The localization of a speaker inside a closed environment is often approached by real-time processing of multiple audio signals captured by a set of microphones. One of the leading related methods for sound source localization, the steered-response power (SRP), searches for the point of maximum power over a spatial grid. High-accuracy localization calls for a dense grid and/or many microphones, which tends to impractically increase computational requirements. This paper proposes a new method for sound source localization (called H-SRP), which applies the SRP approach to space regions instead of grid points. This arrangement makes room for the use of a hierarchical search inspired by the branch-and-bound paradigm, which is guaranteed to find the global maximum in anechoic environments and shown experimentally to also work under reverberant conditions. Besides benefiting from the improved robustness of volume-wise search over point-wise search as to reverberation effects, the H-SRP attains high performance with manageable complexity. In particular, an experiment using a 16-microphone array in a typical presentation room yielded localization errors of the order of 7 cm, and for a given fixed complexity, competing methods' errors are two to three times larger.Index Terms-Sound source localization, steered-response power, microphone array, computational complexity, hierarchical search, branch-and-bound.
We provide new evidence on the presence and distribution of racial bias in the criminal justice system. In many states, the punishment for speeding increases discontinuously with the speed of the driver, exhibiting large jumps in fine amounts. It is a common practice for officers to reduce the charged speed to just below this jump, avoiding an onerous punishment for the driver. Using data from the Florida Highway Patrol, we find evidence of significant bunching in ticketed speeds below a jump in punishment for all drivers but significantly more for whites than for blacks and Hispanics. We estimate the bias of each officer by comparing his lenience towards whites and non-whites, allowing us to recover the full distribution of bias. The total disparity in lenience across races can be explained by a small percentage (∼20%) of officers. Officers tend to favor drivers of their own racial group, and younger, female, and collegeeducated officers are less biased. We then estimate a model that allows for both heterogeneity in officer preferences and driver speeds across races. Because minorities tend to live in areas where officers are harsher to all drivers, policies targeting bias have little effect on the aggregate speed gap. We find that racial bias in lenience explains 16% of the minority-white speed gap, and spatial differences in race-blind lenience explain 30% of the gap.
-The dynamics of flash drums is simulated using a formulation adequate for phase modeling with equations of state (EOS). The energy and mass balances are written as differential equations for the internal energy and the number of moles of each species. The algebraic equations of the model, solved at each time step, are those of a flash with specified internal energy, volume and mole numbers (UVN flash). A new aspect of our dynamic simulations is the use of direct iterations in phase volumes (instead of pressure) for solving the algebraic equations. It was also found that an iterative procedure previously suggested in the literature for UVN flashes becomes unreliable close to phase boundaries and a new alternative is proposed. Another unusual aspect of this work is that the model expressions, including the physical properties and their analytical derivatives, were quickly implemented using computer algebra.
We estimate the degree to which individual police officers practice racial discrimination. Using a bunching estimation design and data from the Florida Highway Patrol, we show that minorities are less likely to receive a discount on their speeding tickets than White drivers. Disaggregating this difference to the individual police officer, we estimate that 42 percent of officers practice discrimination. We then apply our officer- level discrimination measures to various policy-relevant questions in the literature. In particular, reassigning officers across locations based on their lenience can effectively reduce the aggregate disparity in treatment (JEL H76, J15, K42)
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.