An unsatisfactory property of particle filters is that they may become inefficient when the observation noise is low. In this paper we consider a simple-to-implement particle filter, called ‘LIS-based particle filter’, whose aim is to overcome the above mentioned weakness. LIS-based particle filters sample the particles in a two-stage process that uses information of the most recent observation, too. Experiments with the standard bearings-only tracking problem indicate that the proposed new particle filter method is indeed a viable alternative to other methods.
標簽： unsatisfactory inefficient property particle
上傳時間： 2014-01-11
上傳用戶：大三三
Sequential Monte Carlo without Likelihoods 粒子濾波不用似然函數的情況下 本文摘要:Recent new methods in Bayesian simulation have provided ways of evaluating posterior distributions in the presence of analytically or computationally intractable likelihood functions. Despite representing a substantial methodological advance, existing methods based on rejection sampling or Markov chain Monte Carlo can be highly inefficient, and accordingly require far more iterations than may be practical to implement. Here we propose a sequential Monte Carlo sampler that convincingly overcomes these inefficiencies. We demonstrate its implementation through an epidemiological study of the transmission rate of tuberculosis.
標簽： Likelihoods Sequential Bayesian without
上傳時間： 2016-05-26
上傳用戶：離殤
The radio spectrum is one of the most precious resources which must be managed to ensure efficient access for the wireless communication services which use it. The allocation and management of spectrum are administered by the regulatory authorities. Traditionally, spectrum allocation is carried out exclusively of its use in large geographic areas and assigning frequency bands to specific users or service providers is proved to be inefficient. Recently, substantial knowledge about dynamic spectrum access scheme has been accumulated to enable efficient spectrum sharing.
標簽： Management Spectrum Access and
上傳時間： 2020-06-01
上傳用戶：shancjb
Although state of the art in many typical machine learning tasks, deep learning algorithmsareverycostly interms ofenergyconsumption,duetotheirlargeamount of required computations and huge model sizes. Because of this, deep learning applications on battery-constrained wearables have only been possible through wireless connections with a resourceful cloud. This setup has several drawbacks. First, there are privacy concerns. Cloud computing requires users to share their raw data—images, video, locations, speech—with a remote system. Most users are not willing to do this. Second, the cloud-setup requires users to be connected all the time, which is unfeasible given current cellular coverage. Furthermore, real-time applications require low latency connections, which cannot be guaranteed using the current communication infrastructure. Finally, wireless connections are very inefficient—requiringtoo much energyper transferredbit for real-time data transfer on energy-constrained platforms.
標簽： Embedded_Deep_Learning Algorithms
上傳時間： 2020-06-10
上傳用戶：shancjb
