A cluster based interference mitigation scheme for performance enhancement in IEEE

Transcription

1 756 Journal of Scientific & Industrial Research J SCI IND RES VOL 7 SEPTEMBER 2 Vol. 7, September 2, pp A cluster based interference mitigation scheme for performance enhancement in IEEE G M Tamilselvan* and A Shanmugam Department of ECE, Bannari Amman Institute of Technology, Sathyamangalam 638 4, India Received 22 September 2; revised 26 July 2; accepted August 2 This study proposed an adaptive transmission power aware cluster scheduling algorithm using multiple channels in a wireless personal area network (WPAN) in the presence of wireless local area network (WLAN) interference. Algorithm includes node identification, channel allocation, clustering and time scheduling. In proposed algorithm, performance metrics [bit error rate (BER), packet error rate (PER), throughput, average end-end delay and average jitter] were measured through Qualnet simulation and it is found effective in an IEEE cluster-tree network in the presence of multiple IEEE 82. interferers. Keywords: Clustering, Coexistence, Heterogeneous, Packet Error Rate (PER), WLAN and WPAN (Zigbee) Introduction As a low power and low cost technology, IEEE is establishing its place on the market as an enabler for emerging wireless sensor networks (WSNs). IEEE defines three physical layers (2.4 GHz, 868 MHz and 95 MHz frequency bands; 2.4 GHz band is available worldwide, while 868 MHz is available in Europe and 95 MHz in North America) and medium access control (MAC) sub layer of OSI Zigbee stack. It has a total of 27 channels with three different data rates (6 channels with a data rate of 25 kbps at 2.4 GHz band, channels with a data rate of 4 kbps at 95 MHz band, and channel with a data rate of 2 kbps at 868 MHz band). Like IEEE 82.b and IEEE 82.g, IEEE is also used in 2.4 GHz ISM band (Fig. ). IEEE 82. standard has channels, each of which occupies 22 MHz and up to 3 channels can be used simultaneously without mutual interference. On the other hand, IEEE standard defines 6 channels (2 MHz), channels through 27 in 2.4 GHz ISM band, all of which can be used simultaneously without mutual interference. Regarding coexistence -3, IEEE has a little impact on IEEE 82. performance. However, IEEE 82. can have a serious impact on IEEE performance if channel allocation is not carefully taken *Author for correspondence tamiltamil@rediffmail.com into account,4. It is believed the studies so far have dealt with only limited cases of coexistence scenarios. Channel conflict probabilities (CCPs) between IEEE 82.5 based wireless personal area networks (WPANs) have been modeled 5. Packet error rate (PER) of IEEE under IEEE 82.b interference 6 and that of IEEE 82.b under IEEE interference 7 has been analyzed. CCPs 8 between IEEE 82.b and IEEE have been modeled. Channel collision 9 between IEEE and IEEE 82.b for circular and grid topology has been analyzed with mobility model. Effect of inter packet delay has been analyzed. This study proposes power aware time slot and frequency based spectrum access analysis for the performance metrics [bit error rate (BER), PER, throughput, average end-end delay and average jitter] of IEEE , by focusing on coexistence of IEEE and IEEE 82.b in 2.4 GHz ISM band. Experimental Section Overview of IEEE 82.b and IEEE IEEE 82.b IEEE 82.b standard defines MAC sub layer and physical (PHY) layer for wireless local area networks (WLANs). The standard operates at 3 overlapping channels (bandwidth of each channel, 22 MHz) in 2.4 GHz ISM band. IEEE 82.b MAC employs carrier sense multiple access with collision avoidance (CSMA/ CA) mechanism. Before initiating a transmission, an IEEE

2 TAMILSELVAN & SHANMUGAM: CLUSTER SCHEDULING ALGORITHM FOR PERFORMANCE ENHANCEMENT IN IEEE Fig. 82. and channels in 2.4GHz ISM band (a) Star Topology PAN Coordinator Device Coordinator (b) Peer-Peer Topology Communication Flow 82.b node senses the channel to determine whether another node is transmitting. If medium is sensed idle for a distributed coordination function inter-frame space (DIFS) time interval, transmission will be preceded. If medium is busy, node defers transmission. When medium becomes idle for a DIFS interval, node will generate a random back off delay uniformly chosen in an interval. This interval [, W] is called contention window, where W is size. Initial W is set to CWmin. Back off timer is decreased by one as long as medium is sensed idle for a back off time slot. Back off counter will become frozen when a transmission is detected on the medium, and resumed when channel is sensed idle again for a DIFS interval. When back off timer reaches zero, node transmits a DATA packet. Immediately after receiving a packet correctly, destination node waits for a short inter frame spacing (SIFS) interval and then transmits an ACK back to source node. IEEE IEEE MAC sub layer is based on CSMA/ CA with two modes of operation: i) unslotted-csma (beaconless mode); and ii) slotted-csma (beacon enabled mode). Basic responsibilities for MAC sub layer is transmitting beacon frames, synchronization and providing a reliable transmission between Zigbee devices. Link layer acknowledgments (LLAs) are optional in IEEE , which can provide extra link level reliability. For present simulations, unslotted-csma is used as all sources will be continuously contending for the channel. LLAs are used in order to make transmission more reliable. To minimize energy consumption of Zigbee nodes, slotted CSMA/CA should be taken into consideration, since it uses beacon frames that contain information about when nodes can go into sleep mode. Proposed Scheme In proposed scheme, WPAN devices are clustered. Each cluster will have one PAN coordinator and four end devices (Fig. 2). A heterogeneous network is considered with random topology for IEEE Here, performance of IEEE under interference of IEEE 82.b and interference among IEEE nodes because of multiple transmissions is analyzed using Qualnet 4.5 simulation. For simulation, unslotted CSMA/ CA of IEEE model is developed using Qualnet 4.5. PHY of IEEE at 2.4 GHz uses offset quadrature phase shift keying (OQPSK) modulation. Denote that E b / N is the ratio of average energy (E b ) per information bit to noise power spectral density (N ) at receiver input, in the case of an additive white Gaussian noise (AWGN) channel. BER, P B, can be expressed as P B = Q where 2E N b Q( x) = Fig. 2 IEEE topologies x x u exp 2 2 du () (2) P B decreases when E b / N o increases (Fig. 3). N increases when collision increases. As number of WLAN sources increases, BER of IEEE increases because contentions among multiple WLANs increase the channel usage and cause collisions, which is more powerful interference. Here AWGN is considered and there will be no fading. PER is calculated as a function of BER, P b. Probability of not having a bit error is the probability that all bits are received correctly. Therefore, conditional probability of PER is one minus the probability of no bit errors and is computed as ( ) N PER = P b (3)

3 758 J SCI IND RES VOL 7 SEPTEMBER 2 where N represents number of bits in a packet. For experimental setting, each packet is composed of 5 bytes in the case of WPAN node and 5 bytes in the case of WLAN node. If there is an error correction mechanism, then PER utilizing BER should be computed differently. However, experimental platform does not provide an error correction mechanism and Eq. (3) is the final form of PER. Fig. 3 Theoretic bit error rate of OQPSK In this study, conventional TDMA scheme (Fig. 4) and two clusters based scheduling schemes [intra cluster scheduling (ICS, Fig. 5) and cluster reformation scheduling (CRS, Fig. 6)] are proposed and results are compared. In ICS, nodes are separated based on their transmission power. Output power of devices is typically as low as dbm, whereas output power of 82.b devices is 5 dbm or above. Then WLAN nodes are grouped under one operating frequency and WPAN nodes are clustered with cluster size 5. Each cluster will have one PAN coordinator and four end devices. Each cluster is allotted unique channel frequency for error free transmission. After frequency scheduling, in each channel, specific time slot is allotted for packet transmission. In CRS, cluster members from different clusters are grouped under one channel and specific time slots are allotted for packet transmission. Simulation Results and Discussion To evaluate effectiveness of proposed scheme in a coexistence heterogeneous wireless network, a simulation study was conducted using Qualnet 4.5 simulator for three different schemes. For conventional TDMA scheme, all nodes are linked with single channel, time slots are allotted for transmission and simulation time is Fig. 4 Random topology scenario with conventional TDMA

4 TAMILSELVAN & SHANMUGAM: CLUSTER SCHEDULING ALGORITHM FOR PERFORMANCE ENHANCEMENT IN IEEE Fig. 5 Random topology scenario with intra cluster scheduling Fig. 6 Random topology scenario with reformed cluster scheduling

5 76 J SCI IND RES VOL 7 SEPTEMBER x BER, % PER a) b) Throughput,bits/s 8 x Average end-end delay, s c) d) x -3 Average Jitter in sec Average jitter, s e) Fig. 7 For random topology, analysis of: a) BER; b) PER; c) Throughput; d) Average end-end delay; and e) Average jitter

6 TAMILSELVAN & SHANMUGAM: CLUSTER SCHEDULING ALGORITHM FOR PERFORMANCE ENHANCEMENT IN IEEE fixed as 53s. Simulation configuration and parameters used for IEEE 82.b and IEEE , respectively, are as follows: number of nodes, 2, 2; transmission power, 5, 3 dbm; modulation, CCK, OQPSK; MAC protocol, 82., ; routing protocol, bellman ford, AODV; no. of packets,, ; payload size, 5, 5 bytes; simulation time, 35 s, -; packet interval,, ms; packet transmission time, 5, s; test bed size, 4 m 4 m, -; and topology, random, -. Effectiveness of proposed scheme was measured with different metrics (BER, PER, throughput, average end-end delay and average jitter). In BER analysis for random topology (Fig. 7a), when RCS is adopted, BER becomes zero. When ICS is used, BER is reduced by 7-83%. When conventional TDMA and ICS is adopted, time slot mechanism is not helpful in WPAN network because ZigBee is a mesh networking technology, indicating that devices can automatically route messages on each other s behalf (often called multi-hopping). This allows deploying larger networks without immoderately increasing transmission power since direct communications occur only in a geographically-restricted area. In PER analysis for random topology (Fig. 7b), PER becomes zero when RCS is used. ICS improves the performance by reducing 36% of PER. In throughput analysis for random topology (Fig. 7c), throughput is increased by.3% when ICS is used. After implementing RCS, throughput is increased by 9.3%. In average endend delay analysis for random topology (Fig. 7d), endend delay is decreased by 3% when ICS is used. When RCS is used, it is decreased by 27%. In average jitter analysis for random topology (Fig. 7e), average jitter is decreased by 2.8% when ICS is used. When RCS is used, it is decreased by 2%. In order to measure easily the effect of proposed scheme for various nodes, instead of measuring the results for IEEE nodes, the results are plotted for both IEEE as well as IEEE 82.b nodes. In Fig. 7, comparison of all the three scheduling schemes for various performance measures can be clearly observed for different nodes. Conclusions This study presented performance analysis of coexistence heterogeneous networks. A new power based scheme using ICS and CRS mechanism for the coexistence of multiple IEEE LRWPAN and IEEE 82.b WLAN is proposed. Simulation results are compared with conventional TDMA scheme. Performance metrics (BER, PER, throughput, average end-end delay and average jitter) of IEEE network is analyzed when nodes are static. Simulation results show that proposed scheme is effective in performance improvement for coexistence network of IEEE for random topology. In future, the analysis can be extended with mobility model and the same proposed scheme can be implemented with Exata emulator and Free scale processor. References Petrova M, Riihijarvi J, P. Mahonen & Labella S, IEEE : Low rate - wireless personal area network coexistence issues, in Proc IEEE WCNC 6 (Las Vegas, USA) 26, Howitt I & Gutierrez J, Low-rate wireless personal area networks - Enabling wireless sensors with IEEE , in Proc IEEE WCNC 3, vol 3 (New Orleans, La, USA) 23, Sikora A, Coexistence of IEEE (ZigBee) with IEEE82. (WLAN), in Bluetooth and Microwave Ovens in 2.4 GHz ISM-Band,(online) 24: stzedn/ 4 Shin S Y, Choi S, Park H S & Kwon W H, Packet error rate analysis of IEEE under IEEE 82.b interference, in Proc 3rd Int Conf on Wired/Wireless Internet Commun (WWIC 5), vol 35 of Lecture Notes in Computer Science (Xanthi, Greece) May 25, Chen L-J, Sun T & Gerla M, Modeling channel conflict probabilities between IEEE 82.5 based wireless personal area networks, in IEEE Int Conf on Commun, vol (Istanbul, Turkey) -5 June 26, Shin S Y, Parky H S, Choi S & Kwon W H, Packet error rate analysis of IEEE under IEEE 82.b interference, IEICE Trans Commun, 9-B (27) Yoon D G, Shin S Y, Kwon W H & Park H S, Packet error rate analysis of IEEE 82.b under IEEE interference, in Vehicular Technol Conf, vol 3 (Grand Hyatt Melbourne, Melbourne, Australia) 7- May 26, Tamilselvan G M & Shanmugam A, Modeling channel conflict probabilities and interference analysis of coexistent heterogeneous networks, Int J Compu Sci & Knowl Engg (IJCSKE), 3 (29) Tamilselvan G M & Shanmugam A, Probability analysis of channel collision between IEEE and IEEE 82.b using Qualnet simulation for various topologies, Int J Compu Theory & Engg (IJCTE), (29) Tamilselvan G M & Shanmugam A, Effect of inter packet delay in performance analysis of coexistence heterogeneous wireless packet networks, Int J Network Security & Applic (IJNSA), (29) 4-49.