The need for addressing geoprivacy in location based services has increased the offer of mechanisms that protect location information, however, these algorithms are not always developed to ensure the usability of the data and therefore, their adoption is not wide. In this work, a framework is presented to evaluate the effects of geoprivacy mechanisms on the quality of geodata to provide insights into how the data is affected for geospatial analysis. For this purpose, a toolkit of indices was developed to evaluate different characteristics of the data before and after a geoprivacy mechanism is implemented, providing a criterion to select one of them. The indices measure the changes in the presence of clusters through the quantification of hotspots in hotspot analysis and the difference observed in heatmaps of the concentration of the geodata. Variations in global indices like the Nearest Neighbor Index (NNI) and the orientation of the standard deviational ellipse are also measured. For demonstration, the data of crime arrests in New York was used for the month of January in 2017 and 2018. Five mechanisms were tested with different settings, resulting with the NRand-K algorithm producing fewer alterations to the reference data, preserving its initial characteristics better than the other mechanisms.

In recent years Location-Based Information Systems have increased its popularity in the market of mobile applications, however, due to the ability of smartphones and similar devices to estimate location in real time using GPS or through network providers; it is critical to implement techniques to protect such sensitive information while still make it available to the service provider. This paper presents Near-Rand, a new random noise-based location obfuscation technique. This algorithm generates random points around the real location of the user within a neighbor-size squared area and calculates the n nearest points average to the user's location giving as result an obfuscated point. Compared to Pinwheel, another noise-based obfuscation mechanism, it shows a similar performance against Exponentially Moving Average-based filtering attacks, while being able to use non-uniformly distributions for random points.

In the last few years, the number of applications that use the location of the users in order to provide a more personalized service have been increasing, mainly due to easy access to low cost smartphones, geographical positioning systems and other factors, like social networking. The concern of protecting this information is also increasing due to the capabilities than an eventual attacker could have if the location information is obtained. Some protection techniques have been proposed in the literature; for example, location obfuscation which slightly alters the location to hide the real one. However, this technique could be filtered out with time series-based mechanisms. In this work, the Pinwheel obfuscation technique is proposed in order to reduce the possibility of EMA-based filtering based on high variability and asymmetry of the induced noise. The results show that the level of filtered noise is reduced from 35% in N-RAND and 30% in θ-Rand obfuscation techniques, to 15% in Pinwheel, with asymmetric scenarios, while preserving a long final average distance from the original path after a filtering attack.

Location-Based Information Systems have been increasingly adopted by users of both modern Web applications and mobile phones, this last group being the more susceptible for geographic tracking due to their devices' ability to calculate their location in real time via GPS or other positioning technique. This information needs to be protected, while still being available to the service provider so services can be offered. This paper presents a new random noise-based location obfuscation technique to alter the real location of the user, based on the N-RAND technique, while being more resistant to noise filtering than its predecessor. Results show that the θ-RAND technique can reduce by 20% the amount of noise filtered by an exponential moving average filtering technique.