The smart city is an ecosystem that interconnects various devices like sensors, actuators, mobiles, and vehicles. The intelligent and connected transportation system (ICTS) is an essential part of this ecosystem that provides new real-time applications. The emerging applications are based on Internet-of-Things (IoT) technologies, which bring out new challenges, such as heterogeneity and scalability, and they require innovative communication solutions. The existing routing protocols cannot achieve these requirements due to the surrounding knowledge supported by individual nodes and their neighbors, displaying partial visibility of the network. However, the issue grew ever more arduous to conceive routing protocols to satisfy the ever-changing network requirements due to its dynamic topology and its heterogeneity. Software-Defined Networking (SDN) offers the latest view of the entire network and the control of the network based on the application’s specifications. Nonetheless, one of the main problems that arise when using SDN is minimizing the transmission delay between ubiquitous nodes. In order to meet this constraint, a well-attended and realistic alternative is to adopt the Machine Learning (ML) algorithms as prediction solutions. In this paper, we propose a new routing protocol based on SDN and Naive Bayes solution to improve the delay. Simulation results show that our routing scheme outperforms the comparative ones in terms of end-to-end delay and packet delivery ratio.

Background: Same-day HIV testing and antiretroviral therapy (ART) initiation is being widely implemented. However, the optimal timing of ART among patients with tuberculosis (TB) symptoms is unknown. We hypothesized that same-day treatment (TB treatment for those diagnosed with TB; ART for those not diagnosed with TB) would be superior to standard care in this population. Methods and findings: We conducted an open-label trial among adults with TB symptoms at initial HIV diagnosis at GHESKIO in Haiti; participants were recruited and randomized on the same day. Participants were randomized in a 1:1 ratio to same-day treatment (same-day TB testing with same-day TB treatment if TB diagnosed; same-day ART if TB not diagnosed) versus standard care (initiating TB treatment within 7 days and delaying ART to day 7 if TB not diagnosed). In both groups, ART was initiated 2 weeks after TB treatment. The primary outcome was retention in care with 48-week HIV-1 RNA <200 copies/mL, with intention to treat (ITT) analysis. From November 6, 2017 to January 16, 2020, 500 participants were randomized (250/group); the final study visit occurred on March 1, 2021. Baseline TB was diagnosed in 40 (16.0%) in the standard and 48 (19.2%) in the same-day group; all initiated TB treatment. In the standard group, 245 (98.0%) initiated ART at median of 9 days; 6 (2.4%) died, 15 (6.0%) missed the 48-week visit, and 229 (91.6%) attended the 48-week visit. Among all who were randomized, 220 (88.0%) received 48-week HIV-1 RNA testing; 168 had <200 copies/mL (among randomized: 67.2%; among tested: 76.4%). In the same-day group, 249 (99.6%) initiated ART at median of 0 days; 9 (3.6%) died, 23 (9.2%) missed the 48-week visit, and 218 (87.2%) attended the 48-week visit. Among all who were randomized, 211 (84.4%) received 48-week HIV-1 RNA; 152 had <200 copies/mL (among randomized: 60.8%; among tested: 72.0%). There was no difference between groups in the primary outcome (60.8% versus 67.2%; risk difference: −0.06; 95% CI [−0.15, 0.02]; p = 0.14). Two new grade 3 or 4 events were reported per group; none were judged to be related to the intervention. The main limitation of this study is that it was conducted at a single urban clinic, and the generalizability to other settings is uncertain. Conclusions: In patients with TB symptoms at HIV diagnosis, we found that same-day treatment was not associated with superior retention and viral suppression. In this study, a short delay in ART initiation did not appear to compromise outcomes. Trial registration: This study is registered with ClinicalTrials.gov NCT03154320.

Mobile networks are expected to face major problems such as low network capacity, high latency, and limited resources but are expected to provide seamless connectivity in the foreseeable future. It is crucial to deliver an adequate level of performance for network services and to ensure an acceptable quality of services for mobile users. Intelligent mobility management is a promising solution to deal with the aforementioned issues. In this context, modeling user mobility behaviour is of great importance in order to extract valuable information about user behaviours and to meet their demands. In this paper, we propose a hybrid user mobility prediction approach for handover management in mobile networks. First, we extract user mobility patterns using a mobility model based on statistical models and deep learning algorithms. We deploy a vector autoregression (VAR) model and a gated recurrent unit (GRU) to predict the future trajectory of a user. We then reduce the number of unnecessary handover signaling messages and optimize the handover procedure using the obtained prediction results. We deploy mobility data generated from real users to conduct our experiments. The simulation results show that the proposed VAR-GRU mobility model has the lowest prediction error in comparison with existing methods. Moreover, we investigate the handover processing and transmission costs for predictive and non-predictive scenarios. It is shown that the handover-related costs effectively decrease when we obtain a prediction in the network. For vertical handover, processing cost and transmission cost improve, respectively, by 57.14% and 28.01%.

Network densification with Small Cells (SCs) has emerged as a key technique to increase the 5G network capacity. However, in a densified network, fast mobile nodes will experience frequent handovers with a high signaling load, handover latency, and packet loss, due to the short cell radius. Indeed, Distributed Mobility Management (DMM) protocols aim to solve the shortcomings of centralized mobility management solutions such as poor scalability and non-optimal routing. However, when the cell residence time is short, DMM protocols might suffer from increased costs and limited performance. This paper proposes a localized mobility management protocol in 5G dense SCs, based on the locator identifier separation protocol, local mobility anchoring, and fast handovers concepts. The proposed scheme divides a local domain into several location service areas, each controlled by a local anchor. We provide the analytical models of several handover metrics, namely the average total signaling cost, the data delivery cost, the handover latency, and the packet loss. Numerical and simulation results show significant cost savings, up to 30% in signaling overhead, up to 53% of packet loss, and up to 90% of processing load reduction at the core of the network compared to the existing lisp mobile node protocol.