Efraim Pecht

Effective Lifetime-Aware Routing in Wireless Sensor Networks

Lifetime-aware routing and desired sensing spatial coverage (SSC) are two main challenges ahead of an ad-hoc, sensor-based, battery operated monitoring system known as wireless sensor network (WSN). Depending on the application, a necessary SSC level is essential to comply with the needed surveillance quality. On the other hand, network lifetime is of a major concern due to limited energy available to each sensor node. Formerly proposed lifetime-aware routing algorithms have usually defined lifetime as the duration before the first node runs out of energy. This criterion is not consistent with real-world WSN, where a number of sensors are likely to “die” due to hardware failures, natural impacts, etc. Initially, we propose a method that determines the network resource specifications, i.e., the number of available nodes and their sensing range, according to the required SSC and the necessary confidence level. Later on, a novel lifetime criterion which considers the SSC as the WSN effective operation criterion is introduced. Afterward, the new criterion is embedded into the flow augmentation algorithm and the normalized network lifetime is calculated for several scenarios using the solutions of the corresponding linear programming equations. Simulation results show significant improvement achieved in the lifetime. It is also shown that this new method is considerably more robust to the routing algorithm parameters compared to the performance achieved by the published Flow Augmentation algorithm.

FOG-based navigation in downhole environment during horizontal drilling utilizing a complete inertial measurement unit: directional measurement-while-drilling surveying

Presently used surveying sensors in directional drilling processes include accelerometers and magnetometers arranged in three orthogonal directions. The magnetometers in these setups are negatively affected by external magnetic interferences induced by various sources. Therefore, expensive, heavy and lengthy protective nonmagnetic collars need to be installed. Fiber-optic gyroscopes (FOGs) in an inertial navigation setup have been proposed as an alternative to magnetometer-based downhole surveying. The present study explored the feasibility of utilizing a FOG-based tactical-grade inertial measurement unit (IMU) as a complete surveying sensor for measurement-while-drilling (MWD) processes downhole. Alignment and real-time navigation under laboratory conditions were demonstrated. Analysis of vibrations and temperature as possible factors limiting the accuracy of the navigation process was performed. Severe vibration effects were reduced using software techniques, and a shock-absorbing housing was suggested. The temperature range of the IMU is limited by the optical components of the device, but dynamic temperature changes within this range did not present a major problem. A downhole sub design demonstrated that the actual integration of the IMU requires only minor changes in the presently used drilling tools. The utilization of a tactical-grade IMU eliminates the necessity of nonmagnetic collars, which results in lower costs and improved accuracy.

Observability Analysis for INS Alignment in Horizontal Drilling

Contemporary surveying in measurement-while-drilling (MWD) processes incorporates measurements from three-axes accelerometers and magnetometers. Unfortunately, magnetometer-related problems limit the navigation performance of this technique. The introduction of fiber-optic-gyroscope (FOG)-based inertial navigation system (INS) in MWD aims at overcoming these limitations. However, drifts in the measurements provided by the INS might be prohibitive for the long-term utilization of this modern navigation-method downhole. One of the main obstacles precluding the elimination of these measurement drifts is the limited observability of the azimuth angle state provided by the INS. This paper explores the feasibility of utilizing a FOG-based tactical-grade inertial measurement unit (IMU) as a complete surveying sensor for a MWD processes downhole by implementing an innovative in-drilling alignment (IDA) procedure. During IDA, the IMU is exposed to controlled dynamics that excites azimuth-related states. This allows better and faster alignment that can reduce long-term navigation drifts, thus improving the overall accuracy in INS-based MWD processes. It is suggested that one take advantage of the longitudinal space available in the drilling-pipe system and impose controlled motion on the IMU to excite its states and increase its observability. Theoretical simulations and analytical approximations exploring the IDA idea have shown reduction in the steady-state azimuth-error variance and in the time required to achieve convergence with the increase of the acceleration-controlled motion. Several practical aspects of implementing this approach are evaluated and compared.

Modeling of Observability During In-Drilling Alignment for Horizontal Directional Drilling

Navigation performance is an important factor in horizontal directional drilling. In-drilling alignment (IDA) was previously suggested to improve downhole navigation performance when utilizing an inertial navigation system (INS). It was shown that the IDA process enhances the ability to estimate INS bias and drift errors and, particularly, their azimuth-related components. It was suggested that this improvement was related to a better observability that is achieved with the help of the induced dynamics during the IDA phase. The observability of a system is an important parameter that facilitates the estimation of the state parameters and the achievable accuracy of the system. However, observability models that are related to the IDA technique are lacking. This paper presents observability modeling of the newly suggested IDA process to aid horizontal drilling. The presented methodology clearly demonstrates that an induced motion during the IDA process increases system observability and converts the azimuth angle into an observable system state. Adequate system modeling profoundly influences the overall system observability. The utilization of the vertical damped model with a reduced state order is preferable for a faster and more efficient performance due to a decreased computational load but remains inferior when compared to a full system model.

Improved lifetime routing for Wireless Sensor Networks

Wireless Sensor Networks (WSNs) are used for various monitoring tasks. Their utilizations in remote areas, where battery replacement is not feasible, require extending the network lifetime. A modified version of the popular Collection Tree Protocol (CTP) is presented - the Lifetime Improved CTP (LICTP). It is shown that the LICTPs achieved lifetime for the evaluated scenario increases by a factor of 2.55 and for actual real world cases can get, for the defined scenario, up to 7 years.

Performance evaluation of lifetime-aware routing in Wireless Sensor Networks with practical design considerations

Energy resource management is considered essential in battery-operated Wireless Sensor Networks (WSNs). Radio communication circuitry is a significant power consumer in a WSN. Therefore, it is desired to design an efficient routing algorithm which conserves the battery power as much as possible, resulting in a longer network lifetime. Lifetime-aware routing protocols were suggested to distribute the flow of information through different routes. These works were mainly based on simplified assumptions which ignored practical considerations and assumed perfect knowledge of the channel conditions and the ability to adjust the transmission power accordingly. In this work, the practical case of a constant transmission level is applied to the Flow Augmentation algorithm and the lifetime performance of the proposed modified Flow Augmentation algorithm called Constant Transmission Power Flow Augmentation (CTPFA) is evaluated and compared with the original Flow Augmentation. The CTPFA analysis proves that lifetime of a WSN with nodes that have constant transmission power decreases significantly due to the excess transmission power used. Methods to improve the lifetime performance are discussed.

Confidence level analysis of Sensing Spatial Coverage in Wireless Sensor Networks

Achieving required Sensing Spatial Coverage (SSC) is considered a main challenge in Wireless Sensor Networks (WSN). Earlier works used the asymptotic analysis to determine the required number of sensors to meet the necessary sensing coverage in a randomly deployed WSN. Recent works have shown that there is an overestimation in the asymptotic analysis and proposed other analytical solutions to overcome this issue. In this work, a novel concept called the coverage ‘confidence level’ is introduced. It overcomes a lack of reliability that previous works ignored in their calculations. The analysis provides an important practical tool that relates the required sensing spatial coverage to WSN key parameters: the number of available nodes and their sensing range ability.

Lifetime-improved Collection Tree Protocol for Wireless Sensor Networks

Wireless Sensor Networks (WSNs) are generally deployed in remote sites in order to sense the environment and acquire a set of data useful for analysis, monitoring and optimization algorithms. The use of the WSNs in remote and rugged areas, where battery replacement is not practical, requires extending the network lifetime. The proposed solution in this work, is using low-power Raspberry Pi platform alongside TinyOS operating system. This platform utilizes the novel Lifetime Improved CTP (LICTP), an enhanced version of the Collection Tree Protocol (CTP) presented in this work. The LICTPs performance is analyzed and studied by various simulations using different numbers of MEMSIC IRIS motes. Further, simulation results are verified by an experiment using a network composed of 25 IRIS motes. A demonstration of the LICTP lifetime improvement is presented and evaluated. The proposed WSN improvements achieved a significant enhancement in lifetime duration. For the implemented case-study, the lifetime improvement with LICTP utilization is by a factor of 2.

Sensing Spatial Coverage analysis - for fire detection purposes

Sensing Spatial Coverage (SSC) is one of the main Quality of Service (QoS) measures in a Wireless Sensor Network (WSN). This quantifies the WSN monitoring ability. The initial phase of WSN implementation is to determine how many sensors with specific sensing ability are required for the intended application or scenario. For this purpose, the sensing profile of the sensors should be known. Most of the previous works considered a Boolean model for the sensors, limiting the sensing ability to a fixed radius known as the sensing range. The sensing ability of the sensors depends on the nature of the application they are used for. In this work fire detection is assumed to be the application and the sensor consists of a simple Infra-red (IR) photodiode. A realistic sensing model will be introduced and the detection ability of the sensor will be studied in detail and closed form formulas will be derived for the probability of fire detection in different distances.