Apple Inc.

Venturi Wireless, Inc.

Venturi IP LLC

David S Choi

Kanapathipillai Ketheesan

Amit K Vyas

Abhinav Pathak

Anand Ramadurai

Krishna Ramadas

Franco Travostino

Michael S Ferris

Albert S Liu

Olivier Mardinian

Kevin C Milden

Henri S Berger

Drew A Schmitt

Sathish K Narayanaswamy

Daniel Lertpratchya

Sanket V Bedare

Vividh Siddha

Chico Jayanthan

Hari Gorak

Sean Shi

Christopher R D'Angelo

Eisuke Arai

Chico Jayanathan

Embodiments of the present disclosure include methods and systems for generating and processing energy consumption reports received from a plurality of devices. The reports may include a plurality of call stacks captured and stored by remote device as a result of an energy consumption of the corresponding device. For each energy consumption report, the plurality of call stacks may be analyzed to determine a representative call stack of the energy consumption report. The energy consumption reports may be assigned to one or more of a plurality of energy consumption categories based on distances between the representative call stacks of the energy consumption reports and corresponding representative call stacks of the energy consumption categories. A request may be received for energy consumption reports for a particular energy consumption category. Upon receiving the request, providing a representative set of energy consumption reports assigned to the particular energy consumption category.

Detection and analysis of energy consumption of software applications

Some embodiments of the invention provide a novel architecture for debugging devices. This architecture includes numerous devices that without user intervention automatically detect and report bug events to a set of servers that aggregate and process the bug events. When a device detects a potential bug event, the device in some embodiments generates a description of the potential bug event, and sends the generated description to the server set through a network. In addition to generating such a description, the device in some embodiments directs one or more of its modules to gather and store a collection of one or more data sets that are relevant to the potential bug event, in case the event has to be further analyzed by the server set. In the discussion below, the generated bug-event description is referred to as the event signature, while the gathered collection of data sets for an event is referred to as the event's data archive. The server set aggregates and processes the bug-event signatures that it receives from the various devices. For only a subset of the reported bug-event signatures, the server set then directs the devices that sent these signatures to also send the data archives that these devices have gathered and stored for the events associated with these signatures. These data archives can be further analyzed to identify the root causes of the bug events.

Auto bug capture

A bandwidth estimation algorithm on shared links detects peaks and/or average per-user bandwidth. Estimating is performed at the transport or IP layer with no assistance from lower layer (PHY, MAC, etc.) and the techniques can be used for any of adjusting the level of video optimization to the available bandwidth; driving QoS decisions at the transmitter based on available bandwidth; improving QoS enforcement during transitions among hybrid technologies on a wireless links; and correcting estimates on devices delivering bursty payload.

Channel bandwidth estimation on hybrid technology wireless links

A system for optimizing communications on a radio network by altering transitions between different link states that includes several modules. The activity, environment, and load module monitor monitors the link layer based on spectral-load metrics and radio-link metrics. The state transition control module determines when user equipment transitions between different states based on the type of user equipment, user equipment battery life, whether the user equipment is connected to an alternating current outlet, a spectral cost, and a backhaul cost. The channel state influencer module uses any of direct messages, ping messages, and keep-alive messages to influence the link state. The policy and preference handler enables or disables transitions based on the bearer technology type, the type of user equipment, the user's subscription plan, and the load level on the network.

Systems, structures and associated processes for optimization of state transitions within wireless networks

Channel state transition optimization

A bandwidth estimation algorithm detects peaks and/or average per-user bandwidth of data communication networks, such as narrowband and broadband wide-area radio access networks. Estimation can be performed at the TCP/IP layer with no lower layer (PHY, MAC, etc.) information assumed to be available. However, the bandwidth estimation algorithm can be applied to anywhere bandwidth needs to be estimated as well.

Method and apparatus for estimating channel bandwidth

Growing community of inventors

Cupertino, CA, United States of America

David S Choi