125

Proc. of Int. Conf. on Advances in Electrical & Electronics 2010

Energy Efficiency and Latency Analysis of IEEE 802.15.4 MAC Layer for Wireless Body Sensor Networks Ajeya B Dept. of Electronics and Communication Canara Engineering College Mangalore, India ajaybolar@yahoo.com

Durga Prasad Dept. of Electronics and Communication NMAM Institute of Technology, Nitte Karkala, India Abstract- Wireless body sensor networks (WBSN) are a particular type of wireless sensor networks (WSN) that thanks to the development of innovative wearable, wireless and implantable biosensors have gained tremendous international interest in recent years. The applications of WBSN extend from in-vivo monitoring and intervention to everyday healthcare, as well as fitness, sport and security. This paper provides a brief introduction to the 802.15.4 standard and investigates its use in wireless body sensor networks (WBSNs).

Keywords: WBSN, NS-2, IEEE 802.15.4

I.

INTRODUCTION

While the IEEE 802.15.4 wireless networking standard is quickly gaining popularity in industry as the physical (PHY) and media access control (MAC) layer of choice for developing LR-WPAN applications, the academic community has tended to neglect the impact this standard is having in the field of wireless networking.

In contrast to other MAC protocols that have been developed for wireless sensor networks (WSNs), 802.15.4 is highly configurable. This comes at the cost of increased code size, which is an issue for resource-constrained WSN platforms. For this reason, implementing the entire standard on a WSN platform has proved to be a challenge, and as of this writing we are not aware of any such opensource implementations. Since the standard is still in its relative infancy, however, it is not widely understood how it should be configured to optimize performance in aspects that might be important for WSNs. 802.15.4 Supports acknowledgments, which may be turned on and off in certain data transfer modes. If acknowledgments are turned on then transmissions should always be reliable. What is more interesting, however, is what can be termed as "efficient reliability". Efficient reliability is the idea of supporting reliability with the fewest number of retransmissions. Fewest retransmissions mean less energy consumption, which is more important for the WBSN applications. Our aim is to determine the energy efficiency of 802.15.4 with respect to other MAC protocols for a single hop WBSN.

Š 2010 ACEEE DOI: 02.AEE.2010.01.125

II.

INTRODUCTION TO 802.15.4

The IEEE 802.15.4 standard was created for low-rate, wireless personal area networks. Other existing standards for wireless communication are optimized for throughput, and are often not concerned with power consumption. Devices in these networks are either mains powered or their batteries are easily recharged. 802.15.4 is targeted for low cost, resource constrained devices that are deployed for lengthy periods of time without such maintenance as battery replacement. The application domain for such a standard includes wireless sensor networks, industrial and commercial control and monitoring, and home automation.

These devices would typically act as stick-on sensors, virtual wires, wireless hubs or cable replacements.

The standard is divided into two layers, the physical layer (PHY) and the media access control (MAC) layer.

These layers sit below the routing, e.g. Zigbee, or application layers (as shown in figure 1). The PHY and MAC layers provide building blocks for creating different network topologies, including star, mesh, and cluster tree networks. It is designed to operate on two classes of devices: reduced function devices (RFDs) and fully functional devices (FFDs). FFDs have the capability to communicate with any device in a network within range of them, while RFDs are only able to directly communicate with FFDs. Every network consists of multiple FFDs and RFDs, with one of the FFDs designated as the personal area network (PAN) coordinator.

In the following subsections, we give a brief overview of the 802.15.4 PHY and MAC layers, followed by an introduction into the configurability of these two layers.

A. 802.15.4 PHY The PHY layer specification dictates how 802.15.4 devices may communicate with each other over the wireless channel. It allows for the use of three frequency bands with varying data rates. The bit rates are 20 kb/s in the European

145

Proc. of Int. Conf. on Advances in Electrical & Electronics 2010

Fig. 2 The 802.15.4 MAC super-frame structure

Fig. 1 The 802.15.4 Protocol Stack

868 MHz band (868-868.6 MHz), 40 kb/s in the North American 915 MHz band (902-928 MHz), and 250 kb/s in the worldwide 2.45 GHz band (2.4-2.4835 GHz). This layer is responsible for activation and deactivation of the transceiver, channel frequency selection, and data transmission/reception. In addition, it performs channel energy detection (ED), link quality indication (LQI) for received packets, and clear channel assessment (CCA) for the MAC's carrier sense multiple access with collision avoidance (CSMA-CA) protocol. In addition to the packet length information and the PHY payload (the MAC protocol data unit), a PHY packet includes a 5 byte synchronization header (SHR) which allows devices to synchronize with the bit stream which forms the message.

B. 802.15.4 MAC The MAC protocol specifies when devices may access the channel for communication. The basic services provided by the MAC are beacon generation and synchronization, supporting PAN association and disassociation, supporting optional device security, managing channel access via CSMA-CA, maintaining guaranteed time slot (GTS) communication, providing message validation, and providing message acknowledgments.

A PAN may be set up in one of two basic configurations: beacon-enabled and nonbeacon-enabled. In a nonbeacon-enabled network, devices may communicate with each other at any time after an initial association phase. Channel access and contention are managed using an un-slotted CSMA-CA mechanism and any node-level synchronization must be performed at some higher layer. In a beacon-enabled network, the PAN coordinator periodically transmits a beacon which other devices use both for synchronization and for determining when to enable transmission and reception of messages. This beacon message is used to define a super-frame structure that all nodes in the PAN should synchronize to. This super-frame structure is shown in Figure 2.

Š 2010 ACEEE DOI: 02.AEE.2010.01.125

C. Beacon-Enabled Mode: A Detailed Description In beacon-enabled mode, beacon frames are periodically sent by the PAN coordinator every Beacon Interval (BI) to identify its PAN, to synchronize devices that are associated with it, and to describe the super-frame structure (Figure 2), comprising an active period and, optionally, an inactive period.

During the active period communication takes place, and during the inactive period, devices may turn off their transceivers in order to conserve energy.

The active portion of the super-frame structure is divided into 16 equally-spaced slots. Each one of these slots is further decomposed into smaller slots of length 320Âľs called "backoff periods" or "backoff slots", as shown in Figure 2. These backoff periods are the time unit of this mode of operation. Furthermore, there are three well differentiated parts in the active portion of the super-frame: 1) The Beacon: the beacon frame is transmitted at the start of slot 0. It contains the information on the addressing fields, the super-frame specification, the GTS fields, the pending address fields and other PAN related information.

2) The Contention Access Period (CAP): Immediately following the beacon is the CAP. During this period, devices may communicate using a slotted CSMA/CA mechanism. This is similar to un-slotted CSMA/CA, except that the back-off periods are aligned with slot boundaries, meaning that the devices are contending for the right to transmit over entire slots. The CAP must contain at least nine active period slots but may take up all available (16) slots.

3) The Contention Free Period (CFP): Following the CAP is an optional CFP, which may last up to seven active period slots. In the CFP, devices are allocated GTS slots by the PAN coordinator. During a GTS a device has exclusive access to the channel and does not perform CSMA/CA. Inside a GTS, a device may either transmit data to or receive data from its PAN coordinator, but not both. The length of a GTS must be an integer multiple of an active period slot. All GTSs must be contiguous in the CFP and are located at the end of the super-frame active period. A device may disable its transceiver during a GTS designated for another device in order to conserve energy.

146

Proc. of Int. Conf. on Advances in Electrical & Electronics 2010

III.

FUNCTIONAL OVERVIEW IN WBSN

There are three different types of data transmission in the IEEE 802.15.4 standard:

• Backoff Exponent (BE): enabling the computation of the number of waiting backoff slots before attempting to access the medium during a given backoff_ stage.

SIMULATION SETUP AND RESULTS

V.

•

Transmission from a device to the coordinator

•

Transmission from the coordinator to the device

• Transmission between any two devices.

However in the considered WBSN, only device to coordinator transmissions will take place. Since we will typically have sensing devices and a single central node in a star topology scheme, all the data traffic will be uplink, and only the first transmission type will hold. Steps followed by a transmission from a device to a coordinator n a beacon-enabled PAN:

In the simulations with ns-2.33 a many-to-one communication model is used representing a WBSN network application scenario where a number of nodes send data simultaneously to a single sink, the Coordinator (PAN coordinator). The simulation area is fixed to 50x50m. For 6 nodes the node density will be 6/(50*50) = 0.0024 nodes per m2. The network performance is measured in terms of energy consumption.

TABLE I SIMULATION PARAMETERS FOR NS-2

•

The device first listens to the network beacon.

•

When the beacon is found, the device synchronizes to the super-frame structure.

MAC Type

802.15.4/802.11/SMAC

Area

5 * 5 m2

•

The device transmits its data frame, using slotted CSMA/CA, to the coordinator.

Traffic Type

CBR

Packet Size

60 bytes

•

On its turn, it will transmit the data to the coordinator.

Simulation Time

200 and 1000 seconds

•

The coordinator may acknowledge the successful reception of the data by transmitting an optional acknowledgment frame. Acknowledgment frames are transmitted without using the slotted CSMA/CA mechanism to access the channel.

Capture Threshold

10 dB

Carrier Sense Threshold

6.879 X 10-11 watt

Receive Threshold

6.879 X 10-11watt

Bandwidth

250KHz

Transmit Power

0.0001W

Frequency

2.4 GHz

Fig. 3 Communication from a device to a coordinator in a beacon-enabled PAN

IV.

SLOTTED CSMA/CA MECHANISM

We have already discussed how the MAC of the IEEE 802.15.4 is based on either slotted or un-slotted CSMA/CA, depending on the network operation mode: beacon-enabled or non beacon-enabled modes, respectively. The CSMA/CA mechanism is based on backoff periods with duration of 20 symbols (or 320 µs).

Three variables are used to schedule medium access: • Number of Backoff’s (NB): representing the number of failed attempts to access the medium.

• Contention Window (CW): representing the number of backoff periods that the channel must be sensed before starting transmission.

© 2010 ACEEE DOI: 02.AEE.2010.01.125

Simulation is set-up using 7 nodes with node ‘0’ being the coordinator and all other nodes being the transmitting devices. Table 1 shows the parameters and their values for the simulation in NS-2. Figure 4 shows the Network Animator (NAM) instance of the simulation.

The 802.15.4 showed considerable energy efficiency for our single hop network when compared with other MAC protocols such as SMAC and 802.11(Figures 5, 6) for similar duty cycles. But if the duty cycle of 802.15.4 is kept very low keeping higher beacon orders, then the latency of association of nodes will be more (Figure 7).

However, the WBSN applications are meant to operate for years, and hence we can neglect the effect of initial association delay.

VI. CONCLUSION To evaluate the general performance of this new standard, we setup an NS2 simulator, which covers all the 802.15.4 PHY and MAC primitives, and carry out experiments for comparing the performance between 802.15.4, SMAC and802.11 for energy efficiency. We analysed the energy consumption and association time required by the considered scenario under IEEE 802.15.4 for different Beacon orders (Bo) and Superframe (So) orders. From our experiments and analysis it was found 147

Proc. of Int. Conf. on Advances in Electrical & Electronics 2010

that IEEE 802.15.4 protocol was far more energy efficient than the SMAC and IEEE 802.11 standards and more suitable for WBSN applications. But the association time required for higher beacon orders is a cause of concern, but can be ignored considering the prolonged usage of the application.

Fig. 7 Association time under different Beacon Orders

REFERENCES Fig. 4 NAM instance of simulation in NS-2

Fig. 5 Comparing 802.15.4 with other MAC Protocols for energy efficiency

Fig. 6 Energy consumption under different beacon and superframe orders

In this paper we have created an ideal scenario where all nodes are equidistant from the coordinator, but a more realistic WBSN configuration could also be made for future work. Also, the interference issues should be further investigated.

© 2010 ACEEE DOI: 02.AEE.2010.01.125

[1] Radosveta Sokullu, Cagdas Donertas, “Combined Effects of Mobility, Congestion and Contention on Network Performance for IEEE 802.15.4 Based Networks”, 978-14244-2881-6/08 IEEE 2008.

[2] Bruno Bougard, Francky Catthoor, “Energy Efficiency of the IEEE 802.15.4 Standard in Dense Wireless Micro sensor Networks”, Proceedings of Design, Automation and Test in Europe Conference and Exhibition (DATE’05) 15301591/05 IEEE [3] J. Heidemann, “An Energy-efficient MAC Protocol for Wireless Sensor Networks”, Proc INFOCOM 2002, New York, NY, June 2002.

[4] I. Lamprinos, A. Prentza, E. Sakka, and D. Koutsouris, “Energy-efficient MAC Protocol for Patient Personal Area Networks”, Conf Proc IEEE Eng Med Bio Soc, 4:3799, 2005.

[5] H. Li and J. Tan. “An Ultra-low-power Medium Access Control Protocol for Body Sensor Network”, Conf Proc IEEE Eng Med Biol Soc, 3:2451_4, 2005.

[6] Sofie Pollin, Mustafa Ergen, Sinem Coleri Ergen, “Performance Analysis of Slotted Carrier Sense IEEE 802.15.4 Medium Access Layer”, IEEE Transactions on Wireless Communications, VOL.7, NO. 9, SEPTEMBER 2008.

[7] V. Raghunathan et al., “Energy-aware Wireless Microsensor networks”, IEEE Sig. Processing, vol. 19, no. 2, Mar. 2002.

[8] J. Zheng and MJ Lee. “Will IEEE 802.15.4 make ubiquitous networking a reality?: a discussion on a potential low power, low bit rate standard”. IEEE Communications Magazine, 42(6):140_146, 2004.

[9] Wei Ye, Member, John Heidemann, “Medium Access Control With Coordinated Adaptive Sleeping for Wireless Sensor Networks”, IEEE/ACM transactions on networking, vol. 12, NO. 3, June 2004 [10] IEEE 802.15.4-2006 standard, Wireless Medium Access Control (MAC) and Physical Layer (PHY) specifications.

148