The COVID-19 pandemic has necessitated rapid top-down dissemination of reliable and actionable information. This presents unique challenges in engaging low-literate communities that live in poverty and lack access to the Internet. We describe the design and deployment of a voice-based social media platform, accessible over simple phones, for actively engaging such communities in Pakistan with reliable COVID information. We developed three strategies to overcome users’ hesitation, mistrust, and skepticism in engaging with COVID content. Users were: (1) encouraged to listen to reliable COVID advisory, (2) incentivized to share authentic content with others, and (3) prompted to critically think about COVID-related information behaviors. Using a mixed-methods evaluation, we show that users approached with all three strategies had a significantly higher engagement with COVID content compared to others. We discuss how new designs of social media can enable users to engage with and propagate authentic information.

What determines the success of community cellular networks? We leverage unique circumstances where all households in seven localities were interviewed before the launch of cellular networks. We observed substantial differences in network adoption across communities. Four communities displayed high and regular usage, while usage dissipated shortly after the network launch in three sites. Sixty-five percent of households made or received at least one call or text message. We find that a one standard deviation increase in household wealth is correlated with a three percentage point increase in network adoption and 43 additional cellular network transactions. Agricultural households were ten percentage points more likely to adopt the network than other households and female-headed households were five percentage points more likely to use the network at least once.

Significant sampling is an adaptive monitoring technique proposed for highly dynamic networks with centralized network management and control systems. The essential spirit of significant sampling is to collect and disseminate network state information when it is of significant value to the optimal operation of the network, and in particular when it helps identify the shortest routes. Discovering the optimal sampling policy that specifies the optimal sampling frequency is referred to as the significant sampling problem. Modeling the problem as a Markov Decision process, this paper puts forth a deep reinforcement learning (DRL) approach to tackle the significant sampling problem. This approach is more flexible and general than prior approaches as it can accommodate a diverse set of network environments. Experimental results show that, 1) by following the objectives set in the prior work, our DRL approach can achieve performance comparable to their analytically derived policy φ' - unlike the prior approach, our approach is model-free and unaware of the underlying traffic model; 2) by appropriately modifying the objective functions, we obtain a new policy which addresses the never-sample problem of policy φ', consequently reducing the overall cost; 3) our DRL approach works well under different stochastic variations of the network environment - it can provide good solutions under complex network environments where analytically tractable solutions are not feasible.

We face a growing ecosystem of applications that produce and consume data at unprecedented rates and with strict latency requirements. Meanwhile, the bursty and unpredictable nature of their traffic can induce highly dynamic environments within networks which endanger their own viability. Unencumbered operation of these applications requires rapid (re)actions by Network Management and Control (NMC) systems which themselves depends on timely collection of network state information. Given the size of today's networks, collection of detailed network states is prohibitively costly for the network transport and computational resources. Thus, judicious sampling of network states is necessary for a cost-effective NMC system. This paper proposes a deep reinforcement learning (DRL) solution that learns the principle of significant sampling and effectively balances the need for accurate state information against the cost of sampling. Modeling the problem as a Markov Decision Process, we treat the NMC system as an agent that samples the state of various network elements to make optimal routing decisions. The agent will periodically receive a reward commensurate with the quality of its routing decisions. The decision on when to sample will progressively improve as the agent learns the relationship between the sampling frequency and the reward function. We show that our solution has a comparable performance to the recently published analytical optimal without the need for an explicit knowledge of the traffic model. Furthermore, we show that our solution can adapt to new environments, a feature that has been largely absent in the analytical considerations of the problem.

In this paper we propose a novel approach to deliver better delay-jitter performance in dynamic networks. Dynamic networks experience rapid and unpredictable fluctuations and hence, a certain amount of uncertainty about the delay-performance of various network elements is unavoidable. This uncertainty makes it difficult for network operators to guarantee a certain quality of service (in terms of delay and jitter) to users. The uncertainty about the state of the network is often overlooked to simplify problem formulation, but we capture it by modeling the delay on various links as general and potentially correlated random processes. Within this framework, a user will request a certain delay-jitter performance guarantee from the network. After verifying the feasibility of the request, the network will respond to the user by specifying a set of routes as well as the proportion of traffic which should be sent through each one to achieve the desired quality of service. We propose to use mean-variance analysis as the basis for traffic distribution and route selection, and show that this technique can significantly reduce the end-to-end jitter because it accounts for the correlated nature of delay across different paths. The resulting traffic distribution is often non-uniform and the fractional flow on each path is the solution to a simple convex optimization problem. We conclude the paper by commenting on the potential application of this method to general transportation networks.

Future networks have to accommodate an increase of 3-4 orders of magnitude in data rates with very heterogeneous session sizes and sometimes with strict time deadline requirements. The dynamic nature of scheduling of large transactions and the need for rapid actions by the Network Management and Control (NMC) system, require timely collection of network state information. Rough estimates of the size of detailed network states suggest large volumes of data with refresh rates commensurate with the coherence time of the states (can be as fast as 100 ms), resulting in huge burden and cost for the network transport (~300 Gbps/link) and computation resources. Thus, judicious sampling of network states is necessary for a cost-effective network management system. In this paper, we consider a construct of an NMC system where sensing and routing decisions are made with cognitive understanding of the network states and short-term behavior of exogenous offered traffic. We have studied a small but realistic example of adaptive monitoring based on significant sampling techniques. This technique balances the need for accurate and updated state information against the updating cost and provides an algorithm that yields near optimum performance with significantly reduced burden of sampling, transport and computation. We show that our adaptive monitoring system can reduce the NMC overhead by a factor of 100 in one example. The essential spirit of the cognitive NMC is that it collects network states ONLY when they matter to the network performance.

We introduce a novel feedback protocol, called SMART, for wireless broadcast networks that use linear network coding. We consider transmission of packets from a single source to many receivers over a single-hop broadcast erasure channel with heterogeneous links. We propose a predictive model to minimize feedback as well as extraneous data transmissions by the source. In addition, we use the method of types to provide a lower bound for the expected total transmission time, and use simulations to show that our protocol operates close to this lower bound. We show that with SMART, counter to conventional wisdom, the average user's QoE improves slightly as the number of users increases. We demonstrate that SMART's algorithmic simplicity enables multicast transmissions that on average take fewer than 2 feedback rounds to complete. We show the favorable scalability of our technique with the number of users, which enables reliable quality of experience. We also show the robustness of this scheme to uncertainty in the number of receiving nodes, and packet erasure probability, as well as to partial loss of the feedback. Furthermore, we show that SMART performs nearly as well as an omniscient transmitter that requires no feedback.

We present a novel feedback protocol for wireless broadcast networks that utilize linear network coding. We consider transmission of packets from one source to many receivers over a single-hop broadcast erasure channel. Our method utilizes a predictive model to request feedback only when the probability that all receivers have completed decoding is significant. In addition, our proposed NACK-based feedback mechanism enables all receivers to request, within a single time slot, the number of retransmissions needed for successful decoding. We present simulation results as well as analytical results that show the favorable scalability of our technique as the number of receivers, file size, and packet erasure probability increase. We also show the robustness of this scheme to uncertainty in the predictive model, including uncertainty in the number of receiving nodes and the packet erasure probability, as well as to losses of the feedback itself. Our scheme, SMART, is shown to perform nearly as well as an omniscient transmitter that requires no feedback. Furthermore, SMART, is shown to outperform current state of the art methods at any given erasure probability, file size, and numbers of receivers.