Crowdsourced Tracking and Localization over Wireless Networks
This note organizes the locally collected literature around one concrete theme:
crowdsourced location tracking in mobile or motion-rich scenarios over wireless networks
The emphasis is on papers that involve one or more of the following:
- mobile users carrying phones or tags,
- wireless measurements such as Wi-Fi, BLE, Bluetooth broadcasts, or RSS/CSI,
- crowdsourced data collection or crowd-assisted discovery,
- motion, trajectories, path reconstruction, mobility analysis, or proximity tracking.
1. Fast Reading Order
If the goal is to quickly build a mental map of the area, read in this order:
SecureFindfor crowd-assisted object finding.Who Can Find My Devices?for a real deployed crowd-finding system.Zeefor zero-effort crowdsourced indoor localization.Smartphones Based Crowdsourcing for Indoor Localizationfor smartphone-driven crowdsourced fingerprinting.Communicating Is Crowdsourcingfor motion-aware Wi-Fi localization with trajectory information.A Robust Crowdsourcing-Based Indoor Localization Systemfor trajectory radio maps and robust crowdsourced localization.Hotspot Ranking Based Indoor Mapping and Mobility Analysis Using Crowdsourced Wi-Fi Signalfor mobility analysis from crowdsourced Wi-Fi traces.ProTrackfor passive wireless proximity and trajectory inference.CrowdLOC-SandSemi-Self Representation Learning for Crowdsourced WiFi Trajectoriesfor newer ML-oriented extensions.
2. Group A: Crowd-Assisted Object Finding and Device Tracking
These papers are the closest match if the task is "use nearby participants or nearby devices to help find a moving or misplaced object."
2.1 SecureFind: Secure and Privacy-Preserving Object Finding via Mobile Crowdsourcing
- Local file:
papers/crowdsourcing/securefind_secure_privacy_preserving_object_finding_mobile_crowdsourcing_2015.pdf - Why it matters: It is a foundational paper for crowd-assisted object finding. The target object is found through mobile participants, and the design focuses on privacy and security.
- Best use: Read this first if the problem resembles "many mobile users help search for a tagged object."
- Keywords: object finding, mobile crowdsourcing, privacy-preserving search, wireless tagging.
2.2 Who Can Find My Devices? Security and Privacy of Apple's Crowd-Sourced Bluetooth Location Tracking System
- Local file:
papers/tracking/who_can_find_my_devices_security_privacy_apples_crowd_sourced_bluetooth_location_tracking_system_2021.pdf - Why it matters:
It studies Apple's deployed
Find Myecosystem and reconstructs how offline devices broadcast rotating BLE identifiers and how nearby finder devices upload encrypted location reports. - Best use: Read this when you want a real-world large-scale crowd-finding architecture rather than only an academic prototype.
- Keywords: Bluetooth, BLE, finder devices, encrypted reports, crowd-assisted tracking.
3. Group B: Mobile Crowdsourcing for Indoor Localization and Trajectory Reconstruction
These papers are most relevant if the task is "users move around while their devices passively contribute wireless or sensor data that improves localization."
3.1 Zee: Zero-Effort Crowdsourcing for Indoor Localization
- Local file:
papers/tracking/zee_zero_effort_crowdsourcing_for_indoor_localization_2012.pdf - Why it matters: This is one of the classic papers on zero-effort crowdsourced indoor localization. It uses inertial sensing and Wi-Fi scans collected while users naturally walk through the environment.
- Best use: Read this first if you want the standard reference for passive, no-extra-effort crowdsourced localization.
- Keywords: Wi-Fi, inertial sensing, walking trajectories, zero-effort crowdsourcing.
3.2 Smartphones Based Crowdsourcing for Indoor Localization
- Local file:
papers/tracking/smartphones_based_crowdsourcing_for_indoor_localization_2015.pdf - Why it matters: This paper is a strong bridge between classical fingerprinting and practical smartphone-based crowdsourcing. It is useful for understanding how ordinary user motion can be turned into localization data.
- Best use:
Read it after
Zeeto see how smartphone-based crowdsourcing was pushed toward more systematic indoor localization. - Keywords: smartphones, wireless fingerprints, crowdsourced localization, mobile sensing.
3.3 Communicating Is Crowdsourcing: Wi-Fi Indoor Localization with CSI-based Speed Estimation
- Local file:
papers/tracking/communicating_is_crowdsourcing_wifi_indoor_localization_csi_based_speed_estimation_2013.pdf - Why it matters: This paper is especially relevant for motion-rich scenarios because it uses Wi-Fi CSI and speed estimation, making movement itself part of the localization signal.
- Best use: Read this if your problem explicitly involves motion, walking speed, or trajectory-aware localization.
- Keywords: Wi-Fi CSI, speed estimation, motion-aware localization, trajectory inference.
3.4 Scalable Indoor Localization via Mobile Crowdsourcing and Gaussian Process
- Local file:
papers/crowdsourcing/scalable_indoor_localization_mobile_crowdsourcing_gaussian_process_2016.pdf - Why it matters: It focuses on how to build a scalable localization system from crowdsourced mobile data, which is useful when the data collection process itself is decentralized.
- Best use: Read this when the main challenge is scaling the radio map or fingerprint model with crowd data.
- Keywords: Gaussian process, scalable localization, crowdsourced radio maps.
3.5 A Robust Crowdsourcing-Based Indoor Localization System
- Local file:
papers/crowdsourcing/robust_crowdsourcing_based_indoor_localization_system_2017.pdf - Why it matters: This paper is particularly useful because it explicitly works with trajectory radio maps and robustness issues in crowdsourced indoor localization.
- Best use: Read this when you need a practical design for noisy, inconsistent, or partially unreliable crowd-contributed traces.
- Keywords: trajectory radio map, robust localization, crowdsourced indoor positioning.
3.6 Indoor Localization Based on Weighted Surfacing from Crowdsourced Samples
- Local file:
papers/crowdsourcing/indoor_localization_weighted_surfacing_crowdsourced_samples_2018.pdf - Why it matters: It focuses on weighting and surfacing strategies for crowdsourced samples, which is useful when crowd data quality varies significantly.
- Best use: Read this when the central issue is how to aggregate noisy or unevenly distributed crowd samples.
- Keywords: weighted surfacing, sample quality, crowdsourced fingerprints.
3.7 Towards Robust Crowdsourcing-Based Localization: A Fingerprinting Accuracy Indicator Enhanced Wireless/Magnetic/Inertial Integration Approach
- Local file:
papers/crowdsourcing/robust_crowdsourcing_localization_fingerprint_accuracy_indicator_2020.pdf - Why it matters: It combines wireless, magnetic, and inertial signals and explicitly models fingerprint accuracy, which is useful for motion-rich data collection and quality-aware localization.
- Best use: Read this if you care about multi-modal sensing and reliability scoring of crowd-contributed trajectories.
- Keywords: wireless integration, magnetic sensing, inertial sensing, fingerprint quality.
3.8 TuRF: Fast Data Collection for Fingerprint-based Indoor Localization
- Local file:
papers/crowdsourcing/turf_fast_data_collection_fingerprint_indoor_localization_2017.pdf - Why it matters: It is less about end-to-end tracking and more about speeding up the crowd data collection process needed for fingerprint-based localization.
- Best use: Read this when your bottleneck is how to gather localization fingerprints quickly in real deployments.
- Keywords: fingerprint collection, data efficiency, indoor localization.
4. Group C: Mobility Analysis and Passive Wireless Tracking
These papers are most relevant when the task is not only localization but also inferring paths, proximity, or social movement patterns.
4.1 Hotspot Ranking Based Indoor Mapping and Mobility Analysis Using Crowdsourced Wi-Fi Signal
- Local metadata page:
html/hotspot_ranking_indoor_mapping_mobility_analysis_crowdsourced_wifi_signal_2017.html - Local full text: not available in the current environment due to publisher-side bot blocking.
- Why it matters: It explicitly targets indoor mapping and mobility analysis using crowdsourced Wi-Fi signal traces, which makes it one of the strongest matches for motion-driven crowd sensing.
- Best use: Read the metadata page first, then fetch the full text manually in a browser if you need the clustering and hotspot ranking details.
- Keywords: mobility analysis, Wi-Fi RSS, hotspot ranking, indoor mapping.
4.2 ProTrack: Detecting Proximity and Trajectory from Passive Wireless Traces of Mobile Devices
- Local metadata page:
html/protrack_detecting_proximity_trajectory_passive_wireless_traces_mobile_devices_2022.html - Local full text: not available in the current environment due to publisher-side bot blocking.
- Why it matters: This paper moves beyond localization and targets pairwise proximity and trajectory relationships from passive Wi-Fi and Bluetooth traces.
- Best use:
Read this if your target problem is closer to
passive tracking,co-movement,follow behavior, orsocial proximity inference. - Keywords: passive wireless sensing, proximity detection, trajectory relationship, Bluetooth, Wi-Fi.
5. Group D: Newer Extensions and ML-Oriented Directions
These papers help if the research goal is no longer only classical fingerprinting, but also learning-based trajectory modeling and seamless indoor/outdoor localization.
5.1 CrowdLOC-S: Crowdsourced Seamless Localization Framework Based on CNN-LSTM-MLP Enhanced Quality Indicator
- Local metadata page:
html/crowdloc_s_crowdsourced_seamless_localization_framework_2024.html - Local full text: not available in the current environment due to publisher-side blocking.
- Why it matters: It extends the crowdsourcing story into seamless indoor/outdoor localization with learned quality indicators and multi-source fusion.
- Best use: Read this when the setting spans multiple scenes, such as indoor to outdoor transitions or mixed GNSS and Wi-Fi operation.
- Keywords: seamless localization, CNN-LSTM-MLP, quality indicator, multi-source fusion.
5.2 Semi-Self Representation Learning for Crowdsourced WiFi Trajectories
- Local file:
papers/tracking/semi_self_representation_learning_for_crowdsourced_wifi_trajectories_2025.pdf - Why it matters: It represents the newer learning-based line of work on crowdsourced Wi-Fi trajectories and is useful if your interest is trajectory representation learning rather than only handcrafted fingerprinting.
- Best use: Read this after the older fingerprinting papers to see how the field is shifting toward learned representations.
- Keywords: representation learning, Wi-Fi trajectories, semi-supervised localization.
6. Best Matches by Research Goal
If the goal is "crowdsourced finding of a device or object"
SecureFindWho Can Find My Devices?
If the goal is "localization from users moving through space"
ZeeSmartphones Based Crowdsourcing for Indoor LocalizationCommunicating Is CrowdsourcingA Robust Crowdsourcing-Based Indoor Localization System
If the goal is "mobility analysis or trajectory inference from wireless traces"
Hotspot Ranking Based Indoor Mapping and Mobility Analysis Using Crowdsourced Wi-Fi SignalProTrackSemi-Self Representation Learning for Crowdsourced WiFi Trajectories
If the goal is "robust or scalable crowd-built localization databases"
Scalable Indoor Localization via Mobile Crowdsourcing and Gaussian ProcessIndoor Localization Based on Weighted Surfacing from Crowdsourced SamplesTowards Robust Crowdsourcing-Based LocalizationTuRF
If the goal is "multi-modal or seamless localization"
Towards Robust Crowdsourcing-Based LocalizationCrowdLOC-S
7. Practical Recommendation
If the research question is specifically:
How can a wireless system crowdsource the tracking of moving users or devices?
then the best compact reading set is:
SecureFindWho Can Find My Devices?ZeeCommunicating Is CrowdsourcingA Robust Crowdsourcing-Based Indoor Localization SystemProTrack
This six-paper set covers:
- crowd-assisted object discovery,
- large-scale real-world BLE tracking,
- zero-effort mobile crowdsourcing,
- trajectory-aware Wi-Fi localization,
- robust localization with crowdsourced traces,
- passive proximity and movement inference.