**VTrack: Accurate, Energy-aware Road Traffic Delay Estimation Using Mobile Phones** **Authors:** Arvind Thiagarajan et al. **Abstract:** Traffic congestion is a significant issue, causing inefficiencies, wasted fuel, and commuter frustration. This paper presents VTrack, a system for travel time estimation using sensor data from smartphones, addressing the challenges of energy consumption and sensor unreliability. VTrack uses a hidden Markov model (HMM) for map matching and travel time estimation, which can handle noisy and sparse data. The system evaluates the quality of travel time estimates from different sensor combinations and sampling rates, demonstrating that WiFi localization alone can provide accurate enough estimates for route planning, despite individual segment errors. VTrack also shows robustness to noise and outages in location estimates, making it suitable for real-time applications like route planning and hotspot detection. **Key Contributions:** 1. **Robust HMM-based Map Matching:** VTrack uses an HMM to match noisy or sparse position samples to road segments, achieving median errors less than 10% even with only WiFi data. 2. **Accurate Travel Time Estimates:** WiFi localization alone can provide accurate travel times for route planning, with the 90th percentile optimality gap being 10-15%. 3. **Energy Efficiency:** VTrack can save energy by using less frequent GPS sampling, making it suitable for battery-powered devices. 4. **Hotspot Detection:** VTrack can accurately detect hotspots ( roads with travel times significantly higher than expected) using WiFi data, despite the limitations of noisy and sparsely sampled data. **Evaluation:** The evaluation uses a large dataset of GPS and WiFi location estimates from real drives, demonstrating that VTrack can achieve accurate travel time estimates and route planning using WiFi localization or sparsely sampled GPS, while saving energy. The system is also robust to significant amounts of simulated Gaussian noise, further validating its effectiveness. **Conclusion:** VTrack is a real-time traffic monitoring system that overcomes the challenges of energy consumption and sensor unreliability, providing accurate travel time estimates and route planning using smartphone sensor data.**VTrack: Accurate, Energy-aware Road Traffic Delay Estimation Using Mobile Phones** **Authors:** Arvind Thiagarajan et al. **Abstract:** Traffic congestion is a significant issue, causing inefficiencies, wasted fuel, and commuter frustration. This paper presents VTrack, a system for travel time estimation using sensor data from smartphones, addressing the challenges of energy consumption and sensor unreliability. VTrack uses a hidden Markov model (HMM) for map matching and travel time estimation, which can handle noisy and sparse data. The system evaluates the quality of travel time estimates from different sensor combinations and sampling rates, demonstrating that WiFi localization alone can provide accurate enough estimates for route planning, despite individual segment errors. VTrack also shows robustness to noise and outages in location estimates, making it suitable for real-time applications like route planning and hotspot detection. **Key Contributions:** 1. **Robust HMM-based Map Matching:** VTrack uses an HMM to match noisy or sparse position samples to road segments, achieving median errors less than 10% even with only WiFi data. 2. **Accurate Travel Time Estimates:** WiFi localization alone can provide accurate travel times for route planning, with the 90th percentile optimality gap being 10-15%. 3. **Energy Efficiency:** VTrack can save energy by using less frequent GPS sampling, making it suitable for battery-powered devices. 4. **Hotspot Detection:** VTrack can accurately detect hotspots ( roads with travel times significantly higher than expected) using WiFi data, despite the limitations of noisy and sparsely sampled data. **Evaluation:** The evaluation uses a large dataset of GPS and WiFi location estimates from real drives, demonstrating that VTrack can achieve accurate travel time estimates and route planning using WiFi localization or sparsely sampled GPS, while saving energy. The system is also robust to significant amounts of simulated Gaussian noise, further validating its effectiveness. **Conclusion:** VTrack is a real-time traffic monitoring system that overcomes the challenges of energy consumption and sensor unreliability, providing accurate travel time estimates and route planning using smartphone sensor data.