The advancement of the Internet of Things(IoT)brings new opportunities for collecting real-time data and deploying machine learning models.Nonetheless,an individual IoT device may not have adequate computing resources...The advancement of the Internet of Things(IoT)brings new opportunities for collecting real-time data and deploying machine learning models.Nonetheless,an individual IoT device may not have adequate computing resources to train and deploy an entire learning model.At the same time,transmitting continuous real-time data to a central server with high computing resource incurs enormous communication costs and raises issues in data security and privacy.Federated learning,a distributed machine learning framework,is a promising solution to train machine learning models with resource-limited devices and edge servers.Yet,the majority of existing works assume an impractically synchronous parameter update manner with homogeneous IoT nodes under stable communication connections.In this paper,we develop an asynchronous federated learning scheme to improve training efficiency for heterogeneous IoT devices under unstable communication network.Particularly,we formulate an asynchronous federated learning model and develop a lightweight node selection algorithm to carry out learning tasks effectively.The proposed algorithm iteratively selects heterogeneous IoT nodes to participate in the global learning aggregation while considering their local computing resource and communication condition.Extensive experimental results demonstrate that our proposed asynchronous federated learning scheme outperforms the state-of-the-art schemes in various settings on independent and identically distributed(i.i.d.)and non-i.i.d.data distribution.展开更多
Self-encoded spread spectrum (SESS) is a unique realization of random spread spectrum. SESS eliminates the need for the traditional transmitting and receiving PN code generators. Instead, the time varying, random spre...Self-encoded spread spectrum (SESS) is a unique realization of random spread spectrum. SESS eliminates the need for the traditional transmitting and receiving PN code generators. Instead, the time varying, random spreading sequence is obtained from the data source. Cooperative diversity (CD) has been attracting increas-ing attention as a novel and promising diversity technique. This paper analyzes the cooperative SESS for Amplify and Forward CD links in Rayleigh channels. The results show that our cooperative SESS improves the system performance significantly over MRC-based cooperative systems.展开更多
Background:Debate on treatment for young patients with coronary artery disease still exists.This study aimed to investigate the intermediate-and long-term outcomes between coronary artery bypass grafting (CABG) and...Background:Debate on treatment for young patients with coronary artery disease still exists.This study aimed to investigate the intermediate-and long-term outcomes between coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) in patients aged 18-45 years with diabetes mellitus (DM).Methods:Between January 2006 and March 2016,a total of 2018 DM patients aged 18-45 years including 517 cases of CABG and 1501 cases of PCI were enrolled in the study.Using propensity score matching (PSM),406 patients were matched from each group.The intermediate-and long-term data were collected.The primary end point of this study was long-term death.The secondary end points included long-term major adverse cardiovascular and cerebrovascular events (MACCEs),stroke,angina,myocardial infarction (MI),and repeat revascularization.Results:Before PSM,the in-hospital mortality was 1.2% in the CABG group and 0.1% in the PCI group,with statistically significant difference (P 〈 0.0001).The 10-year follow-up outcomes including long-term survival rate and freedom from MACCEs were better in the CABG group than those in the PCI group (97.3% vs.94.5%,P =0.0072;93.2% vs.86.3%,P 〈 0.0001),but CABG group was associated with lower freedom from stoke compared to PCI group (94.2% vs.97.5%,P =0.0059).After propensity score-matched analysis,these findings at 10-year follow-up were also confirmed.Freedom from MACCEs was higher in CABG group compared to PCI group,but no significant difference was observed (93.1% vs.89.2%,P =0.0720).The freedom from recurrent MI was significantly higher in CABG patients compared with PCI patients (95.6% vs.92.5%,P =0.0260).Furthermore,CABG was associated with a higher rate of long-term survival rate than PCI (97.5% vs.94.6%,P =0.0403).There was no significant difference in the freedom from stroke between CABG and PCI groups (95.3% vs.97.3%,P =0.9385).The hospital cost was greater for CABG (13,936 ± 4480 US dollars vs.10,926 ± 7376 US dollars,P 〈 0.0001).Conclusions:In DM patients aged 18-45 years,the cumulative survival rate,and freedom from MI and repeat revascularization for CABG were superior to those of PCI.However,a better trend to avoid stroke was observed with PCI.展开更多
文摘The advancement of the Internet of Things(IoT)brings new opportunities for collecting real-time data and deploying machine learning models.Nonetheless,an individual IoT device may not have adequate computing resources to train and deploy an entire learning model.At the same time,transmitting continuous real-time data to a central server with high computing resource incurs enormous communication costs and raises issues in data security and privacy.Federated learning,a distributed machine learning framework,is a promising solution to train machine learning models with resource-limited devices and edge servers.Yet,the majority of existing works assume an impractically synchronous parameter update manner with homogeneous IoT nodes under stable communication connections.In this paper,we develop an asynchronous federated learning scheme to improve training efficiency for heterogeneous IoT devices under unstable communication network.Particularly,we formulate an asynchronous federated learning model and develop a lightweight node selection algorithm to carry out learning tasks effectively.The proposed algorithm iteratively selects heterogeneous IoT nodes to participate in the global learning aggregation while considering their local computing resource and communication condition.Extensive experimental results demonstrate that our proposed asynchronous federated learning scheme outperforms the state-of-the-art schemes in various settings on independent and identically distributed(i.i.d.)and non-i.i.d.data distribution.
文摘Self-encoded spread spectrum (SESS) is a unique realization of random spread spectrum. SESS eliminates the need for the traditional transmitting and receiving PN code generators. Instead, the time varying, random spreading sequence is obtained from the data source. Cooperative diversity (CD) has been attracting increas-ing attention as a novel and promising diversity technique. This paper analyzes the cooperative SESS for Amplify and Forward CD links in Rayleigh channels. The results show that our cooperative SESS improves the system performance significantly over MRC-based cooperative systems.
文摘Background:Debate on treatment for young patients with coronary artery disease still exists.This study aimed to investigate the intermediate-and long-term outcomes between coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) in patients aged 18-45 years with diabetes mellitus (DM).Methods:Between January 2006 and March 2016,a total of 2018 DM patients aged 18-45 years including 517 cases of CABG and 1501 cases of PCI were enrolled in the study.Using propensity score matching (PSM),406 patients were matched from each group.The intermediate-and long-term data were collected.The primary end point of this study was long-term death.The secondary end points included long-term major adverse cardiovascular and cerebrovascular events (MACCEs),stroke,angina,myocardial infarction (MI),and repeat revascularization.Results:Before PSM,the in-hospital mortality was 1.2% in the CABG group and 0.1% in the PCI group,with statistically significant difference (P 〈 0.0001).The 10-year follow-up outcomes including long-term survival rate and freedom from MACCEs were better in the CABG group than those in the PCI group (97.3% vs.94.5%,P =0.0072;93.2% vs.86.3%,P 〈 0.0001),but CABG group was associated with lower freedom from stoke compared to PCI group (94.2% vs.97.5%,P =0.0059).After propensity score-matched analysis,these findings at 10-year follow-up were also confirmed.Freedom from MACCEs was higher in CABG group compared to PCI group,but no significant difference was observed (93.1% vs.89.2%,P =0.0720).The freedom from recurrent MI was significantly higher in CABG patients compared with PCI patients (95.6% vs.92.5%,P =0.0260).Furthermore,CABG was associated with a higher rate of long-term survival rate than PCI (97.5% vs.94.6%,P =0.0403).There was no significant difference in the freedom from stroke between CABG and PCI groups (95.3% vs.97.3%,P =0.9385).The hospital cost was greater for CABG (13,936 ± 4480 US dollars vs.10,926 ± 7376 US dollars,P 〈 0.0001).Conclusions:In DM patients aged 18-45 years,the cumulative survival rate,and freedom from MI and repeat revascularization for CABG were superior to those of PCI.However,a better trend to avoid stroke was observed with PCI.