Source-linked AI summary
Coverage Protocols for Wireless Sensor Networks: Review and Future Directions
Riham Elhabyan, Wei Shi, Marc St-Hilaire
TL;DR
The paper addresses how effectively WSN fields are monitored and synthesizes the fragmented coverage-protocol literature. It proposes a taxonomy and three-stage classification, then identifies realism and connectivity as central directions for future protocol design.
Problem
The survey addresses the WSN coverage problem of effectively monitoring a network field, which directly affects sensor energy consumption and network lifetime.
Method
It develops a taxonomy and reviews coverage-aware deployment, flat-network sleep scheduling, and cluster-based sleep scheduling protocols.
Results
The review finds that coverage protocols are mainly limited by unrealistic coverage assumptions, including location awareness, uniform signal strength, and idealized energy models.
Takeaways & Limitations
Future protocol designs should consider realistic sensing, energy, connectivity, and localization models, with network connectivity treated as crucial.
Abstract
from arXiv · showhide
The coverage problem in wireless sensor networks (WSNs) can be generally defined as a measure of how effectively a network field is monitored by its sensor nodes. This problem has attracted a lot of interest over the years and as a result, many coverage protocols were proposed. In this survey, we first propose a taxonomy for classifying coverage protocols in WSNs. Then, we classify the coverage protocols into three categories (i.e. coverage aware deployment protocols, sleep scheduling protocols for flat networks, and cluster-based sleep scheduling protocols) based on the network stage where the coverage is optimized. For each category, relevant protocols are thoroughly reviewed and classified based on the adopted coverage techniques. Finally, we discuss open issues (and recommend future directions to resolve them) associated with the design of realistic coverage protocols. Issues such as realistic sensing models, realistic energy consumption models, realistic connectivity models and sensor localization are covered.
I. Introduction
The survey frames coverage as a fundamental WSN problem involving effective field monitoring, with consequences for sensor energy consumption and network lifetime. Coverage problems and protocols can be classified along multiple dimensions.
- Coverage measures how effectively sensor nodes monitor the network field.
- Coverage is fundamental because it directly affects sensor energy consumption and network lifetime.
- Coverage problems are classified by monitoring frequency, region of interest, and required coverage degree.These dimensions distinguish continuous versus sweep coverage, area/point/barrier coverage, and 1-coverage versus K-coverage.
- Coverage protocols are classified by connectivity requirement, algorithm characteristics, and network system model.Connectivity-aware versus non-connectivity-aware and distributed versus centralized protocols are explicitly identified.
A. Related Reviews
Earlier reviews addressed important subsets of WSN coverage, but the survey argues that they did not comprehensively classify the problem’s features and protocols. It therefore compares reviews across surveyed dimensions and presents a broader taxonomy.
- Existing surveys often focus on subsets of coverage features and do not classify coverage problems in their entirety.
- Some reviews restrict analysis to area coverage, static nodes, Boolean sensing, or directional barrier coverage.
- Prior reviews separately emphasize k-coverage, evolutionary sleep scheduling, energy-efficient coverage, coverage-connectivity relationships, or barrier coverage.Their scopes include particular protocol families, sensing models, deployment settings, or barrier categories.
- This survey develops a comprehensive taxonomy from protocol features and reviews three protocol groups: deployment, flat-network sleep scheduling, and cluster-based sleep scheduling.
- Table 1 compares related reviews with this survey across dimensions of the surveyed protocols.
B. Our Contributions
The survey contributes a broad classification and current review of WSN coverage protocols, together with discussion of realistic energy-efficient designs and future directions for addressing unrealistic assumptions.
- The paper presents a broad discussion and clear classification of coverage protocols in WSNs.
- It provides an in-depth review of up-to-date protocols designed to solve different coverage problems.
- It discusses open issues in designing realistic energy-efficient coverage protocols for WSNs.
- It recommends future directions for addressing unrealistic assumptions in previously proposed protocols.
C. Paper Organization
The paper reviews coverage protocols by the network stage where coverage is optimized, then discusses open research issues and concludes with future directions.
- Sections II through V provide background and review protocols in three categories based on the network stage where coverage is optimized.The categories are coverage-aware deployment, flat-network sleep scheduling, and cluster-based coverage protocols.
- Section VI discusses four open research directions: realistic sensing, coverage-aware clustering, realistic connectivity, and sensor localization.
- Section VII concludes the paper and highlights future research directions.
II. Preliminaries
The preliminaries introduce coverage issues in WSNs, outline the section’s background topics, and provide an acronym reference.
- The section establishes background knowledge about coverage issues in WSNs before discussing design factors, coverage problems, and sensing models.
- The paper provides a dedicated list of acronyms for reference.
A. Design Factors
Coverage protocols are shaped by coverage requirements, deployment settings, algorithm organization, sensing models, and interactions with connectivity and energy consumption.
- Coverage requirements: Coverage requires every region point to be monitored by at least one sensor, while k-coverage requires monitoring by k or more sensors.Higher coverage degree provides redundancy, fault tolerance, and stronger monitoring in applications such as intrusion detection and military operations.
- Deployment models: Sensor deployment may be random or deterministic depending on application requirements and environmental accessibility.Random deployment suits large-scale, dangerous, or inaccessible environments, whereas deterministic deployment suits controlled, human-friendly settings.
- Algorithm organization: Coverage optimization can use distributed or centralized algorithms, with many coverage problems formulated as NP-hard optimization problems.Distributed nodes use neighbor information, whereas centralized protocols forward data to a central unit such as a base station.
- Coverage types: Coverage problems may involve continuous or periodical monitoring, including area, point, barrier, and sweep coverage.Sweep coverage uses mobile sensors to collect data about points of interest when continuous field coverage is unnecessary.
- Connectivity: Coverage solutions should consider connectivity because achieving coverage alone cannot guarantee reliable communication with a base station.The paper notes that coverage and connectivity should be jointly investigated for proper deployment.
- Sensing models: Sensing models represent sensor sensing ability and quality, including directional, omnidirectional, and Boolean models.Directional sensing uses a finite field of view, while Boolean sensing treats points within radius R as covered.
C.4 Probabilistic Sensing Models
Probabilistic sensing models extend Boolean sensing by representing uncertain, distance-dependent detection and shadowing effects in realistic environments.
- Probabilistic sensing: Probabilistic sensing models address imprecise detection by expressing coverage as a probability that decreases with distance from the sensor.They are presented as a more realistic extension of the Boolean sensing model.
- The Elfes Sensing Model: The Elfes model uses Rmin and Rmax to separate definite detection, uncertain detection, and definite non-detection.Detection is certain within Rmin, probabilistic between Rmin and Rmax, and absent beyond Rmax.
- The Elfes Sensing Model: The Elfes model becomes Boolean when Rmin = Rmax.The λ and γ parameters are adjusted according to the physical properties of the sensor.
- The Shadow Fading Sensing Model: The shadow fading model accounts for obstructions and non-uniform, asymmetric sensing caused by environmental and hardware effects.Its coverage probability depends on shadowing, distance, and an average sensing radius.
- Survey scope: The survey compares coverage protocols across sensing models and includes a realistic model reflecting radio irregularity in WSNs.Its taxonomy organizes protocols by the network stage where coverage is optimized.
V. Cluster-based Sleep Scheduling Protocols
Cluster-based coverage protocols jointly address clustering, coverage, connectivity, energy efficiency, or sensor activity rather than selecting cluster heads solely for energy management. The reviewed approaches include distributed, centralized, optimization-based, and mobile-sensor strategies, while some assumptions can create coverage holes.
- Existing clustering protocols often prioritize energy-based Cluster Head selection while leaving effective network-area coverage unaddressed.
- ECDC favors Cluster Head candidates with higher residual energy, denser deployment, or greater point-of-interest coverage.
- BCA forms approximately coverage-balanced clusters and favors candidates in densely populated areas.
- Some cluster-based procedures impose a Cluster Head threshold without redundancy checking, creating potential coverage holes.
- MOEAD-CCP encodes four sensor states and uses two objectives to jointly address clustering and coverage, but its objective functions are not well-defined.
- A centralized protocol for 3D WSNs jointly optimizes each sensor’s cluster-head, active, or inactive/sleep status to maintain connectivity and coverage.
A. Realistic Sensing Model
The survey argues that idealized sensing assumptions limit realistic coverage analysis, because sensing strength varies spatially and radio signals can be directionally irregular. It presents shadow fading and RIM as more realistic modeling directions, while noting that RIM has not been applied to WSN coverage.
- Most coverage protocols use a boolean sensing model that assumes an abrupt transition from full sensing strength to zero at a uniform range.
- The boolean model can undercount coverage beyond its defined range, keeping redundant sensors active and reducing network lifetime.
- The same model can overestimate coverage within its range, producing coverage holes and unmet application requirements.
- Elfes sensing model II.C.4.a adds two sensing ranges, but both remain uniformly distributed and retain boolean-model limitations.
- Empirical studies report that coverage-protocol performance degrades as location errors, sensing-signal irregularity, and packet losses increase.
- RIM models directional radio irregularity through DOI-adjusted path loss values, yet it has not been studied for WSN coverage.
B. Realistic Coverage-aware Clustering Protocols
Coverage-aware clustering should jointly optimize energy efficiency and area coverage using realistic sensing, radio, and energy models. The survey identifies sleep scheduling and integrated realistic-model solutions as future directions.
- Clustering improves network energy efficiency, but many protocols overlook effective area coverage and face challenges in determining accurate radio models.
- The commonly used radio energy model applies distance-dependent free-space or multipath amplification terms after a distance threshold.Eelec denotes transmitter or receiver circuitry energy; d0 is the distance threshold; εfs and εmp denote amplification energies.
- The conventional energy model ignores listening energy and assumes infinite communication range, making it idealized for sensor-network radio power consumption.
- A discrete radio model can calculate power consumption more realistically and determine which sensor links are available for transmission.The proposed realistic model can use Chipcon CC2420 transceiver characteristics.
- Realistic node energy accounting includes sleep, reception, and transmission states, their durations, and transition energy.
- Further energy-efficiency gains may come from adding sleep scheduling to realistic clustering solutions and integrating realistic energy and sensing models.
C. Realistic Connectivity Model
Realistic connectivity requires link-quality models beyond inter-sensor distance and should be considered jointly with coverage and localization. RSSI provides a supported alternative, while localization assumptions remain costly.
- Coverage and connectivity should be investigated jointly, but existing protocols often evaluate links by distance and assume coverage plus sufficient range guarantees connectivity.
- Distance-based connectivity requires locating or distance-measuring hardware, increasing solution cost, and fails to represent link asymmetry.
- Several studies show that link quality is not necessarily correlated with distance.
- RSSI and LQI are chip-provided link-quality metrics; RSSI represents received signal strength, whereas LQI characterizes received-packet strength and quality.
- An RSSI of -87dBm or stronger is almost, but not completely, associated with achieving a ≥99% packet delivery rate.Below this level, a 2 dBm RSSI shift can change a good link to a bad link.
- RSSI shows significant bidirectional correlation and should be used to assess link quality, whereas LQI shows weak correlation between link directions.
- GPS-based location awareness simplifies protocol design but can make coverage solutions costly and unrealistic.
- Range-based localization measures distances or angles, while range-free localization estimates positions using network connectivity.
F.4 Multi-objective Approach
The survey critiques weighted-sum formulations for multi-objective WSN optimization and motivates solution encodings that jointly represent deployment, localization, clustering, connectivity, and lifetime objectives.
- Weighted-sum formulations are computationally efficient and straightforward but return only one optimal solution.
- Weighted sums cannot find an optimum when the feasible objective-space solution set is non-convex.
- Weight-vector selection can bias solutions toward one sub-objective, especially when objectives conflict and trade-off selection is required.
- Jointly solving deployment, localization, clustering, connectivity, and lifetime maximization requires a new individual encoding scheme representing the combined solution.
VII. Conclusions
The survey concludes that realistic sensing, connectivity, and energy assumptions are central limitations of existing coverage protocols. It recommends discrete radio models, anisotropic sensing models, and connectivity-aware future designs.
- Existing coverage protocols are mainly limited by unrealistic coverage assumptions, including location awareness and uniform signal strength within sensing or communication ranges.
- The survey recommends discrete radio models for more accurate power calculations and better transmission-link selection.
- The authors compare known protocols across design factors and identify realistic anisotropic sensing and network connectivity as future research priorities.