devices. inspires to bring forth innovative solutions almost in every The rest of the paper is structured as follows. We firstly sector such as logistics, automotive industry, cities, homes, present an overview of the leading communication tech- schools, and workplaces [8]. Approximately 90 % of the nologies used in ITS and logistics, followed by a generic companies worldwide have adopted M2M technology and overview of the two latest technologies, i.e., the Institute of imparted it as one of the most favorable technologies for Electrical and Electronics Engineers (IEEE) 802.11p and achieving noticeable outcomes. Automotive industry is one 3GPP LTE-Advanced in Sect. 2. An overview of major of the top sectors for adopting M2M technology. Approx- M2M services in ITS and logistics are discussed in Sect. 2. imately 32 and 17 % increasing growth rates for adopting Then, we discuss problem definitions by highlighting how M2M technology have been noticed in automotive and LTE-Advanced resources can be used inefficiently by logistic sectors, respectively, as shown in Fig. 1. Thus, ITS smart M2M devices in Sect. 3. Section 4 presents the and logistics are considered as one of the potential M2M proposed uplink M2M data multiplexing scheme [13]. The users worldwide [9]. In addition, NOKIA forecasted that simulation environment and parameter settings are pre- the use of M2M technology in automotive industry and sented in Sect. 5. We discuss our simulation results in logistical processes will dominate other applications in the Sect. 6. In the end, conclusions are drawn in Sect. 7.In future [10]. One of the major motivations is to deliver a addition, a list of most frequently used acronyms is pre- fully managed infrastructure which primarily guarantees, sented in Table 1. e.g., safe and secure driving, in time delivery, smart monitoring, and tracking of assets. Resultantly, this can revolutionize the existing methods of transportation and 2 Mobile M2M communications freight movements. In addition, optimum system perfor- mance can be achieved by reducing factor of costs, pol- This section presents an overview of ETSI (European lution and emission of harmful gases. Telecommunications Standards Institute) M2M architec- Mobile M2M communication greatly differs from tra- ture followed by the major M2M use cases and services in ditional human-to-human (H2H) communication in terms transportation and logistics. of traffic density, data packet size, and quality of service (QoS) requirements [11]. For instance, an experimental 2.1 ETSI M2M architectural overview study done in [12] shows that M2M traffic exhibits a sig- nificantly different behavior than the traditional smart- The high-level ETSI M2M network architecture is shown phone traffic in various aspects. For example, unlike in Fig. 2 [14]. The major components of mobile M2M traditional mobile traffic, M2M is an uplink dominant communication architecture include the device, communi- traffic which particularly generates bursty traffic volumes. cation as well as server domains. The primary functionality of the device domain is to collect and send sensor data such Sending information from M2M devices to base station. 123 Logist. Res. (2016) 9:15 Page 3 of 11 15 Table 1 List of used abbreviations technology for Vehicular Ad hoc Networks (VANET). IEEE 802.11p wireless access introduces minimum delay Abbreviation Acronym in ITS. However, its employment is limited due to its MAC Medium Access Control decentralized nature. The recent research has revealed that MME Mobility Management Entity maximum operating efficiency of IEEE 802.11p can be PHY Physical Layer achieved by supporting it with the LTE-Advanced mobile PRB Physical Resource Block networks. The main features of LTE-Advanced technology QCI QoS Class Identifiers include wide availability of modules/devices, system RLC Radio Link Control existence, decreased costs as well as easy deployment. RRC Radio Resource Control Since the cellular modules and sensors are easily available, e.g., in vehicles, the applications of LTE-Advanced mobile SAE System Architecture Evolution networks are dramatically increasing in automotive sector. S-GWs Serving Gateways The authors in [16–18] compared the performance of SGSN Serving GPRS Support Node two different technologies under varying channel and TTI Transmission Time Interval traffic load conditions. Moreover, several issues due to WLAN Wireless Local Area Network decentralized nature of IEEE 802.11p are highlighted in WPAN Wireless Personal Area Network [16]. For instance, the main shortcomings are less relia- WSN Wireless Sensor Network bility due to uncoordinated Medium Access Control (MAC) procedures, risk of network congestion, higher as the internal temperature and humidity level of a con- vehicle mobility, and low scalability. Moreover, the num- tainer, position and speed of a vehicle, and fuel con- ber of network-connected vehicles adversely affect the sumption. The role of the communication network is to performance of the standard system. On the other hand, create a communication path between the devices and LTE-Advanced networks are more reliable, scalable, and servers through either wired or wireless networks such as capable of supporting higher density of vehicles. In addi- Digital Subscriber Line (DSL) and cellular networks (e.g., tion, the authors analyzed the performance of the above LTE-Advanced), respectively. Finally, the server domain technologies in terms of delay, reliability, scalability, and consists of a middleware layer where the collected packets mobility support. The authors concluded that 3GPP LTE- go through several application services and later are used Advanced systems are more efficient in terms of high by related agencies. Thus, M2M technology employs scalability and mobility support than IEEE 802.11p stan- wired, wireless, and hybrid communication opportunities dard. However, it is further concluded that the performance among devices to ensure a fully automatic acquisition, of IEEE 802.11p is more sensitive during increased den- processing, and transmission of data. Thus, M2M repre- sity, traffic load, and higher mobility. sents a broad next-generation technology that is primarily incorporated in modern automobile industries to improve 2.2 M2M use cases the ease, safety, and quality of human life. The wired access technologies provide less delay, high throughput Mobile M2M communication offers manifold applications and are more reliable. Despite the aforementioned capa- and services in modern transport and logistical processes bilities, wired systems cannot be used in transport and such as onboard security, traffic and infrastructure man- logistical systems due to several limitations such as lack of agement, fleet management, and route planning [19]. In scalability, cost efficiency, and mobility. However, above- case of an emergency, the collected data are sent to other mentioned limitations can be overcome by incorporating vehicles as well as infrastructure to gain immediate atten- wireless (cellular) technologies to achieve maximum sys- tion. To avoid further incidents, communication between tem efficiency and reliability. the infrastructure and the vehicles must be very fast to Recently, the key emerging wireless technologies used detect emergency messages and deliver warning messages in modern transport and logistical systems include IEEE immediately. Similarly, traffic and infrastructure manage- 802.11p [15] (standardized to support vehicular commu- ment play a prominent role in handling the problem of road nication) and 3GPP LTE-Advanced mobile networks [16]. congestion. It tackles the problem by providing two-way IEEE 802.11p is achieved by making few advancements in communication opportunities between vehicles and well-known 802.11 Wireless Local Area Network (WLAN) infrastructure. Vehicles can send status updates about the technology. It is also considered as the most feasible position, speed, fuel consumption, and delivery status reports to the infrastructure and can also receive relevant instructions about road accidents and emergency braking 3rd Generation Partnership Project—3GPP—is the leading stan- system [20]. Moreover, M2M communications support dardization organization for mobile networks. 123 15 Page 4 of 11 Logist. Res. (2016) 9:15 Backend server (e.g., transportaon control center) Gateway DeNB: Donor eNB Downlink: Mobile communicaon Uplink: domain (e.g., LTE- Advanced) DeNB DeNB DeNB DeNB Applicaon domain Smart sensors deployed to (e.g., ITS and monitor temperature, humidity, pressure, power, water, etc. logiscs) Inventory Downlink informaon Vehicles cross talk for on-board Warehouse management Accidental/emergency security, status update, fleet (monitoring all sensors deployed unicast/broadcast to uplink informaon management, etc. for various metering operaons) all devices (vehicles) Fig. 2 ETSI mobile M2M communication architecture along with an overview of major M2M applications in intelligent transportation and logistics, based on [14] several operations such as tracking of a stolen vehicle, In addition to the above applications, M2M communi- traffic reports, and route planning as well as infotainment cation provides additional services in logistics such as services [6]. For instance, to recover a stolen vehicle, SVT decreased operational cost, high inventory flexibility, (Stolen Vehicle Tracking) service providers request data increased supply chain visibility, and reduced loss of about the location from Telematic Control Unit (TCU) vehicles and containers [21]. In supply chains, M2M located inside the vehicle. In addition, drivers are also technology enables tracking the status of goods in real-time updated by sending reports regarding traffic in a particular via M2M devices. This increasing visibility allows for region so that they can change or plan new routes in case of significant reduction in uncertainties in supply chain [22]. traffic jam or an emergency. Furthermore, infotainment Similarly, in a warehouse, M2M devices can be deployed services aim to provide news/information to drivers and to track the inventory so that stockholders and enterprises passengers through mobile TV, web-browsing, etc. can respond to the market dynamics and to decide when to Fleet management is also one of the major M2M refill and when to go on sale. Moreover, cross talk among applications in logistics [9]. The movements of vehicles, vehicles can also be effective to get immediate assistance. containers, buses, and cars are being tracked regularly Additionally, direct delivery of inventory from one vehicle through devices which collect data of the location, vehicle to another without storing it in a warehouse can also be speed, temperature, distribution progress, fuel consumption accomplished through mutual information sharing. Con- and send this information to monitoring servers. Through sequently, it can significantly reduce the required space of regular monitoring, several activities of the system can be warehouse, customer’s waiting time as well as the opera- performed in an efficient way. For instance, the goods tional costs for business entities. which are transported from one place to another are mon- itored regularly in order to accomplish in time delivery and to handle any undesirable situation during shipment pro- 3 Problem descriptions cesses. The cargo moves across several regions; therefore, it must be monitored in order to stay updated. Moreover, Mobile systems such as LTE-Advanced are mainly reporting gives the exact location of the freight, and thus designed to support the increasing traditional mobile traffic the conditions of the objects can be easily monitored. In by enabling improved broadband services [23]. For applications such as warehouse management, fleet man- instance, Ericsson in its mobility report [24] anticipates agement, robotics, and control systems, alarms are also approximately 5.6 billion cellular-based active smart- used to detect critical or emergency situations. phones by 2019. Therefore, in order to support such a 123 Logist. Res. (2016) 9:15 Page 5 of 11 15 100.00 Device payload (%) values, e.g., 20–26 representing favorable channel condi- 90.00 Zero padding (%) tions, the capacity of the PRB increases. However, small 80.00 packets with increasing percentage of padded zeros are 70.00 transmitted. This shows that more packets can be accom- 60.00 modated per PRB under favorable channel conditions in 50.00 40.00 the given exemplary scenarios. Since mobile radio 30.00 resources are valuable assets and scarcely available, it is 20.00 required to ensure efficient utilization of radio resources for 10.00 M2M communications. 0.00 0 2 4 6 8 101214161820222426 MCS Index 4 Proposed M2M data multiplexing Fig. 3 An example illustration of inefficient PRB utilization for M2M communication In the proposed data multiplexing scheme, the small-sized M2M data packets are multiplexed at the Packet Data massive mobile traffic, several enhancements in existing Convergence Protocol (PDCP) layer of the RN. A high- LTE systems [25] such as Relay Nodes (RNs) [26], mas- level illustration of E-UTRAN (Evolved UMTS Terrestrial sive Multiple Input–Multiple Output (MIMO) systems Radio Access Network) architecture with the functionali- [27], and femto cells [28] were introduced to fulfill the ties of LTE-Advanced relaying is depicted in Fig. 4. The IMT-Advanced (International Mobile Telecommunications RN communicates with devices using access link (Uu), Advanced) demands of maximum throughput of upto 1 whereas it communicates with DeNB using backhaul link Gbps in downlink and 500 Mbps in uplink [29]. On the (Un). Thus, it behaves like an eNB toward the M2M other hand, mobile standards are foreseen as an attractive devices, whereas it acts as a UE toward the DeNB. option to support future M2M applications such as in The RN PDCP is selected to aggregate M2M packets in transportation, logistics, smart city, and living [2]. Several order to maximize the multiplexing gain. This increasing forecasts have also reported a considerable market growth multiplexing gain is achieved without aggregating the for M2M device volume, making quality of service pro- additional overheads such as those from the PDCP, Radio visioning a huge challenge [10]. Consequently, mobile Link Control (RLC), and Medium Access Control (MAC). networks will not be able to support the increasing number A high-level illustration of the proposed scheme is given in of M2M devices worldwide [30]. Fig. 5. Generally, the PDCP layer is a part of LTE air According to 3GPP, the smallest unit of the radio interface control and user plane protocols. It exists in the spectrum allocatable to a single device is 1 PRB which is UE, Donor eNB (DeNB), and in the RN. The major capable of transmitting several hundred bits under favor- functionalities of PDCP layer for the user plane include able channel conditions. However, allocating 1 PRB to a header compression and decompression, user data transfer, single M2M device used in transportation/logistics could delivery of upper layer Packet Data Units (PDUs) in significantly degrade radio spectrum utilization sequence, and retransmission of PDCP Service Data Units [13, 19, 31]. This is due to the fact that the capacity of a (SDUs). Moreover, the control plane services include PRB can be much higher than the actual size of the device ciphering and integrity protection and transfer of control payload under favorable channel conditions. Figure 3 plane data. The proposed multiplexing scheme can also be illustrates how inefficiently a PRB can be utilized when a applied at the lower layers of the RN such as RLC and device sends small-sized packets when experiencing good MAC on the Uu interface or in the PDCP of Un interface. channel conditions. An exemplary device payload of 4 However, in these cases, the multiplexing gain will be bytes is considered in order to evaluate the utilization of a reduced due to the multiplexing of additional headers along PRB against all the possible values of the Modulation and with the original small-sized data packets [13]. Coding Scheme (MCS). A PRB utilization efficiency is A multiplexing buffer is created at the RN PDCP layer evaluated in terms of percentage of the actual payload and which multiplexes the incoming packets according to the the padded zeros with respect to MCS index. It can be seen available service rate which is equal to Transport Block from Fig. 3 that with the lower MCS values depicting Size (TBS)—RN Un protocol overheads in terms of bits unfavorable channel conditions, the PRB capacity is per TTI, as shown in Fig. 5. The Un protocol overhead smaller, and therefore, it is fully utilized to transmit the includes 12, 28, and 4 bytes for GTP (GPRS Tunneling given payload. On the other hand, with the higher MCS Protocol), UDP (User Datagram Protocol)/IP, and layer 2, respectively. The multiplexed packet is sent to the RN GTP over the Un link. The additional overheads such as from Physical Resource Block. Ulizaon/PRB in % 15 Page 6 of 11 Logist. Res. (2016) 9:15 Applicaon server Serving Overcoming gateway shadowing problems (S-GW) (DeNB) (RN) (M2M) Un Uu Un Uu Un Uu Hotspot coverage for ITS and logiscal operaons Coverage extensions Fig. 4 An illustration of LTE-Advanced relaying technology. Re-drawn based on [26]) IP Applicaon GTP GTP GTP GTP MUX TCP/UDP UDP UDP UDP UDP IP IP IP IP λ λ λ 1 2 λ IP PDCP PDCP PDCP PDCP RLC RLC RLC RLC RLC Layer 2 MAC MAC MAC MAC MAC PHY PHY PHY PHY PHY PHY M2M device RN DeNB aGW Fig. 5 A high-level illustration of the proposed multiplexing scheme along with protocols stack of the M2M device, RN, and DeNB the GTP, UDP/IP, PDCP, and RLC are added. The block low loaded scenario, there is a comparatively longer delay diagram of the data differentiating algorithm at the DeNB due to the low arrival rate. Consequently, the performance is given in Fig. 6. From the physical layer of the RN, the of delay sensitive M2M applications can be degraded. To multiplexed packets are sent to the DeNB over the Un tackle this issue, an expiry timer, T is introduced. The max interface, as depicted in Fig. 5. timer is set with a fixed value such as 10 ms in the current The multiplexing approach significantly improves PRB implementation. This means that the buffer serves the utilization. However, there are certain constraints regarding multiplexed packet after 10 ms at the latest. The value of latency requirements of high priority M2M traffic such as the timer could also be adaptive, i.e., it can change its value accidental information, emergency alerting, and e-health. according to priorities of the incoming packets. For this This is due to the fact that each packet waits until the size purpose, the algorithm must be fully aware of various of the buffer B is equal to the available TBS—RN Un priorities of M2M applications. protocol overheads. In a highly loaded scenario, the wait- In this work, a fixed RN is implemented and used to ing time is not long due to a high arrival rate. However, in a multiplex the uplink traffic from M2M devices deployed 123 Logist. Res. (2016) 9:15 Page 7 of 11 15 Start Router Packet arrival at the PDCP layer DeNB of the DeNB Remote Access Backhaul link Access link server gateway RN Identify the destination IP of the X2 received packet If False The received packet belongs (incoming packet IP == to the standalone user DeNB IP) M2M devices True Send the packet to GTP-U Fig. 7 Project editor of the LTE-Advanced-based model developed in the RIVERBED Modeler simulator The received packet belongs to the RN • The overall service rate is l ¼ l  5ðPRBsÞ b/s. Send to GTP-U via IP and PRB UDP • So, the system utilization q can be given as Arrival rate k Utilization; q ¼ ¼ ð1Þ De-tunnel to extract the Service rate l original transmitted standalone packets N  656 Utilization; q ¼ ð2Þ • Moreover, the maximum number of devices served can Tunnel each packet again be given as follows, and route towards aGW q  321600 N ¼ ð3Þ max end Fig. 6 Block diagram of the data differentiating algorithm at the DeNB 5 Simulation model and parameters in ITS and logistics (i.e., for onboard security and The Optimized Network Engineering Tool (OPNET) metering purposes). The position of the RN corresponds Modeler which is newly named as RIVERBED Modeler is to an MCS of 16, whereas 5 PRBs are allocated to RN to used as a primary modeling, simulation, and analysis tool serve M2M traffic. The given values of MCS and PRBs for this research. It provides a simulation environment for correspond to TB size of 1608 bits per TTI. Since a TTI the performance measurements of communication net- is of 1-ms duration, the TB size can also be given as works [32]. The project editor of the LTE-Advanced-based 1608000 bits/s. According to the 3GPP standardizations, model developed in the simulator with several nodes along the capacity of a single PRB varies according to the with LTE-Advanced functionalities and protocols is shown MCS. Therefore, under the given values of MCS and in Fig. 7. Additionally, an RN is implemented and placed PRBs, 321 bits per TTI or 321000 bits/s can be sent within the coverage of DeNB to relay and multiplex M2M within a single PRB. In general, the system utilization, q data traffic. The DeNB is responsible to connect UEs/M2M can be determined in terms of the arrival rate, k and the devices with the transport network, thus includes both radio service rate, l. Additionally, the maximum number of interface and transport protocols to communicate with served devices, N can be determined according to the max UEs/M2M devices and the core network, respectively. The following procedure, radio protocols traffic coming from aGW toward the RN. The remote server and the aGW (access gateway) are • The size of the each transmitted M2M packet per interconnected with an Ethernet link with an average delay second is P ¼ 656 bits. size of 20 ms. Moreover, aGW acts as the tunneling point for • The arrival rate k due to N number of devices is the downlink and uplink traffic coming from PDN-GW and ðN  656Þb=s. DeNB, respectively. The aGW node protocols include the • The service rate per PRB is l ¼ 321000 b/s. PRB 123 15 Page 8 of 11 Logist. Res. (2016) 9:15 Table 2 Simulation parameters work. Later, we discuss our simulation results in detail with varying load conditions. In this work, we consider M2M Parameters Values devices used in ITS and logistics in order to generate Simulation length 1000 s uplink traffic. eNB coverage radius 350 m Min. eNB UE distance 35 m 6.1 Simulation scenarios description Max terminal power 23 dBm Terminal speed 120 km/h The scenarios are simulated according to three major cat- Mobility model Random way point (RWP) egories, i.e., no multiplexing, multiplexing without timer, Frequency reuse factor 1 and multiplexing with timer. In the first category, M2M data packets are relayed in uplink without multiplexing. In Transmission bandwidth 5 MHz the second category, the data packets from all the active No. of PRBs 25 devices are multiplexed at the RN before being sent to the MCS QPSK, 16QAM, 64QAM, DeNB. However, in this category, no expiry timer is con- Channel models Pathloss, slow fading, fast fading sidered to control the multiplexing process, and the mul- Path loss 128:1 þ 37:6 log ðRÞ, R in km tiplexed packet is served when its size is large enough to Slow fading Log-normal shadowing, correlation 1, deviation 8 dB utilize full capacity of the available spectrum. In the third Fast fading Jakes-like method [33] category, an expiry timer is introduced in order to limit the RN parameters multiplexing delay especially in the low loaded scenarios. In this case, the multiplexed packet is served after T at PRBs for RN 5 max the latest. Each category is further divided into several Corresponding MCS 16 scenarios. The scenarios are simulated for both load con- TBS 1608 bits ditions (i.e., low and high). In the low loads, the number of Simulated scenarios No multiplexing, multiplexing without timer, and multiplexing devices, N, is kept 200 in the first scenario. The value of with timer N is incremented by 200 in the subsequent scenarios till the Timer expiry values 10 ms (Sect. 6.2) limiting case, i.e., when all 5 PRBs are fully utilized. In Timer expiry values 5, 10 and 20 ms (Sect. 6.3) high loads, 2800 devices are placed in the first scenario, Type of RN Fixed and the number is also incremented by 200 in the subse- M2M Traffic model [34] quent scenarios. Message size 38 bytes (constant) at the device PDCP Inter-send time 1 s (constant) 6.2 Simulation results The simulation results for the mean number of used PRBs Internet Protocol (IP) and Ethernet. The aGW and DeNB with 95 % confidence interval (CI) are illustrated in Figs. 8 nodes (names as eNB1..) communicate through IP routers and 9 for the low and high load conditions, respectively. (R1..). QoS parameters at the Transport Network (TN) The values of the upper and lower bound of CI are very ensure QoS parameterization and traffic differentiation. small in most of the scenarios as shown in Figs. 8 and 9. The user movement in a cell is emulated by the mobility These PRBs are used by RN to transmit the multiplexed model by periodically updating the location of the user. data toward the DeNB. The user mobility information is stored in the global user The simulation results clearly show the efficient uti- database (Global-UE-List). The channel model parameters lization of PRBs in uplink with the proposed multiplexing scheme. For instance, in the no multiplexing scenario with for the air interface include path loss, slow fading, and fast fading models. In this paper, the simulation modeling 400 devices, the arrival rate at the PHY layer of the RN is mainly focuses on the user plane to perform E2E perfor- 262.4 bit/TTI which almost utilize 1 PRB. However, in the mance evaluations. Furthermore, Table 2 presents a case of multiplexing, only half of the PRBs are used to detailed description of simulation parameters and settings. serve the 400 devices, see Fig. 8. Similarly, without mul- tiplexing, the RN serves nearly 2400 devices with 5 PRBs in uplink, which is actually the limiting case as shown in 6 Results and discussion 5 PRBs are allocated to RN in order to serve M2M devices. These This section investigates the performance of the proposed PRBs are fully utilized when the number of devices is raised to 2600 multiplexing scheme in the low and high loaded scenarios. without multiplexing. So the high load scenarios start with 2800 We firstly describe all scenarios which are simulated in this devices. 123 Logist. Res. (2016) 9:15 Page 9 of 11 15 5.00 0.03 No mulplexing No mulplexing 4.50 Mulplexing without mer Mulplexing without mer 4.00 Mulplexing with mer Mulplexing with mer 3.50 0.02 3.00 2.50 2.00 0.01 1.50 1.00 0.50 0.00 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 0.00 M2M devices 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 M2M devices Fig. 10 A comparison of packet mean E2E delay in low loaded scenarios. T ¼ 10 ms max Fig. 8 A comparison of mean PRBs used with and without data multiplexing in low loaded scenarios. T ¼ 10 ms max 0.03 5.00 4.50 No mulplexing Mulplexing without mer 4.00 0.02 Mulplexing with mer 3.50 3.00 No mulplexing 0.01 2.50 Mulplexing without mer 2.00 Mulplexing with mer 1.50 0.00 1.00 2800 3000 3200 3400 3600 3800 4000 4200 4400 0.50 M2M devices 0.00 2800 3000 3200 3400 3600 3800 4000 4200 4400 Fig. 11 Comparing packet mean E2E delay in high loaded scenarios M2M devices with timer expiry, T ¼ 10 ms max Fig. 9 Comparing simulation results of mean PRBs used with and without data traffic multiplexing in high loaded scenarios. delay is higher in the case of multiplexing for low loads as T ¼ 10 ms max shown in Fig. 10. This increase in delay time is due to less arriving packets especially in the low loaded scenarios. The buffer aggregates packets until its size plus the additional Fig. 8. However, in case of multiplexing, the number of devices served by the RN nearly doubles, i.e., approxi- Un overhead is equal to the available TBS. However, the use of an expiry timer limits the delay by serving the mately 4200 devices are served as depicted in Fig. 9.This is due to the fact that in the case of no multiplexing, each aggregated packets at the latest after 10 ms. On the other hand, in high loaded scenarios, the E2E delay is slightly data packet contains an additional Un air interface over- head of GTP, UDP, IP, and layer 2. The additional over- higher compared to the case of no multiplexing, see Fig. 11. This is due to high arrival rate and the buffer head causes an extra PRB usage and overall reduces the PRB utilization efficiency. Moreover, in low loaded sce- aggregates the incoming packets to make a large aggre- narios such as with 200 devices (see Fig. 8), the average gated packet within less time. Moreover, the value of the number of PRBs used is slightly higher in the case of E2E delay is very large in fully loaded scenarios as multiplexing with timer. This is due to the fact that the RN depicted in Fig. 10, when the RN utilizes all 5 PRBs with the maximum capacity. serves the traffic at the latest after 10 ms, and thus, the PRB is not necessarily used with its maximum capacity due to 6.3 Impact of timer expiry on data multiplexing the low arrival rate. However, in high load scenarios, the timer has almost no impact, and a nearly equal number of In this section, we investigate the impact of timer expiry, PRBs is used with and without timer. Figures 10 and 11 show the simulation results of the T on mean PRB utilization and E2E delay by varying max input traffic. However, we only consider low loaded sce- mean packet E2E delay in low and high load scenarios, respectively. The results show that the value of packet E2E narios due to the fact that the timer expiry has almost no Mean PRBs used Mean PRBs used Mean E2E delay [s] Mean E2E delay [s] 15 Page 10 of 11 Logist. Res. (2016) 9:15 1.40 average, as it allows more packets to be multiplexed before Tmax = 05 ms timer expiry. Resultantly, it increases the multiplexing Tmax = 10 ms 1.20 Tmax = 15 ms gain. Additionally, it is further noted that larger values of 1.00 T has almost no affect on PRBs used in the case of high max 0.80 loads. For instance, when N ¼ 1000, the average values of PRBs used are 1.219 and 1.213 for the maximum waiting 0.60 time of 10 and 15 ms, respectively. 0.40 Similarly, Fig. 13 compares the simulation results of the 0.20 mean E2E delay for the given timer expiry values. As discussed in the beginning, the larger values of waiting 0.00 100 200 300 400 500 600 700 800 900 1000 time, T , increase the mean packet E2E delay, particu- max M2M devices larly in the low loaded scenarios. However, the effect reduces with the larger values of N . For instance, in the Fig. 12 A comparison of packets mean E2E delay low loaded scenario when N ¼ 100, the given values of maximum waiting time introduce a mean E2E delay of 0.010 0.0045, 0.0070, and 0.0095 ms, respectively. Since the Tmax = 05 ms 0.009 effect of timer on the multiplexing process decreases due to Tmax = 10 ms 0.008 Tmax = 15 ms the increasing arrival rate, the mean E2E delay values are 0.007 reduced to 0.00368, 0.00416, and 0.00419, respectively, for 0.006 N ¼ 1000. Furthermore, it is noted that the values of mean 0.005 E2E delay for T of 10 and 15 ms are almost similar. max 0.004 Moreover, the impact of timer expiry completely vanishes 0.003 when input traffic load is increased beyond 1000, as dis- 0.002 cussed earlier in Figs. 10 and 11. 0.001 0.000 100 200 300 400 500 600 700 800 900 1000 M2M devices 7 Conclusion Fig. 13 A comparison of packets mean E2E delay The latest 3GPP LTE-Advanced networks have primarily focused on dominating non-cellular technologies to support affect on PRB utilization and mean E2E delay in high M2M applications also in the future. Since automotive loaded scenarios in the multiplexing process, as discussed applications and logistical processes demand wide cover- earlier in Figs. 10 and 11. In general, the mean value of ages, increased data rates, high reliability, and low costs, used PRBs decreases and E2E delay increases for larger LTE-Advanced is considered as the ready-to-use technol- values of the timer expiry. Additionally, this trade-off is ogy to fulfill the future M2M service requirements. Nev- heavily dependent on the arrival rate, as discussed earlier. ertheless, the major challenge is to support the myriad of In this work, 10 scenarios are simulated in order to devices sending small-sized data and thus inefficiently evaluate the impact of timer expiry on the multiplexing utilizing radio resources (PRBs). Since the spectrum for process. In the first scenario, 100 M2M devices are con- mobile networks will remain a scarce resource, new con- sidered to generate input traffic. The number of devices is cepts for new traffic type demand careful planning and incremented by 100 in the subsequent scenarios. The above evaluation. In this paper, the data multiplexing scheme is scenarios are simulated for the timer expiry values of 5, 10, proposed to overcome the risk of inefficient PRB utiliza- and 15 ms. The simulation results of mean number of used tion for mobile M2M traffic. Our results show that PRBs and E2E delay are compared for the given values of approximately 40 % more M2M devices in ITS and maximum waiting time. Figures 12 and 13 compare the logistics are served under the present system settings as simulation results of mean PRB utilization and E2E delay, compared to the state of the art without the use of multi- respectively. Figure 12 shows that less PRBs are used in plexing approach. Additionally, the QoS provisioning of the case of larger values of T , when the arrival rate is the M2M traffic is ensured by considering an expiry timer max kept constant. For instance, in the first scenario with which minimizes the multiplexing delays especially in low N ¼ 100, average numbers of PRBs used are 0.48, 0.27, loaded scenarios. More importantly, approximately 60 % and 0.20 for the timer expiry values of 5, 10, and 15 ms, of PRB utilization is improved for a maximum timer expiry respectively. It shows that for the larger values of maxi- value of 15 ms especially in low loaded scenarios. Thus, mum waiting time, less number of PRBs is used on the proposed scheme can dramatically reduce the network Mean PRBs used Mean E2E delay [s] Logist. Res. (2016) 9:15 Page 11 of 11 15 16. Mir ZH, Filali F (2014) Lte and ieee 802.11 p for vehicular load by sharing PRBs among several devices and can be a networking: a performance evaluation. EURASIP J Wirel Com- very useful scheme for network operators as well as service mun Netw 2014(1):1–15 providers. 17. Chou CM, Li CY, Chien WM, Lan KC (2009) A feasibility study on vehicle-to-infrastructure communication: Wifi vs. wimax. In: Compliance with ethical standards Tenth international conference on mobile data management: systems, services and middleware, 2009. MDM’09. IEEE, Conflict of interest The authors have no conflict of interest in this pp 397–398 manuscript. 18. Trichias K, Berg VDJ, Heijenk G, Jongh DJ, Litjens R (2012) Modeling and evaluation of LTE in intelligent transportation systems. In: Joint ERCIM eMobility and MobiSense Work- Open Access This article is distributed under the terms of the shop, Santorini, Greece, pp 48–59 Creative Commons Attribution 4.0 International License (http://crea 19. Mehmood Y, Goerg C, Muehleisen M, Timm-Giel A (2015) tivecommons.org/licenses/by/4.0/), which permits unrestricted use, Mobile M2M communication architectures, upcoming chal- distribution, and reproduction in any medium, provided you give lenges, applications, and future directions. EURASIP J Wirel appropriate credit to the original author(s) and the source, provide a Commun Netw 2015:1–37 link to the Creative Commons license, and indicate if changes were 20. Greenwood DA, Dannegger C, Dorer K, Calisti M (2009) made. Dynamic dispatching and transport optimization-real-world experience with perspectives on pervasive technology integra- tion. In: HICSS, pp 1–9 References 21. Machine to Machine (M2M)—Deutsche Telekom; Transport and Logistics. https://m2m.telekom.com/vertical-industries/transport- 1. Zhang Y, Yu R, Nekovee M, Liu Y, Xie S, Gjessing S (2012) and-logistics. Accessed 1 Sept 2015 Cognitive machine-to-machine communications: visions and 22. Lee HL, Whang S (2004) E-business and supply chain integra- potentials for the smart grid. IEEE Netw 26(3):6–13 tion. In: The practice of supply chain management: where theory 2. Taleb T, Kunz A (2012) Machine type communications in 3GPP and application converge. Springer, pp 123–138 networks: potential, challenges, and solutions. IEEE Commun 23. Chen Y, Wang W (2010) Machine-to-machine communication in Mag 50(3):178–184 LTE-A. In: IEEE 72nd vehicular technology conference fall 3. Atzori L, Iera A, Morabito G (2010) The internet of things: a (VTC 2010-Fall). IEEE, pp 1–4 survey. Comput Netw 54(15):2787–2805 24. ERICSSON Mobility Report on the pulse of the networked 4. Exalted: Expanding LTE for Devices. http://www.ict-exalted.eu/ society. Technical report, ERICSSON (2014) fileadmin/documents/EXALTED_WP2_D2.1.pdf. Accessed 30 25. 3GPP LTE-Advanced. http://www.3gpp.org/technologies/key Sept 2013 words-acronyms/97-lte-advanced. Accessed 28 May 2015 5. Towards 50 billion connected devices. Technical report, 26. 3GPP: Technical specification group radio access network; ERICSSON (2010) evolved universal terrestrial radio access (E-UTRA); relay 6. ETSI: Machine-to-Machine communications (M2M); Use Cases architectures for E-UTRA (LTE-Advanced). Technical report, of automotive applications. In: M2M capable networks. Techni- TR 36.806 V9.0.0 (2010) cal report, TR 102 898 V1.1.1 (2013) 27. Mehmood Y, Afzal W, Ahmad F, Younas U, Rashid I, Mehmood 7. Slicing up the M2M revenue pie; How to get your share and boost I (2013) Large scaled multi-user mimo system so called massive your business. Technical report, COMARCH (white paper) mimo systems for future wireless communication networks. In: 8. Vodafone M2M Adoption Barometer. Technical report, Voda- 2013 19th international conference on automation and computing fone (2015) (ICAC). IEEE, pp 1–4 9. The state of M2M in transport and logistics; Key findings from 28. Chandrasekhar V, Andrews JG, Gatherer A (2008) Femtocell the 2013 Vodafone M2M Adoption Barometer. Technical report, networks: a survey. IEEE Commun Mag 46(9):59–67 Vodafone (2014) 29. Cox C (2012) An introduction to LTE: LTE, LTE-advanced, SAE 10. NOKIA: LTE M2M: optimizing LTE for the internet of things. and 4G mobile communications. Wiley, New york Technical report, Nokia white paper (2014) 30. METIS: Novel radio link concepts and state of the art analysis. 11. Laner M, Svoboda P, Nikaein N, Rupp M (2013) Traffic models Technical report, Deliverable Number ICT-317669-METIS/D2.2 for machine type communications. In: Proceedings of the tenth (2013) international symposium on wireless communication systems 31. Mehmood Y, Marwat SNK, Po¨tsch T, Ahmad F, Go¨rg C, Rashid I (ISWCS). VDE, pp 1–5 (2014) Impact of M2M traffic on LTE data traffic performance. 12. Shafiq MZ, Ji L, Liu AX, Pang J, Wang J (2012) A first look at In: LogDynamics international conference. Springer (in-press) cellular machine-to-machine traffic: large scale measurement and 32. Opnet Modeler Technologies. https://www.opnet.com/services/ characterization. In: ACM SIGMETRICS performance evalua- modeling-simulation-service.html. Accessed 10 Jul 2013 tion review, vol 40. ACM, pp 65–76 33. Cavers JK (2000) Mobile channel characteristics. Springer, 13. Mehmood Y, Nawaz Khan Marwat S, Grg C, Zaki Y, Timm-Giel Berlin A (2015) Evaluation of M2M data traffic aggregation in LTE-A 34. IEEE 802.16p-11/0014: IEEE 802.16p Machine to Machine uplink. In: ITG/VDE mobile communication conference. VDE, (M2M) Evaluation Methodology Document (EMD). Technical pp 24–29 report, IEEE 802.16 Broadband Wireless Access Working Group 14. ETSI: Machine-to-Machine communications (M2M), Functional (2010) architecture. Technical report, ETSI TS 102 690 V2.1.1 (2013) 15. IEEE, Wireless Access in Vehicular Environments. http://stan dards.ieee.org/findstds/standard/802.11p-2010.html. Page Acces- sed 21 Nov 2015

Journal

Logistics ResearchSpringer Journals

Published: Jul 28, 2016

References