Telefonaktiebolaget Lm Ericsson (publ)

Pulin Chhatbar

Miroslav Budic

Narendra Tilwani

Abdennaceaur Lachtar

SaiRamesh Nammi

Carl Cao

Mikhail Royz

Fang Gao

Ravindra Dongre

Gheorghe Spiride

Mayuresh Hooli

Nirav Gohel

A system, method and non-transitory computer readable media for optimizing input data for an ML model associated with a communications network. In one implementation, example ML model(s) may be trained using a modified dataset obtained for a plurality of cellular aggregation units (CAUs) of the RAN infrastructure(s), wherein the modified dataset is derived from data collected for individual CAUs over a data collection period with respect to a plurality of KPI variables. The modified data set is optimized by replacement of null values of variables with corresponding modal values of the variables. The trained ML model may be used for predicting one or more KPIs based on a set of test data associated with the RAN infrastructure(s).

System and method for improving machine learning model performance in a communications network

The present disclosure relates to an advanced digital baseband processor for a base station in a cellular communications network. In general, the advanced digital baseband processor integrates multiple Radio Access Technologies (RATs). As used herein, a RAT is a radio access technology as defined by a corresponding standard. For example, in one embodiment, RATs integrated by the advanced digital baseband processor include a Code Division Multiple Access (CDMA) voice RAT (e.g., 3GPP2 CDMA 1xRTT or 3GPP2 CDMA 1xAdvanced) as well as a CDMA Data Only, or Data Optimized, (DO) RAT (e.g., 3GPP2 CDMA EV-DO or 3GPP2 CDMA EV-DO Rev B). In another embodiment, the RATs integrated by the advanced digital baseband processor also include 3GPP Long Term Evolution (LTE). The RATs given above are only examples. The advanced digital baseband processor may integrate additional or alternative RATs.

Apparatus, systems, methods, and computer products suitable for use in an advanced digital baseband processor

A method of operating a wireless terminal communicating with a base station over a wireless channel may include determining whether the wireless terminal is in an edge area or an interior area of coverage of the base station. Responsive to determining that the wireless terminal is in an edge area of coverage of the base station, a transmit diversity communications rank indicator may be selected to select diversity communications over the wireless channel between the wireless terminal and the base station. Related wireless terminals are also discussed.

Methods of determining rank information and related communications devices and systems

An Optimized Streetlight Operation (OSLO) system that utilizes a cellular network overlay to broadcast control commands issued by a centralized Streetlight Operation Center (SOC) to RF control modules mounted on each streetlight pole. The cellular network may be a CDMA network utilizing Short Message Service-Broadcast (SMS-B) messaging. When a Mobile Telephone Exchange/Mobile Switching Center (MTX/MSC) receives a SOC control command from the SOC, the MTX/MSC causes a plurality of BSs to broadcast the SOC command in the network operating area a defined number of times. Each RF control module that successfully receives the broadcast SOC control command, performs actions to control operation of the module's associated streetlight, and transmits a confirmation message to the module's serving BS indicating the SOC command was received. The modules are configured to stagger transmission of the confirmation messages over a period of time to reduce a peak load imposed on the cellular network.

Optimized streetlight operation (OSLO) using a cellular network overlay

A method and apparatus for controlling a Network Load Balancing (NLB) algorithm that balances a traffic load between multiple downlink (DL) sectors in a cellular telecommunication network. A Connection Integrity Preservation (CIP) algorithm, which runs on top of the NLB algorithm in the Radio Network Controller/Base Station Controller (RNC/BSC), minimizes the risk of degrading network performance due to NLB offload decisions. The CIP algorithm may override an NLB offload decision, for example, if there have been too many offload failures, there are no target DL sectors available to acquire an offloaded Access Terminal (AT), or the offloaded AT is not acquired within a threshold time period. The CIP algorithm ensures required metrics are collected, and minimizes the impact on RNC/BSC processing due to Routing Update messages needed to make offload decisions. The invention enables the NLB algorithm to realize its potential without negative side-effects.

Growing community of inventors

Plano, TX, United States of America

Pulin Chhatbar