USRP - MDPI

sustainability Article

An Experimental Channel Capacity Analysis of Cooperative Networks Using Universal Software Radio Peripheral (USRP) Shujaat Ali Khan Tanoli 1 , Mubashir Rehman 1 , Muhammad Bilal Khan 1 , Ihtesham Jadoon 1 , Farman Ali Khan 2 , Faiza Nawaz 1 , Syed Aziz Shah 3 ID , Xiaodong Yang 4, * and Ali Arshad Nasir 5 1

2 3 4 5

*

Department of Electrical Engineering, COMSATS University Islamabad, Attock Campus 43600, Pakistan; [email protected] (S.A.K.T.); [email protected] (M.R.); [email protected] (M.B.K.); [email protected] (I.J.); [email protected] (F.N.) Department of Computer Science Engineering, COMSATS University Islamabad, Attock Campus 43600, Pakistan; [email protected] School of International Education, Xidian University, Xi’an 710071, China; [email protected] School of Electronic Engineering, Xidian University, Xi’an 710071, China Department of Electrical Engineering, King Fahd University of Petroleum and Minerals, Dhahran 34464, Saudi Arabia; [email protected] Correspondence: [email protected]





Received: 31 March 2018; Accepted: 12 June 2018; Published: 13 June 2018

Abstract: Cooperative communication (CC) is one of the best solutions to overcome channel fading and to improve channel capacity. However, most of the researchers evaluate its performance based on mathematical modeling or by simulations. These approaches are often unable to successfully capture many real-world radio signal propagation problems. Hardware based wireless communication test-bed provides reliable and accurate measurements, which are not attainable through other means. This research work investigates experimental performance analysis of CC over direct communication (DC) in the lab environment. The experimental setup is built using Universal Software Radio Peripheral (USRP) and Laboratory Virtual Instrument Engineering Workbench (LabVIEW). A text message is transmitted by using Phase Shift Keying (PSK) modulation schemes. The setup uses amplify and forward (AF) relaying mode and two time slot transmission protocols. The maximum ratio combining (MRC) technique is used for combining SNR at the receiver. Channel capacity analysis is performed in order to evaluate the performance of CC over DC with and without obstacle. Moreover, optimal position of the relay is also analyzed by varying the position of the relay. Extensive experiments are carried out in the lab environment to evaluate the performance of the system for different hardware setups. The results reveal that cooperative communication attains significant improvement in terms of channel capacity of the system. Keywords: cooperative communication; channel capacity; relay; AF; MRC; USRP

1. Introduction The 5G wireless communication services on one hand are aimed for providing basic conventional telecommunication services, such as speech, web browsing, video download and upload, social networking, etc. On the other hand, they are supposed to provide wide range of services by linking a range of devices and sensors in the internet of things, for example, the smart city applications. The networks providing such services consist of large number of devices that are low cost and consume low energy. These devices cooperate to communicate the data to the desired place. Cooperative

Sustainability 2018, 10, 1983; doi:10.3390/su10061983

www.mdpi.com/journal/sustainability

Sustainability 2018, 10, 1983

2 of 13

communication (CC) is commonly preferred due to high/rapid exchange rate, range, and reliability. Furthermore, it has the tendency to overcome multi-path fading and provide diversity gain. The other features deal with spatial and user diversity by combining various signals that are received through different routes from multiple users at receiver node. The CC can significantly enhance the system performance in terms of reliability and channel capacity. The cooperative communication system consists of a virtual multiple input multiple output (VMIMO), where single-antenna cooperate with other antennas thus forming a multiple antenna system. During cooperation, the cooperative diversity gain of MIMO systems can be achieved by reducing the cost of physical antenna array at each node. Due to the high data rates and cost effective solutions, cooperative communication is considered to be one of the most efficient technique, and it can be exploited for the 5G wireless networks as well. Cover and Gamal [1] introduced the concept of relay in communication system and it acts as a door opening work in this direction. Later on, a lot of research work was carried using different relaying schemes and protocols. In [2,3], evaluated system performance using bit error rate (BER) analysis [4–6]. Similarly, General performance analysis bounds were also found for various CC scenarios [7]. However, majority of the research work evaluated the CC network performance either mathematically or by simulation. Theoretical or mathematically modeling investigation provides guidance for real time system implementation, with many real time factors with assumptions and limitations [8]. Regarding simulation, it has been generally acknowledged that results that are produced through such process often failed to provide the actual hurdle that is faced in real world radio signals. By considering all of the factors involved in radio wireless communication system, there is still a challenge how much performance improvement by cooperative communication can bring in real wireless environment [9]. To analyze real world radio communication factors, a real time testbed should be introduced to analyze the overall CC network performance. It should provide a realistic approach to evaluate the wireless system. Many researchers are working on different existing platforms to evaluate the CC networks. The commodity wireless card (e.g., 802.11 NIC) is one of the examples of these platforms [10]. However, the hardware of cooperative communication has extra requirements to implement physical layer functions, which cannot be resolved by these commodity wireless cards. Aside from these wireless cards, various testbeds are introduced using digital signal processor (DSP) kits or field programmable gate array (FPGA) boards [11]. While using these platforms, real time implementation and hardware issues can be resolved. The cost of implementation and hardware programming sets new era in research communities. But still, the above mentioned testbed is less efficient to solve the basic functions to implement various MAC layer protocols. The best solution against these problems is software defined radio (SDR) that is based on general purpose processor. They are more suitable for implementation of signal processing and MAC layer functions. They provide more efficient hardware programming environment. Universal Software Radio Peripheral (USRP) and GNU Radio, used as hardware and software platform [12,13] are most popularly used SDR platforms in these days. However, these platforms are not widely used for cooperative communication testbed. So, this is one of the reasons that existing one can perform very limited number of radio functions [14]. They are only used for selective cooperative communications. In [15], an experimental testbed was established by using Ettus Research USRPs N210 with SBX daughterboard, VERT2450 antennas, and GNU Radio. The obtained results concluded that the strength of the signal at the receiver is significantly improved by employing the relay in the wireless network. In all of the previous research work, performance metrics of experimental analysis is based on BER, signal strength, and transmit power. In this paper, USRP and LabVIEW are used to implement the cooperative network testbed. LabVIEW software and NI USRP hardware provides affordability, functionality, and flexibility to deliver an ideal software-defined radio prototyping platform. In this experimental hardware, setup is interfaced with LabVIEW by Gigabit Ethernet port. It offers the signal processing capabilities for the modulation and demodulation of signals streaming to and from NI USRP hardware. Toolkits from

Sustainability 2018, 10, 1983

3 of 13

LabVIEW software offer functional blocks for numerous analog and digital modulation methods and signal processing algorithms for real-world radio signals. The methodology that is used for implementation provides a platform for the evaluation of cooperative network performance on USRP and LabVIEW based testbed. This research paper provides the following contributions: I.

II. III. IV. V.

Design and implementation of a complete-functional testbed framework based on USRP and LabVIEW to support the cooperative communication, which includes a signal processing for the physical layer implementation of cooperative communication. Different modulation schemes are used to verify the performance of the system. Capacity analysis of cooperative network using LabVIEW and NI-USRP based test-bed. Extensive experiments are performed with and without obstacles in the lab environment. Determine the optimal position of the relay placement in the cooperative system.

The result shows the significant capacity gain of CC over DC. The response of the system is very similar to the theoretical and simulation analysis. This paper is organized in the following sequence: in Section 2, related research works are reviewed. In Section 3, system model of the cooperative network using USRP is introduced. In Section 4, the experimental setup is discussed. In Section 5, the methodology used in the research work is explained. Section 6 provides the experimental results and discussions. In the last, Section 7, the conclusion and future work directions are presented. 2. Related Work Previous research indicates that the performance of CC networks in real time wireless systems has been analyzed and evaluated on the basis of various testbeds. The testbed comprised of various kind of hardware design and functionalities. Regarding the performance of testbed, the BER approach was commonly adopted. For example, author et al. [8] exploited CoopMAC protocol on 802.11 standard devices in connection with open source platform. They performed a number of experiments on relays ranging from 3 to 10 in a single testbed. Similarly, in [12], commodity hardware multiple relays were utilized for implementing testbed, where the selection of strong signal was either from sender or relay without taking into account the signal combining technique. However, the drawback of this approach is that the actual cooperative diversity gain cannot be achieved at the destination. Furthermore, this hardware is not easy to modify at MAC layer and almost impossible to alter at physical layer, where actually CC technology is applied. To overcome the limitations of commodity hardware, later on, fast processors and boards like DSP and FPGA based testbeds were used to analyze CC system. For example, Zetterberg et al. [11] developed a testbed using a DSP kit with laptop or PC, and commonly used relaying schemes, such as amplify and forward (AF), Decode and forward (DF) and Compress and forward (CF). Furthermore, at the destination for getting the actual diversity gain of CC network, MRC technique was applied. The benefit of these testbeds, on the one hand, is that they provide hardware and software flexibility, and on the other hand they are cost effective. Similarly [16,17] developed a Wireless Open Access Research Platform (WARP) that uses amplify and forward relaying scheme and is based on the orthogonal frequency division multiplexing (OFDM) technique and distributed Alamouti transmit diversity scheme. However, the drawback of this approach on the one hand is that due the absence of external clock interface in hardware platform, the multi-relay synchronization becomes difficult. On the other hand, it is very expensive. The platform of USRP and GNU Radio is becoming very popular because of the implementation of all signals processing in CC testbed on it. The open source availability of GNU radio software made it possible to add new functionality and modifications. Due to its flexibility and popularity, a lot of novel ideas have been proposed and implemented on GNU Radio based testbed, including efficient analog network coding [18–20].

Sustainability 2018, 10, 1983

4 of 13

Distributed asynchronous cooperation (DAC) is introduced in [21], and the tradeoff between the diversity gain the reduction Sustainability 2018, 10, xand FOR PEER REVIEW in multiplexing opportunities is analyzed. Novel cross 4layer of 13 techniques are also proposed to verify the real time environment effect. To analyze the performance of multi-antenna system GNU radio based testbed is also used. In [22], GNU Radio based testbed to build hydra andand analyze the performance of MIMO on basis the experimental hydraprototype prototype analyze the performance of system, MIMO and system, andof on basis of the results, rate adaptation is proposed. [23], range observed is byobserved comparing experimental results, rateprotocol adaptation protocol isInproposed. Inextension [23], rangeis extension by multiple relay network singlewith relay network. While [24] analyzed the performance of DF comparing multiple relaywith network single relay network. While [24] analyzed the performance and DF without signal combining technique with GNU Radio based [25], of DFselective and selective DF without signal combining technique with GNU Radio basedtestbed. testbed. In [25], signal combining of CC was evaluated but optimal position of the relay and channel capacity have not been evaluated on this platform. These are most important facts to evaluate the performance of the system. In this research channel capacity analysis is evaluated for the cooperative network using USRP and LabVIEW. LabVIEW. 3. System System Model Model 3. The system system model model consists consists of of three three main main modules; modules; i.e., i.e., source source (S), (S), relay relay (R), (R), and and destination destination (D), (D), The as shown in the Figure 1. Each module is equipped with single antenna. as shown in the Figure 1. Each module is equipped with single antenna.

Figure 1. System model. Figure 1. System model.

3.1. Source Operation 3.1. Source Operation A text message is generated in PC in LabVIEW environment. Software defined BPSK, 4PSK, and A text message is generated in PC in LabVIEW environment. Software defined BPSK, 4PSK, 8PSK digital modulation schemes are used to transmit data over the wireless channel by using USRP and 8PSK digital modulation schemes are used to transmit data over the wireless channel by using kit. USRP kit. 3.2. 3.2. Relay Relay Operation Operation Relay source using using USRP USRP kit. kit. The Relay receives receives text text message message from from source The received received signal signal is is amplified amplified in in PC PC using LabVIEW. Amplified signal is again forwarded to USRP kit and then retransmitted towards using LabVIEW. Amplified signal is again forwarded to USRP kit and then retransmitted towards the the destination. destination. 3.3. 3.3. Optimal Optimal Position Position of of the the Relay Relay To determinethe theoptimal optimal position of relay, the relay, following three positions of the are To determine position of the following three positions of the relay are relay analyzed. analyzed. At position ‘A’ relay is placed near to the source and away from the destination, as shown in At2.position ‘A’ relay is placed to the source andtoaway the destination, as shown in Figure The distances from source near to destination, source relay,from and relay to destination are 12 m, Figure 2. The from source to destination, source to relay, and relay to destination are 12 m, 4.2 m, and 8.1distances m, respectively. 4.2 m, and 8.1 m, respectively.

Sustainability 2018, 10, 1983

5 of 13

Sustainability 2018, 10, x FOR PEER REVIEW Sustainability 2018, 10, x FOR PEER REVIEW Sustainability 2018, 10, x FOR PEER REVIEW

5 of 13 5 of 13 5 of 13

Figure 2. Position to source. source. Figure 2. Position A: A: Relay Relay is is near near to Figure 2. Position A: Relay is near to source. Figure 2. Position A: Relay is near to source.

At position ‘B’, relay is placed at equally distance from source and destination, as shown in At position position ‘B’, ‘B’, relay relay is placed equally distance from and destination, as shown shown in in At is source placedtoat atthe equally distance fromtosource source and relay destination, as Figure 3.position The distances from destination, source relay and to destination are 12 At ‘B’, relay is placed at equally distance from source and destination, as shown in Figure 3. distances from and relay to to destination areare 12 12 m, Figure 3. The The distances fromsource sourcetotothe thedestination, destination,source sourcetotorelay relay and relay destination m, 6.2 m, and 6.2 m, respectively. Figure 3. The from source to the destination, source to relay and relay to destination are 12 6.2 m, m, and 6.2 distances m, respectively. m, 6.2 and 6.2 m, respectively. m, 6.2 m, and 6.2 m, respectively.

Figure 3. Position B: Relay is center. Figure Figure 3. 3. Position Position B: Relay is center. Figure 3. Position B: Relay is center.

At position ‘C’, relay is placed away from source and near to destination, as shown in Figure 4. At position ‘C’,source relay is away source from source and near to to destination, asare shown Figure 4. The distances from toplaced destination, to relay and relay destination 12 m,in 8.1 m, and At position ‘C’, relay is placed away from from source source and near to destination, destination, as shown shown in Figure 4. At position ‘C’, relay is placed away and near to as in Figure 4. The distances from source to destination, source to relay and relay to destination are 12 m, 8.1 m, and 4.2 m, respectively. The distances from source to destination, source to relay and relay to destination are 12 m, 8.1 m, Them, distances from source to destination, source to relay and relay to destination are 12 m, 8.1 m, and 4.2 respectively. and 4.2respectively. m, respectively. 4.2 m,

Figure 4. Position C: Relay near to destination. Figure 4. Position C: Relay near to destination. Figure 4. 4. Position C: Relay Relay near near to to destination. destination. Figure Position C:

3.4. Transmission Protocols 3.4. Transmission Protocols 3.4. Transmission Protocols The Naber’sProtocols two time slot transmission protocol is applied for the direct path and cooperative 3.4. Transmission The Naber’s two time slot transmission protocol applied the second direct path path The [24].Naber’s This protocol is used because the source is is kept silentfor in the timeand slot,cooperative and hence two time slot transmission protocol is applied for the direct path and cooperative The Naber’s two time slot transmission protocol is applied for the direct path and cooperative path [24]. This protocol is used because the source is kept silent in the second time slot, and hence less transmit power is consumed by the the system. The signalsilent moves from sourcetime to destination and pathtransmit [24]. This This protocol isused usedbecause because source is kept in the second slot, and hence path [24]. protocol is the source is kept silent in the second time slot, and hence less less power is consumed by the system. The signal moves from source to destination and relay path in two time slots. In Table 1, two time slot transmission protocol is given. In time slot 1, less transmit power is slots. consumed bysystem. the system. The signal moves from source to destination and transmit power is consumed by the The signal moves from source to destination and relay relay path in two time In Table 1, two time slot transmission protocol is given. In time slot 1, the node S transmits the signal to node D and node R. In the time slot 2, only the node R transmits relay twoslots. time slots. In1,to Table 1, D two slot protocol is given. In Rtime 1, path inpath two time In Table two time slottime transmission protocol is given. Inthe time slot 1,transmits theslot node the node S in transmits the signal node and node R.transmission In the time slot 2, only node signal to node D. The relay is using AF scheme. node S transmits signal node D and R. In slot the 2, time slot 2, node only R thetransmits node R transmits Sthe transmits theD. signal to node Dtoand R. Innode the time only the signal to signal to node The the relay is using AFnode scheme. signal to node D. The relay is using AF scheme. node D. The relay is using AF scheme. Table 1. Transmission Protocol [26]. Table 1. Transmission Protocol [26]. Table 1. Transmission Protocol Number Time Slot 1 [26].

Protocol ProtocolIINumber Protocol Number II II

Time S →Slot R 11 Time Slot S→R S→R

Time Slot 2 Time R →Slot D 22 Time Slot R→D R→D

Sustainability 2018, 10, 1983

6 of 13

Table 1. Transmission Protocol [26]. Protocol Number

Time Slot 1

Time Slot 2

II

S→R S→D

R→D

3.5. Input-Output Equations The data from node S is transmitted via direct and a relay path. Time Slot 1: The received data equation at node D for time slot 1 is shown as in Equation (1) [8]. YSD = hSD

p

PSD x + nSD

(1)

where hSD is the channel response between is source and destination, PSD is the transmit signal power from source to destination, x is the signal transmitted, and nSD is the channel noise between S node to D node. The signal that is received at the node R in first time slot is shown by: YSR = hSR

p

PSR x + nSR

(2)

where hSR is channel response between source and relay are, PSR is the transmit signal power from source to relay, x is the transmitted signal, and nSR is the channel noise between S node to R node. Time Slot 2: In second time slot, the relay receives the signal from S node, amplifies it, and retransmits it to the D node. The received signal through the R node in time slot 2 is given as: YRD = h RD

p

PRD YSR + n RD

(3)

where channel response between relay and destination is h RD , PRD is the signal power, YSR is the transmitted signal, and n RD is the channel noise between the S node to R node. Signal from the S and R node is combined using MRC at D node. The MRC at the D node is given as: YMRC = YSD + YRD

(4)

3.6. Receiver In the receiver operation SNR received from the transmitter and relay are measured and combined at receiver. Performance of the system is measure by using Shannon capacity theorem that is given in Equation (5). C = B × log2 (1 + SNR) (5) In Equation (5), C is the channel capacity, B is the bandwidth, and SNR is the signal to noise ratio. 4. Experimental Setup Following experimental testbed is set in the lab environment for capacity analysis of cooperative network using USRP kits. In setup 1, as shown in Figure 5, three USRPs and PCs are used for the source destination and relay in the lab environment without any obstacle. USRPs moves to the maximum distance in the lab and relay USRP is placed at three different positions.

Sustainability 2018, 10, 1983

7 of 13

Sustainability 2018, 10, x FOR PEER REVIEW

7 of 13

Sustainability 2018, 10, x FOR PEER REVIEW

7 of 13

Figure 5. Experimental Setup of cooperative communication (CC) without obstacle.

Figure 5. Experimental Setup of cooperative communication (CC) without obstacle. Figure 5. Experimental Setup of cooperative communication (CC) without obstacle.

In setup 2, as shown in Figure 6, a wooden obstacle of 60 × 45 × 1.5 cm dimension is placed In setup 2, source as shown in Figure 6, a wooden obstacle of in 60the × relay 45 × destination 1.5 cm dimension is placed between the and the destination. No obstacle is placed path. In setup 2, as shown in Figure 6, a wooden obstacle of 60 × 45 × 1.5 cm dimension is placed between the the source and thethe destination. is placed placedininthe therelay relaydestination destination path. between source and destination.No Noobstacle obstacle is path.

Figure 6. Experimental Setup of CC with obstacle.

5. Methodology

Figure 6. Experimental Setup of CC with obstacle.

Figure 6. Experimental Setup of CC with obstacle.

5. Methodology The methodology that is used in this research is as follows. A text message is transmitted over

5. Methodology the direct path and relay path by using different digital M-PSK modulation schemes. The The methodology that is used in this research is as follows. A text message is transmitted over performance is analyzed byused varying the research transmit antenna gain andtext the received SNR is measured at Thedirect methodology that is this is as follows. is transmitted the path and relay pathin by using different digital A M-PSKmessage modulation schemes. over The the the destination. The capacity of channel is measured by using Shannon capacity theorem. In CC, the performance is analyzed byusing varying the transmit antenna and the received SNR is measured at is direct path and relay path by different digital M-PSKgain modulation schemes. The performance AF relaying mode is used with amplifying gain of 10 dB. Two time slot transmission protocols are the destination. The capacity of channel measured by using Shannon capacity theorem. Indestination. CC, the analyzed by varying the transmit antennaisgain and the received SNR is measured at the used. In first time slot, text message is transmitted to relay and destination for 50 s. In the second time relaying mode is is used with amplifying of 10 capacity dB. Two theorem. time slot transmission protocols The AF capacity of channel measured by using gain Shannon In CC, the AF relayingare mode slot after 20 s delay, relay again retransmits text to the receiver for 50 s. The receive SNR form direct used. In first time slot, text message is transmitted to relay and destination for 50 s. In the second timetime is used with amplifying gain of 10 dB. Two time slot transmission protocols are used. In first 20 s delay, relay again retransmits text to the receiver for 50 s. The receive SNR form direct slot,slot textafter message is transmitted to relay and destination for 50 s. In the second time slot after 20 s

Sustainability 2018, 10, 1983

8 of 13

Sustainability 2018, 10, x FOR PEER REVIEW

8 of 13

delay, relay again retransmits text to the receiver for 50 s. The receive SNR form direct path and relay path and relay path combined at the using the MRC technique. first capacity analysis path is combined at theisreceiver using thereceiver MRC technique. At first capacityAt analysis for experimental, for experimental, setup 1 is performed and then the same analysis is performed for setup 2. Optimal setup 1 is performed and then the same analysis is performed for setup 2. Optimal position of the relay position ofby the relay isthe analyzed byof varying the position of is theplaced relay. near First relay is placed the is analyzed varying position the relay. First relay the source thennear at center, source then at center, and finally near the destination. Extensive experiments are performed within and finally near the destination. Extensive experiments are performed within the lab environment. the lab environment. The following setup and system parameters are analyzed for the performance The following setup and system parameters are analyzed for the performance investigation of the investigation of the system shown in Table 2. system shown in Table 2. Table 2. System parameters. Table 2. System parameters.

Transmit Antenna Gain Operating Frequency Transmit Antenna Gain Operating Frequency Bandwidth Bandwidth Data type Data type Message Message Guards Guards Synchronization Synchronization Packets Packets Modulation Modulation Channel Channel Relaying mode Relaying mode Transmission protocols Transmission protocols Amplification Gain Amplification Gain Number of relays Number of relays Combining Technique Combining Technique

0–30 dB 915 MHz 0–30 dB 915 MHz 400 kHz 400 message kHz Text Text message 128 bits 128 bits 30 bits 30 bits bits 2020bits 4646 BPSK, 8PSK BPSK,QPSK, QPSK, 8PSK Real Realtime time AF AF Two time slot Two time slot 10 dB 10 1 dB 1 MRC MRC

6. Results and Discussion 6. Results and Discussion Based upon the system model and experimental setups, the capacity analysis is performed for Based upon the system model and experimental setups, the capacity analysis is performed for several M-PSK modulation schemes, as discussed below. several M-PSK modulation schemes, as discussed below. 6.1. Experimental Analysis of CC over DC Without Obstacle 6.1. Experimental Analysis of CC over DC Without Obstacle The following experimental analysis is based on the performance evaluation of DC with CC. The following experimental analysis is based on the performance evaluation of DC with CC. In In this setup, no obstacle is placed in either direct or relay path. this setup, no obstacle is placed in either direct or relay path. Figure 7 shows the experimental results of DC and CC without any obstacle. The performance of Figure 7 shows the experimental results of DC and CC without any obstacle. The performance CCofdominates the DC. From Table 3, the average channel capacity gain ofof CC over CC dominates the DC. From Table 3, the average channel capacity gain CC overDC DCisis14.44% 14.44%for BPSK, 14.59% for QPSK, and 14.75% for 8PSK, respectively. for BPSK, 14.59% for QPSK, and 14.75% for 8PSK, respectively.

Figure Performanceanalysis analysisof of CC CC over over direct direct communication Figure 7. 7.Performance communication(DC) (DC)without withoutobstacle. obstacle.

DC 0 2.01 5 2.07 10 2.08 15 2.12 20 2.16 Sustainability 2018, 10, 1983 25 2.26 30 2.29 Average Gain

CC 2.27 2.37 2.42 2.48 2.49 2.55 2.57

Gain (%) 12.93 14.49 16.34 16.98 15.27 12.83 12.22 14.44

DC 2.05 2.08 2.16 2.19 2.26 2.29 2.31

CC 2.36 2.45 2.5 2.51 2.55 2.59 2.61

Gain (%) 15.12 17.78 15.74 14.61 12.83 13.10 12.98 14.59

DC 2.08 2.12 2.17 2.25 2.27 2.31 2.33

CC 2.42 2.51 2.52 2.56 2.58 2.6 2.62

Table 3. Performance comparison of DC with CC without any obstacle.

Gain (%) 16.34 18.39 16.12 13.77 13.65 9 of 13 12.55 12.44 14.75

6.2. Experimental Analysis of CC over DC with Obstacle

Capacity (Mbits/s/Hz)

The following experimental analysis is based on the performance evaluation of DC with CC. A BPSK QPSK 8PSK wooden obstacle is placed in the direct path to analyze the performance of the system. DC CC Gain (%) DC CC Gain (%) DC CC Gain (%) Figure 8 shows the experimental results of DC and CC with obstacle. Here, significant 0 2.01 2.27 12.93 2.05 2.36 15.12 2.08 2.42 16.34 performance improvement is obtained for CC as compared to DC. From Table 4, it can be seen that 5 2.07 2.37 14.49 2.08 2.45 17.78 2.12 2.51 18.39 average10channel capacity gain2.42 of CC over DC is 15.97% for2.5 BPSK, 16.12% for QPSK, and 17.01%16.12 for 2.08 16.34 2.16 15.74 2.17 2.52 8PSK respectively. 15 2.12 2.48 16.98 2.19 2.51 14.61 2.25 2.56 13.77

Tx Antenna Gain (dB)

20 25 30

2.16 2.49 15.27 2.26 2.55 12.83 2.27 2.58 13.65 2.26 4. Performance 2.55 12.83 2.29 2.59 CC with 13.10 2.6 12.55 Table comparison of DC with obstacle.2.31 2.29 2.57 12.22 2.31 2.61 12.98 2.33 2.62 12.44 Capacity (Mbits/s/Hz) Tx Antenna Average Gain 14.44 14.59 14.75 BPSK QPSK 8PSK Gain (dB) DC CC Gain (%) DC CC Gain (%) DC CC Gain (%) 0 1.94 of2.26 2.04 2.35 15.19 1.99 2.43 22.11 6.2. Experimental Analysis CC over 16.49 DC with Obstacle 5 2.04 2.36 15.68 2.07 2.44 17.87 2.08 2.49 19.71 The following experimental analysis is based on2.49 the performance evaluation of 17.28 DC with CC. 10 2.08 2.41 15.86 2.11 18.00 2.14 2.51 15 15.71 path2.15 2.5 the performance 16.27 2.18of the 2.52 15.59 A wooden obstacle is2.1 placed2.43 in the direct to analyze system. 2.13experimental 2.48 16.43 2.53 with 14.47 2.57 16.28 Figure20 8 shows the results of 2.21 DC and CC obstacle. 2.21 Here, significant performance 25 2.19 2.57 17.35 2.22 2.56 15.31 2.26 2.58 14.15 improvement is obtained for CC as compared to DC. From Table 4, it can be seen that average channel 30 2.24 2.56 14.28 2.29 2.65 15.72 2.29 2.61 13.97 capacity gainAverage of CC over respectively. Gain DC is 15.97% 15.97for BPSK, 16.12% for QPSK, 16.12 and 17.01% for 8PSK17.01

Figure 8. Performance analysis of CC over DC with obstacle. Figure 8. Performance analysis of CC over DC with obstacle.

Table 4. Performance comparison of DC with CC with obstacle. Capacity (Mbits/s/Hz) BPSK

Tx Antenna Gain (dB) 0 5 10 15 20 25 30

QPSK

8PSK

DC

CC

Gain (%)

DC

CC

Gain (%)

DC

CC

Gain (%)

1.94 2.04 2.08 2.1 2.13 2.19 2.24

2.26 2.36 2.41 2.43 2.48 2.57 2.56

16.49 15.68 15.86 15.71 16.43 17.35 14.28

2.04 2.07 2.11 2.15 2.21 2.22 2.29

2.35 2.44 2.49 2.5 2.53 2.56 2.65

15.19 17.87 18.00 16.27 14.47 15.31 15.72

1.99 2.08 2.14 2.18 2.21 2.26 2.29

2.43 2.49 2.51 2.52 2.57 2.58 2.61

22.11 19.71 17.28 15.59 16.28 14.15 13.97

Average Gain

15.97

16.12

17.01

Sustainability 2018, 10, 1983

10 of 13

Sustainability 2018, 10, x FOR PEER REVIEW

10 of 13

6.3. Optimal Position of the Relay without and with Obstacle 6.3. Optimal Position of the Relay without and with Obstacle Capacity analysis is performed for three different positions of relay for evaluating the optimal Capacity analysisgain is performed forfor three differentposition positions relaythe forfollowing evaluating the optimal position. The capacity is achieved the optimal byofusing cases: position. The capacity gain is achieved for the optimal position by using the following cases: • Case 1: Capacity gain of Position A over B • Case 1: Capacity gain of Position A over B • Case 2: Capacity gain of Position A over C • Case 2: Capacity gain of Position A over C •• Case Case 3: 3: Capacity Capacity gain gain of of Position Position B B over over C C In this setup, setup,no noobstacle obstacleisisplaced placed between source destination In this inin between thethe source andand the the destination and and relayrelay path.path. The The performance is analyzed by varying position of the relay at position B, and Capacity gain performance is analyzed by varying the the position of the relay at position A, A, B, and C. C. Capacity gain is is compared among all the three position of the relay. compared among all the three position of the relay. Figure position of of the the relay relay without without obstacle. obstacle. Figure 99 shows shows the the results results of of evaluation evaluation for for the the optimal optimal position The The relay relay position position B B performance performance is is better better than than Position Position A A and and Position Position C. C. While While the the relay relay position position A A performance is better than C. performance is better than C.

Figure 9. 9. Optimal Optimal position position without without obstacle. obstacle. Figure

From Table 5, the average capacity gain of relay position B over A and C are 1.35% and 1.90%, From Table 5, the average capacity gain of relay position B over A and C are 1.35% and 1.90%, respectively. On the other hand average capacity gain of relay position at A is 0.54% over C. The respectively. On the other hand average capacity gain of relay position at A is 0.54% over C. The resulted resulted analysis shows that optimal position of the relay is at point B. This means if relay is placed analysis shows that optimal position of the relay is at point B. This means if relay is placed equally equally distance from the source and destination it provides more capacity gain. distance from the source and destination it provides more capacity gain. Table 5. Optimal position of the relay without obstacle. Table 5. Optimal position of the relay without obstacle. Capacity (Mbits/s/Hz) Capacity Gain Tx Antenna Gain (dB) Capacity (Mbits/s/Hz) Capacity Gain2 Position A Position B Position C Case 1 Case Case 3 Tx Antenna Gain (dB) Position C Case 1.28 1 Case1.72 2 Case0.43 3 0 2.33 A Position 2.36 B Position 2.32 100 2.41 2.45 2.33 2.36 2.322.40 1.281.65 1.722.08 0.430.41 2.41 2.45 2.402.45 1.651.62 2.082.04 0.410.40 1510 2.46 2.5 2.46 2.5 2.452.46 1.620.80 2.042.03 0.401.21 2015 2.49 2.51 20 2.49 2.51 2.46 0.80 2.03 1.21 25 2.51 2.55 2.50 1.59 2.00 0.40 25 2.51 2.55 2.50 1.59 2.00 0.40 3030 2.56 2.59 2.55 1.17 1.56 2.56 2.59 2.55 1.17 1.56 0.390.39 Average Gain 1.35 1.90 0.54 Average Gain 1.35 1.90 0.54

In this setup, when the obstacle is introduced between source and destination, performance is In this when the obstacle is introduced between andfor destination, performance is analyzed forsetup, all three position A, B, and C. Capacity gain is source calculated all the above three cases. analyzed for all three position A, B, and C. Capacity gain is calculated for all the above three cases. In Figure 10 shows the results of evaluation for the optimal position of the relay with obstacle. The In Figure 10 shows results of evaluation for the optimal position the position relay with The relay relay position at B the performance is better than A and C. While the of relay atobstacle. A performance is position at B performance is better than A and C. While the relay position at A performance is better better than C. From Table 6, the average capacity gain of relay position at B is 1.22% and 1.70% over A and C. On the other hand, the average capacity gain of relay position at A is 0.47% over B position.

Sustainability 2018, 10, 1983

11 of 13

than C. From Table 6, the average capacity gain of relay position at B is 1.22% and 1.70% over A and C. On the other hand, capacity gain of relay position at A is 0.47% over B position. 11 of 13 Sustainability 2018, 10, x the FORaverage PEER REVIEW

Figure 10. 10. Optimal Optimal position position with with obstacle. obstacle. Figure Table 6. 6. Optimal Optimal position position of of the the relay relay with with obstacle. obstacle. Table Tx Antenna Gain (dB) Tx Antenna Gain (dB)

0 010 1015 15 20 20 2525 3030

Capacity (Mbits/s/Hz) Capacity Gain Capacity Capacity Gain Position A (Mbits/s/Hz) Position B Position C Case 1 Case 2 Case 3 Position A Position B Position C Case 1 Case 2 Case 2.29 2.32 2.28 1.31 1.75 0.433 2.29 2.32 2.46 2.28 2.42 1.31 1.23 1.751.65 0.43 2.43 0.41 2.43 2.46 2.49 2.42 2.45 1.23 1.21 1.651.63 0.41 2.46 0.40 2.46 2.49 2.45 1.21 1.63 0.40 2.5 2.53 2.48 1.20 2.01 0.80 2.5 2.53 2.48 1.20 2.01 0.80 2.53 2.56 2.52 1.18 1.58 0.39 2.53 2.56 2.52 1.18 1.58 0.39 2.56 0.39 2.56 2.59 2.59 2.55 2.55 1.17 1.17 1.561.56 0.39 Average Gain 1.22 1.70 0.47 Average Gain 1.22 1.70 0.47

7. Conclusions Conclusions 7. Cooperative communication communication is one of of the the best best solutions solutions to to overcome overcome the the multipath multipath fading fading and and Cooperative is one provides diversity gain. In this research work, an experimental testbed comprised of USRP and provides diversity gain. In this research work, an experimental testbed comprised of USRP and LabVIEW is used to build a cooperative network SDR platform. Extensive experiments are carried LabVIEW is used to build a cooperative network SDR platform. Extensive experiments are carried out outcapacity for capacity ofover CC over DC lab environment. Experimental performance analysis shows for gaingain of CC DC in labinenvironment. Experimental performance analysis shows that that CC significantly improves the performance of the system than DC. Experimental analysis shows CC significantly improves the performance of the system than DC. Experimental analysis shows an an increase in the transmit antenna gain results in improving the SNR at receiver. High SNR provides increase in the transmit antenna gain results in improving the SNR at receiver. High SNR provides more channel channel capacity capacity by by using using higher higher order order PSK PSK modulation. modulation. The The performance performance of of CC CC significantly significantly more increases the system reliability and the channel capacity of the system with and without obstacle increases the system reliability and the channel capacity of the system with and without obstacle environment. It It is is also also analyzed analyzed that that channel channel capacity capacity increases increases by by placing placing the the relay relay at at center center between between environment. source and destination. source and destination. 8. Future Future Recommendations Recommendations 8. As aa future future extension extension to to this this work, work, aa cooperative cooperative communication communication platform platform can can be be implemented implemented As with advance advance channel channel coding coding techniques techniques and and multiple multiple relays. relays. Automatic detection of of digital digital with Automatic detection modulation can be applied. This system can be made spectral efficient system by using orthogonal modulation can be applied. This system can be made spectral efficient system by using orthogonal frequency division division multiple multiple accesses. accesses. On forward relaying relaying mode mode can can be be frequency On this this testbed, testbed, decode decode and and forward possible by by decoding decoding the the text text message message at at the the relay, relay, which which will will be be similar similar to to receiver receiver operation operation at at the the possible cost of system complexity. The other combining schemes can be implemented at the receiver like cost of system complexity. The other combining schemes can be implemented at the receiver like selection combining, equal gain combiner, etc. Currently, this work is implemented in lab environment but it can be extended to outdoor environments. This research can be further extended to future 5G networks like Internet of thing, vehicle to vehicle communication, machine to machine communication, and fully functional software defined radios.

Sustainability 2018, 10, 1983

12 of 13

selection combining, equal gain combiner, etc. Currently, this work is implemented in lab environment but it can be extended to outdoor environments. This research can be further extended to future 5G networks like Internet of thing, vehicle to vehicle communication, machine to machine communication, and fully functional software defined radios. Author Contributions: S.A.K.T., M.R., M.B.K., I.J., F.N. are the principle investigator and contributed in the designing and establishment of experimental tested, configuring of NI-USRP with LabVIEW, LabVIEW Programing Including generation of text message, designing virtual instruments for transmitter, receiver and relay, real time measurements and analysis of the results respectively; F.A.K. contributed in analysis and presentation of the results, drafting and proof reading; S.A.S., X.Y. are the international research collaborators contributed in simulations, mathematical modeling and funding this research; A.A.N. is also research collaborator helped in drafting author reply, comparative analysis, proof reading and rephrasing. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2. 3.

4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14. 15.

16. 17. 18.

Cover, T.; Gamal, A.E. Capacity theorems for the relay channel. IEEE Trans. Inf. Theory 1979, 25, 572–584. [CrossRef] Bletsas, A.; Khisti, A.; Reed, D.P.; Lippman, A. A simple cooperative diversity method based on network path selection. IEEE J. Sel. Areas Commun. 2006, 24, 659–672. [CrossRef] Laneman, J.N.; Wornell, G.W.; Tse, D.N. An efficient protocol for realizing cooperative diversity in wireless networks. In Proceedings of the 2001 IEEE International Symposium on Information Theory, Washington, DC, USA, 29 June 2001; p. 294. Hunter, T.E.; Sanayei, S.; Nosratinia, A. Outage analysis of coded cooperation. IEEE Trans. Inf. Theory 2006, 52, 375–391. [CrossRef] Kramer, G.; Gastpar, M.; Gupta, P. Cooperative strategies and capacity theorems for relay networks. IEEE Trans. Inf. Theory 2005, 51, 3037–3063. [CrossRef] Laneman, J.N.; Tse, D.N.; Wornell, G.W. Cooperative diversity in wireless net-works: Efficient protocols and outage behavior. IEEE Trans. Inf. Theory 2004, 50, 3062–3080. [CrossRef] Host-Madsen, A. Capacity bounds for cooperative diversity. IEEE Trans. Inf. Theory 2006, 52, 1522–1544. [CrossRef] Khan, I.; Rajatheva, N.; Tanoli, S.; Jan, S. Performance analysis of cooperative network over nakagami and rician fading channels. Int. J. Commun. Syst. 2014, 27, 2703–2722. [CrossRef] Kumar, N.; Bhatia, V. Outage probability and average channel capacity of amplify-and-forward in conventional cooperative communication networks over rayleigh fading channels. Wirel. Pers. Commun. 2016, 88, 943–951. [CrossRef] Korakis, T.; Tao, Z.; Makda, S.; Gitelman, B.; Panwar, S. To Serve is to Receive Implications of Cooperation in a Real Environment; Proceedings of the Networking: Atlanta, GA, USA, 2007. Zetterberg, P.; Mavrokefalidis, C.; Lalos, A.S.; Matigakis, E. Experimental investigation of cooperative schemes on a real-time dsp-based testbed. EURASIP J. Wirel. Commun. Netw. 2009, 1, 368752. [CrossRef] Gnu Radio. Available online: http://www.gnu.org/software/gnuradio (accessed on 17 May 2018). Usrp Family Brochure. Available online: http://www.ettus.com/downloads/er_broch_trifold_v5b.pdf (accessed on 17 May 2018). Bletsas, A.; Lippman, A. Implementing cooperative diversity antenna arrays with commodity hardware. IEEE Commun. Mag. 2006, 44, 33–40. [CrossRef] Wang, H.; Tian, S.; Gao, X.; Wu, L.; Chen, G. Approximation Designs for Cooperative Relay Deployment in Wireless Networks. In Proceedings of the 2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS), Atlanta, GA, USA, 5–8 June 2017; pp. 2270–2275. Building a Cooperative Communications System. Available online: http://warp.rice.edu/trac/attachment/ wiki/JSAC_CooperativeCo/Rice_JSAC_CooperativeComm.pdf (accessed on 17 May 2018). Wireless Open-Access Research Platform. Available online: http://warp.rice.edu/index (accessed on 17 May 2018). Katti, S.; Gollakota, S.; Katabi, D. Embracing wireless interference: Analog network coding. ACM SIGCOMM Comput. Commun. Rev. 2007, 37, 397–408. [CrossRef]

Sustainability 2018, 10, 1983

19.

20. 21. 22.

23.

24. 25.

26.

13 of 13

Katabi, D.; Gollakota, S. Zigzag decoding: Combating hidden terminals in wireless networks. In Proceedings of the ACM SIGCOMM 2008 Conference on Data Communication, New York, NY, USA, 17–22 August 2008; pp. 159–170. Katti, S.; Katabi, D.; Balakrishnan, H.; Medard, M. Symbol-level network coding for wireless mesh networks. ACM SIGCOMM Comput. Commun. Rev. 2008, 38, 401–412. [CrossRef] Zhang, X.; Shin, K.G. Cooperation without synchronization: Practical cooperative relaying for wireless networks. IEEE Trans. Mob. Comput. 2015, 14, 937–950. [CrossRef] Kim, W.; Khan, O.; Truong, K.T.; Choi, S.-H.; Grant, R.; Wright, H.K.; Mandke, K.; Daniels, R.C.; Heath, R.W., Jr.; Nettles, S.M. An experimental evaluation of rate adaptation for multi-antenna systems. In Proceedings of the 2009 IEEE INFOCOM, Rio de Janeiro, Brazil, 19–25 April 2009; pp. 2313–2321. Omar, M.S.; Raza, S.A.; Kabir, S.H.; Hussain, M.; Hassan, S.A. Experimental implementation of cooperative transmission range extension in indoor environments. In Proceedings of the 2015 International Wireless Communications and Mobile Computing Conference (IWCMC), Dubrovnik, Croatia, 24–28 August 2015; pp. 1312–1317. Bradford, G.J. A Framework for Implementation and Evaluation of Cooperative Diversity in Software-Defned Radio. Master’s Thesis, University of Notre Dame, Notre Dame, IN, USA, 2008. Zhang, J.; Jia, J.; Zhang, Q.; Lo, E.M.K. Implementation and evaluation of co-operative communication schemes in software-defined radio testbed. In Proceedings of the 2010 Proceedings IEEE INFOCOM, San Diego, CA, USA, 14–19 March 2010. Nabar, R.U.; Bolcskei, H.; Kneubuhler, F.W. Fading relay channels: Performance limits and space-time signal design. IEEE J. Sel. Areas Commun. 2004, 22, 1099–1109. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).