25-10-2016, 09:18 AM
An Energy Efficient Wireless Sensor Networks for Bicycle Performance Monitoring Application
1460955154-wsnpaper.doc (Size: 1,003.5 KB / Downloads: 5)
Abstract:
Wireless sensor networks (WSNs) have greatly advanced in the past few decades and are now widely used, especially for remote monitoring; the list of potential uses seems endless. Three types of wireless sensor technologies (Bluetooth, ZigBee, and ANT) have been used to monitor the biomechanical and physiological activities of bicycles and cyclists, respectively. However, the wireless monitoring of these activities has faced some challenges. The aim of this paper is to highlight various methodologies for monitoring cycling to provide an effective and efficient way to overcome the various challenges and limitations of sports cycling using wireless sensor interfaces. Several design criteria were reviewed and compared with different solutions for the implementation of current WSN research, such as low power consumption, long distance communications, small size, and light weight. Conclusions were drawn after observing the example of an advanced and adaptive network technology (ANT) network highlighting reduced power consumption and prolonged battery life. The power saving achieved in the slave node was 88–95% compared to the similar ANT protocol used in the medical rehabilitation
Introduction
Cycling requires a very high level of endurance supported by a high fitness level and a suitable training program and teamwork. Cycling is a highly competitive sport in which the margins between victory and defeat are typically measured in seconds or milliseconds. Statistical information can be used to further improve and refine team skills and cyclist training programs to increase competitiveness in international tour-naments. Acquired data may be processed to produce athlete performance profiles. Comparisons of athlete profiles can be made occasionally to measure an athlete’s progress and to monitor and supervise him or her so that a coach can correct any deficient parameters.
In this paper, different types of wireless sensor networks (WSNs), which are used to monitor the biomechanical and physiological parameters of the bicycle and cyclist, respec-tively, will be critically reviewed and compared. A compari-son between these wireless sensor technologies will be carried out on the basis of several performance metrics related to
physical activities involving both the bicycle and cyclist. Some biomechanical and physiological parameters relevant to the sensors are covered in terms of their specifications and applications to cycling monitoring, including heart rate, speed, cadence, and pedal power.
The purpose of this review is not to criticize but rather to formulate potential extensions to existing approaches and to understand future research challenges. In addition, a methodical review of the use of WSN in cycling applications is used to describe current WSN research implementations. The review compares wireless technologies and performance metrics from recent research. Also, an ANT network example was introduced to overcome the limitations of power con-sumption and battery life.
The contribution of this paper is to present a com-prehensive overview of the many challenges that face the wireless sensor networks in the bicycle performance mon-itoring system. In addition, comparisons between different wireless technologies were done based on various perfor-mance metrics. We attempted to highlight several wireless
sensor network technologies that are used in wireless sensor bicycle monitoring, thus providing a comprehensive, up-to-date survey of wireless sensor networks research that is used in bicycle performance monitoring system. A new wireless sensor technology (i.e., ANT) that is used in bicycle performance monitoring was introduced, which aims to reduce the power consumption and prolong battery life of the bicycle performance monitoring system based on message rate reduction, as will be presented in Section 6. From the literature review, we found that ANT wireless technology offers significant potential in the area of energy efficient WSN transmissions. Our focus in this paper is on several wireless protocols which are connected to several bicycle-related sensors. Furthermore, this survey provides a good number of latest references for those interested in the area of wireless sensor bicycle performance monitoring. Based on our limited knowledge, this review paper is the f irst paper that deals with the extensive survey of wireless sensors for bicycle performance monitoring, focusing on energy efficient WSN protocols.
The rest of the paper is structured as follows. Section 2 describes the problem statement of the cycling team. The comparisons of types of wireless technologies used to transfer data from the cyclists will be presented in Section 3. T he common biomechanical and physiological parameters in the cycling sport will be introduced in Section 4. Section 5 cate-gorizes and describes previous work according to the type of wireless technology, highlighting the research methodology, advantages, and disadvantages of each research. In addition, a table of comparison for the research will be summarized. Fur-thermore, the biomechanical and physiological signals and sensors related to both bicycle and cyclist will be described in the same section. Section 6 will explain a proposed example of bicycle wireless sensor monitoring. Future works and a maturity evaluation will be introduced in Section 7. Finally, conclusions will be presented in Section 8.
2. Problem Statement and Challenges
Before advances in sensor development technologies, a cycling training program once depended mainly on the conventional stopwatch [1] to measure the performance of athletes. This method is inaccurate and uninformative when used to measure the performance and capability of an athlete on the track. Furthermore, this method also lacks scientific support, as there is no evidence that training with a stopwatch can improve the competitive performance of an athlete. There are many challenges when attempting to implement wireless sensor devices and networks specifically in bicycle performance monitoring systems as follows.
(i) The monitored parameters of the bicycle must be real time and accurate, so that coaching staff can give instant feedback to improve the performance of cyclists.
(ii) The wireless sensor monitoring system must operate at low power consumption (i.e., the battery must last for at least five hours of training).
There must be a flexible transmission range between the cyclist and coaching teams as it depends on the type of discipline, such as, track, road, or cross-country.
(iii) Ideally, lightweight and small-size wireless sensors and display units (mounted on the bicycle) are desired, especially for a highly competitive professional tournament.
(iv) The system should expand to include additional (new) measurement parameters, such as body temperature, left/right axis (gyro sensors), specific athlete power profiles, and stress level.
(v) An efficient algorithm must be introduced to reduce the power consumption of the three main consumers of power in the WSN, namely the sensors, microcontroller, and transceiver.
3. Wireless Technologies
There are many wireless technology standards which have been adopted for bicycle monitoring systems. Two of the most commonly available wireless protocols are ZigBee [2] and Bluetooth [3]. There is also alternative wireless standard ANT, which is commonly used in sensor networks [4, 5]. As is customary, wireless technology requirements such as transmitted power (dBm), data rate (kbps), the range of communications (m), power consumption (W), and weight (kg) differ among these standards. Practically, the power consumption, weight, and size must be as small as possible and the coverage area of the network must be as large as possible to improve the performance of the cyclist. For example, one worst-case scenario occurs when cyclists and coach are positioned too far apart.
Three wireless standards occupying the 2.4 GHz ISM band used in cycling monitoring are Bluetooth, ZigBee [6], and ANT, where ANT is commonly used in commercial products. Bluetooth operates in the 2.4 GHz ISM band, with a broadband speed 1 to 24 Mbps depending on the Bluetooth version [7] and a communication range of approximately 1 m/10 m/100 m [8]. Moreover, Bluetooth is a “lighter” standard that is extremely ubiquitous (e.g., it is built into most mobile phones, PDAs, and laptops) and can easily be adapted to several other services or networks [9].
ZigBee is a wireless technology that operates in the 868 MHz, 915 MHz, and 2.4 GHz frequency bands [10]. ZigBee was designed mainly for low-power-consumption applications, with additional features such as low cost, low data rate (maximum of 250 Kbps), and long battery life. In many ZigBee applications, the wireless device participation or contribution is very limited; the device spends most of its time in sleep mode. As a result, ZigBee device batteries are capable of working for several years and do not need to be replaced [11]. ZigBee enables broad-based deployment of wireless networks with an approximate communication range of 100 m. This allows it to work for years on small coin batter-ies for a host of control and monitoring applications, such as smart grid [12], patient monitoring [13], automation systems [14], control systems [15], and several other applications.
4. Biomechanical and Physiological Parameters in Cycling
Most research works are concerned with four common bicycle parameters: (i) speed, (ii) heart rate, (iii) cadence, and (iv) power. These four parameters are key factors in improving the performance of cyclists, which are explained as follows.
4.1. Speed or Tire Rotation. The tire rotation or speed of the bicycle is the number of wheel rotations or revolutions per minute (RPM). Knowing the speed allows cyclists to avoid overtraining by regulating the training load [45] and it quantifies the external load [46]. The speed or tire rotation can be measured by using a magnetic sensor (reed switch) and a permanent magnet. The magnetic sensor is placed on the fork blade on the front wheel or on the chainstay of the rear wheel and the permanent magnet is fixed on one spoke on the front or rear wheel. The number of pulses generated by the sensor is directly proportional to the rotations of the rear or front wheel (RPM) and the ground speed of the bicycle can be calculated in kilometers per hour (KPH) or miles per hour (MPH), based on the following equation:
2
=RPM ⋅ ⋅( ) = ⋅ , (1)
60
where is the wheel radius in meter and RPM is the wheel angular velocity and (2 /60) is the unit transformation of revolution per minute into radians per second.
4.2. Heart Rate. The heart rate (HR) is the number of heart beats per minute. The heart rate monitor device is usually mounted on the athlete’s body in the form of a belt around the chest that contains a pulsed wireless sensor. T he HR device is able to measure and evaluate the physical health condition of the athlete. Knowing the heart rate is important to allow each cyclist to achieve his or her training plan as much as possible; the difference between the training heart rate and the reference value of each cyclist must be as small as possible [47]. T he heart rate is also used to assess the physical stress or internal load that results from an external load [46]. The heart rate in cycling is interchangeable with the workload (watts) made by the cyclist; for example, the heart rate is 91 beats/min when the workload is 25 W, while it becomes 137 beats/min when the workload is equal to 100 W [48].
4.3. Cadence. Cadence is the measure of the spinning speed of the pedals. The unit of measurement is revolutions per minute (RPM). The cycling power can be increased and the cyclist can be a productive part of the cycling team if the cyclist optimizes the cadence of the bicycle. Knowing the cadence allows cyclists to remain in a secure and an efficient range of revolutions per minute (RPM) to protect their knees [49], which is especially important over long distances. It is also possible to use the cadence to estimate the number of calories consumed during a bicycle trip. In addition, maintaining an ideal cadence (60–80) RPM helps a cyclist to consume calories efficiently. The cadence can be measured using a cadence sensor, which has two elements: a permanent magnet attached to the right crank arm of the bicycle and a magnetic sensor (reed switch) mounted on the chainstay. The pulses generated by the magnetic sensor are directly proportional to the crank rotations as measured in revolutions per minute (RPM).
Previous Works
Over the past decades, the issues of mobility, size [56], capability of processing (i.e., microprocessors or microcon-trollers), flexible memory, and robust modes of transmission have been solved in wireless sensors due to the advancement in wireless sensor technology. Many new wireless standards provide a reliable and secure platform for a sensor network. Such wireless sensors support wearable [57], unobtrusive [58] equipment and also benefit from remote monitoring. This development benefits many practical applications including those in the realm of sports. In sports applications, a WSN allows the gathering and use of data in real time and it is also possible to compare data among athletes.
The transfer of data from an athlete or patient to a remote data processor or storage may now be performed remotely. The most common way to transfer the data is simply point-to-point, which means that one transmitter on the side of the athlete and one receiver on the monitoring side communicate based on frequency-modulated transmission
In this section, several types of bicycle wireless sensor monitoring (BWSM) are critically reviewed. This review will be categorized based on the type of wireless technology or platform.
5.1. Systems Based on ZigBee Platforms. Muscle work, sweat level, and heart rate were monitored by Armstrong [6]. This study represents a beneficial introduction to how the ZigBee wireless technologies can be employed to improve the measurement process.
Advantage. Power consumption is reduced.
Disadvantage. Small coverage area limits communication due to limitations of the ZigBee standard.
The pedal power, speed, cadence, and heart rate data were transmitted separately for each cyclist to his computer via a ZigBee network in the work of Kuhn et al. [45]. They compared these measured parameters while developing the ABT system with commercial Schoberer Rad Messtechnik (SRM) power meters for twelve to thirty hours of bicycle riding. They found this system to be quite convincing.
Advantages. The system has low cost and lightweight over-head for cyclists and the transceiver unit can be used with multiple radio channels. Thus, many networks may be active at the same place in the same time period.
Disadvantages. The transmission range is 50 m in outdoor conditions and the notebook size is large. Another limiting factor is battery capacity, especially when the power plug is not available.
Le et al. [47] built an assisted bicycle trainer (ABT) with a predictive controller to measure the parameters of the bicycle and rider: speed (RPM), position in the group, and heart rate (bpm) of the cyclists. The collected data were transferred via a ZigBee network to a notebook computer for monitoring and supervision.
Advantage. The predictive controls of the ABT system were an effective way of training the cyclist group.
Disadvantages. The system did not consider other parameters such as power, velocity, cadence of the bicycle, and wind velocity, which were left for future work.
Walker et al. [59] monitored pulse heart rate (bpm), SpO2 (%), skin temperature (∘C), ambient temperature (∘C), bicycle speed or wheel speed (KPH or MPH), and bicycle location based on a remote mobile monitoring system (RMMS). In addition, this system may be used to monitor vital events for the elderly.
Advantages. Low-power wireless sensors (ZigBee module), light weight, and real-time application (the time to transmit data from the sensor location to the remote device is approx-imately 10 ms); monitoring can be performed by persons not admitted to the hospital or in a physician’s office This last advantage improves retrieval of the history of a person’s status and allows a medical care center to service a huge number of people, reducing hospital visits and health care expenses, and enables the observer to recognize emergency situations quickly.
Disadvantage. The wireless sensor nodes (BioTE) have dif-ficulty communicating when an obstacle blocks the line-of-sight between transmitter and receiver nodes.
Measurements of wheel spin rates (RPM), angular veloc-ity (RPM) of the bicycle, steer angle, and the acceleration of three points on the frame were performed by Peterson et al. [60]. They were more ambitious than others in using a low-power-consumption Xbee-pro wireless module interfaced to a microcontroller based on the Arduino platform to transmit the measured parameters to a monitoring system, based on GUI toolkits.
Advantages. The system created an accurate assessment of the lean via numerical integration and precise filtering of gyroscope signals, and the optical encoders are frictionless and have high reliability at low speeds with a fast dynamic response.
Disadvantages. T he roll of the bicycle frame is dif f icult to directly measure, while the high precision of the optical methods is countered by both the weight and expense of the head unit. Optical methods can also be af fected by road surface bending; reflectance and a larger battery size were required to supply the system.
Location estimation and radio reachability of the bicycle were determined by Hayashi et al. [61] using the ZigBee WSN. The bicycle’s location can be estimated based on base sta-tions, internet protocol gateways, and network management servers.
Advantages. Low power consumption, wide-area sensor net-work, monitoring over a long distance of 3.5 km, remote read-ing, and remote control and bicycle management (including a temperature sensor, GPS sensor, luminance sensor, and acceleration sensor). The project found new services that are secure, eco-friendly, and safe.
Disadvantages. Large infrastructure, more components in the network, and the required line-of-sight between the transmission terminal and the base station.
Olieman et al. [62] measured vibrations of the bicycle produced from a combination of the wheels, speed, tire pressure, and road surface. Wireless inertial acceleration sensors (ProMove2) were used to measure the vibration of the bicycle in real time.
Advantages. Real-time measurements; the WSN nodes are connected to a computer by the fast gateway. GUI software is used to display the acquired data for further processing, while the receiver is designed to store the data every second, and automatic pausing is turned off.