Search

We consider a status update system consisting of a finite-state Markov source, an energy-harvesting-enabled transmitter, and a sink. The forward and feedback channels between the transmitter and the sink are error-prone. We study the problem of minimizing the long-term time average of a (generic) distortion function subject to an energy causality constraint. Since the feedback channel is error-prone, the transmitter has only partial knowledge about the transmission results and, consequently, about the estimate of the source state at the sink. Therefore, we model the problem as a partially observable Markov decision process (POMDP), which is then cast as a belief-MDP problem. The infinite belief space makes solving the belief-MDP difficult. Thus, by exploiting a specific property of the belief evolution, we truncate the state space and formulate a finite-state MDP problem, which is then solved using the relative value iteration algorithm (RVIA). Furthermore, we propose a low-complexity transmission policy in which the belief-MDP problem is transformed into a sequence of per-slot optimization problems. Simulation results show the effectiveness of the proposed policies and their superiority compared to a baseline policy. Moreover, we numerically show that the proposed policies have switching-type structures.

We address the problem of real-time remote tracking of a partially observable Markov source in an energy harvesting system with an unreliable communication channel. We consider both sampling and transmission costs. Different from most prior studies that assume the source is fully observable, the sampling cost renders the source partially observable. The goal is to jointly optimize sampling and transmission policies for two semantic-aware metrics: i) a general distortion measure and ii) the age of incorrect information (AoII). We formulate a stochastic control problem. To solve the problem for each metric, we cast a partially observable Markov decision process (POMDP), which is transformed into a belief MDP. Then, for both AoII under the perfect channel setup and distortion, we express the belief as a function of the age of information (AoI). This expression enables us to effectively truncate the corresponding belief space and formulate a finite-state MDP problem, which is solved using the relative value iteration algorithm. For the AoII metric in the general setup, a deep reinforcement learning policy is proposed to solve the belief MDP problem. Simulation results show the effectiveness of the derived policies and, in particular, reveal a non-monotonic switching-type structure of the real-time optimal policy with respect to AoI., Comment: The preliminary results were accepted for Asilomar 2023 and IEEE Globecom 2023

We study a real-time tracking problem in an energy harvesting status update system with a Markov source and an imperfect channel, considering both sampling and transmission costs. The problem primary challenge stems from the non-observability of the source due to the sampling cost. By using the age of incorrect information (AoII) as a semantic-aware performance metric, our main goal is to find an optimal policy that minimizes the time average AoII subject to an energy-causality constraint. To this end, a stochastic optimization problem is formulated and solved by modeling it as a partially observable Markov decision process (POMDP). More specifically, to solve the main problem, we use the notion of a belief state and cast the problem as a belief MDP problem. Then, for the perfect channel setup, we effectively truncate the corresponding belief space and solve the MDP problem using the relative value iteration method. For the general setup, a deep reinforcement learning policy is proposed. The simulation results show the efficacy of the derived policies in comparison to an AoI-optimal policy and an opportunistic baseline policy., Comment: Accepted for presentation at Globecom 2023

We study status updating under inexact knowledge about the battery levels of the energy harvesting sensors in an IoT network, where users make on-demand requests to a cache-enabled edge node to send updates about various random processes monitored by the sensors. To serve the request(s), the edge node either commands the corresponding sensor to send an update or uses the aged data from the cache. We find a control policy that minimizes the average on-demand AoI subject to per-slot energy harvesting constraints under partial battery knowledge at the edge node. Namely, the edge node is informed about sensors' battery levels only via received status updates, leading to uncertainty about the battery levels for the decision-making. We model the problem as a POMDP which is then reformulated as an equivalent belief-MDP. The belief-MDP in its original form is difficult to solve due to the infinite belief space. However, by exploiting a specific pattern in the evolution of beliefs, we truncate the belief space and develop a dynamic programming algorithm to obtain an optimal policy. Moreover, we address a multi-sensor setup under a transmission limitation for which we develop an asymptotically optimal algorithm. Simulation results assess the performance of the proposed methods., Comment: 36 Pages. arXiv admin note: text overlap with arXiv:2203.10400, arXiv:2212.05979

We consider an energy harvesting (EH) IoT network, where users make on-demand requests to a cache-enabled edge node to send status updates about various random processes, each monitored by an EH sensor. The edge node serves users' requests by either commanding the corresponding sensor to send a fresh status update or retrieving the most recently received measurement from the cache. We aim to find a control policy at the edge node that minimizes the average on-demand AoI over all sensors subject to per-slot transmission and energy constraints under partial battery knowledge at the edge node. Namely, the limited radio resources (e.g., bandwidth) causes that only a limited number of sensors can send status updates at each time slot (i.e., per-slot transmission constraint) and the scarcity of energy for the EH sensors imposes an energy constraint. Besides, the edge node is informed of the sensors' battery levels only via received status update packets, leading to uncertainty about the battery levels for the decision-making.We develop a low-complexity algorithm -- termed relax-then-truncate -- and prove that it is asymptotically optimal as the number of sensors goes to infinity. Numerical results illustrate that the proposed method achieves significant gains over a request-aware greedy policy and show that it has near-optimal performance even for moderate numbers of sensors., Comment: 7 pages. arXiv admin note: substantial text overlap with arXiv:2201.12277

This letter provides query-age-optimal joint sampling and transmission scheduling policies for a heterogeneous status update system, consisting of a stochastic arrival and a generate-at-will source, with an unreliable channel. Our main goal is to minimize the average query age of information (QAoI) subject to average sampling, average transmission, and per-slot transmission constraints. To this end, an optimization problem is formulated and solved by casting it into a linear program. We also provide a low-complexity near-optimal policy using the notion of weakly-coupled constrained Markov decision processes. The numerical results show up to 32% performance improvement by the proposed policies compared with a benchmark policy., Comment: Accepted by IEEE COML

We study status updating under two-way delay in a system consisting of a sampler, a sink, and a controller residing at the sink. The controller controls the sampling process by sending request packets to the sampler. Upon receiving a request, the sampler generates a sample and transmits the status update packet to the sink. Transmissions of both request and status update packets encounter random delays. We develop optimal control policies to minimize the average age of information (AoI) using the tools of Markov decision processes in two scenarios. We begin with the system having at most one active request, i.e., a generated request for which the sink has not yet received a status update packet. Then, as the main distinctive feature of this paper, we initiate pipelining-type status updating by studying a system having at most two active requests. Furthermore, we conduct AoI analysis by deriving the average AoI expressions for the Zero-Wait-1, Zero-Wait-2, and Wait-1 policies. According to the Zero-Wait-1 policy, whenever a status update packet is delivered to the sink, a new request packet is inserted into the system. The Zero-Wait-2 policy operates similarly, except that the system holds two active requests. According to the Wait-1 policy, whenever a status update packet is delivered to the sink, a new request is sent after a waiting time which is a function of the current AoI. Numerical results illustrate the performance of each status updating policy under different system parameter values.

We consider a multi-source hybrid automatic repeat request (HARQ) based system, where a transmitter sends status update packets of random arrival (i.e., uncontrollable sampling) and generate-atwill (i.e., controllable sampling) sources to a destination through an error-prone channel. We develop transmission scheduling policies to minimize the average number of transmissions subject to an average age of information (AoI) constraint. First, we consider known environment (i.e., known system statistics) and develop a near-optimal deterministic transmission policy and a low-complexity dynamic transmission (LC-DT) policy. The former policy is derived by casting the main problem into a constrained Markov decision process (CMDP) problem, which is then solved using the Lagrangian relaxation, relative value iteration algorithm, and bisection. The LC-DT policy is developed via the drift-plus-penalty (DPP) method by transforming the main problem into a sequence of per-slot problems. Finally, we consider unknown environment and devise a learning-based transmission policy by relaxing the CMDP problem into an MDP problem using the DPP method and then adopting the deep Q-learning algorithm. Numerical results show that the proposed policies achieve near-optimal performance and illustrate the benefits of HARQ in status updating.

We consider status updating under inexact knowledge of the battery level of an energy harvesting (EH) sensor that sends status updates about a random process to users via a cache-enabled edge node. More precisely, the control decisions are performed by relying only on the battery level knowledge captured from the last received status update packet. Upon receiving on-demand requests for fresh information from the users, the edge node uses the available information to decide whether to command the sensor to send a status update or to retrieve the most recently received measurement from the cache. We seek for the best actions of the edge node to minimize the average AoI of the served measurements, i.e., average on-demand AoI. Accounting for the partial battery knowledge, we model the problem as a partially observable Markov decision process (POMDP), and, through characterizing its key structures, develop a dynamic programming algorithm to obtain an optimal policy. Simulation results illustrate the threshold-based structure of an optimal policy and show the gains obtained by the proposed optimal POMDP-based policy compared to a request-aware greedy (myopic) policy., Comment: 12 pages

We consider a multi-source relaying system where independent sources randomly generate status update packets which are sent to the destination with the aid of a relay through unreliable links. We develop transmission scheduling policies to minimize the weighted sum average age of information (AoI) subject to transmission capacity and long-run average resource constraints. We formulate a stochastic control optimization problem and solve it using a constrained Markov decision process (CMDP) approach and a drift-plus-penalty method. The CMDP problem is solved by transforming it into an MDP problem using the Lagrangian relaxation method. We theoretically analyze the structure of optimal policies for the MDP problem and subsequently propose a structure-aware algorithm that returns a practical near-optimal policy. Using the drift-plus-penalty method, we devise a near-optimal low-complexity policy that performs the scheduling decisions dynamically. We also develop a model-free deep reinforcement learning policy for which the Lyapunov optimization theory and a dueling double deep Q-network are employed. The complexities of the proposed policies are analyzed. Simulation results are provided to assess the performance of our policies and validate the theoretical results. The results show up to 91% performance improvement compared to a baseline policy., Comment: Conditionally accepted for a publication

We consider a resource-constrained IoT network, where multiple users make on-demand requests to a cache-enabled edge node to send status updates about various random processes, each monitored by an energy harvesting sensor. The edge node serves users' requests by deciding whether to command the corresponding sensor to send a fresh status update or retrieve the most recently received measurement from the cache. Our objective is to find the best actions of the edge node to minimize the average age of information (AoI) of the received measurements upon request, i.e., average on-demand AoI, subject to per-slot transmission and energy constraints. First, we derive a Markov decision process model and propose an iterative algorithm that obtains an optimal policy. Then, we develop an asymptotically optimal low-complexity algorithm -- termed relax-then-truncate -- and prove that it is optimal as the number of sensors goes to infinity. Simulation results illustrate that the proposed relax-then-truncate approach significantly reduces the average on-demand AoI compared to a request-aware greedy (myopic) policy and also depict that it performs close to the optimal solution even for moderate numbers of sensors., Comment: 31 Pages

We consider a multi-source status update system, where each source generates status update packets according to a Poisson process which are then served according to a generally distributed service time. For this multi-source M/G/1/1 queueing model, we introduce a source-aware preemptive packet management policy and derive the moment generating functions (MGFs) of the age of information (AoI) and peak AoI of each source. According to the policy, an arriving fresh packet preempts the possible packet of the same source in the system. Furthermore, we derive the MGFs of the AoI and peak AoI for the source-agnostic preemptive and non-preemptive policy, for which only the average AoI and peak AoI have been derived earlier. Finally, we use the MGFs to derive the average AoI and peak AoI in a two-source M/G/1/1 queueing model under each policy. Numerical results show the effect of the service time distribution parameters on the average AoI: for a given service rate, when the tail of the service time distribution is sufficiently heavy, the source-agnostic preemptive policy is the best policy, whereas for a sufficiently light tailed distribution, the non-preemptive policy is the best policy. The results also highlight the importance of higher moments of the AoI., Comment: 31 pages, 9 figures

We consider a multi-source relaying system where the sources independently and randomly generate status update packets which are sent to the destination with the aid of a bufferaided relay through unreliable links. We formulate a stochastic optimization problem aiming to minimize the sum average age of information (AAoI) of sources under per-slot transmission capacity constraints and a long-run average resource constraint. To solve the problem, we recast it as a constrained Markov decision process (CMDP) problem and adopt the Lagrangian method. We analyze the structure of an optimal policy for the resulting MDP problem that possesses a switching-type structure. We propose an algorithm that obtains a stationary deterministic near-optimal policy, establishing a benchmark for the system. Simulation results show the effectiveness of our algorithm compared to benchmark algorithms., Comment: Accepted by IEEE Globecom, 20201

In this work, we consider the problem of jointly minimizing the average cost of sampling and transmitting status updates by users over a wireless channel subject to average Age of Information (AoI) constraints. Errors in the transmission may occur and a scheduling policy has to decide if the users sample a new packet or attempt for retransmission of the packet sampled previously. The cost consists of both sampling and transmission costs. The sampling of a new packet after a failure imposes an additional cost on the system. We formulate a stochastic optimization problem with the average cost in the objective under average AoI constraints. To solve this problem, we propose three scheduling policies; a) a dynamic policy, that is centralized and requires full knowledge of the state of the system, b) two stationary randomized policies that require no knowledge of the state of the system. We utilize tools from Lyapunov optimization theory in order to provide the dynamic policy, and we prove that its solution is arbitrary close to the optimal one. In order to provide the randomized policies, we model the system by utilizing Discrete Time Markov Chain (DTMC). We provide closed-form and approximated expressions for the average AoI and its distribution, for each randomized policy. Simulation results show the importance of providing the option to transmit an old packet in order to minimize the total average cost., Comment: 30 pages, submitted on a journal

We present our vision for a departure from the established way of architecting and assessing communication networks, by incorporating the semantics of information for communications and control in networked systems. We define semantics of information, not as the meaning of the messages, but as their significance, possibly within a real time constraint, relative to the purpose of the data exchange. We argue that research efforts must focus on laying the theoretical foundations of a redesign of the entire process of information generation, transmission and usage in unison by developing: advanced semantic metrics for communications and control systems; an optimal sampling theory combining signal sparsity and semantics, for real-time prediction, reconstruction and control under communication constraints and delays; semantic compressed sensing techniques for decision making and inference directly in the compressed domain; semantic-aware data generation, channel coding, feedback, multiple and random access schemes that reduce the volume of data and the energy consumption, increasing the number of supportable devices., Comment: 10 pages, 5 figures, 1 table, 1500 words

We consider a multi-source status update system in which status updates are transmitted as packets containing the measured value of the monitored process and a time stamp representing the time when the sample was generated. The packets of each source are generated according to the Poisson process and the packets are served according to an exponentially distributed service time. We assume that the received status update packets needs further processing before being used (hence, computation-intensive). This is mathematically modeled by introducing an additional server at the sink node. The sink server serves the packets according to an exponentially distributed service time. We introduce two packet management policies, namely, i) a preemptive policy and ii) a blocking policy and derive the moment generating function (MGF) of the AoI of each source under both policies. In the preemptive policy, a new arriving packet preempts any possible packet that is currently under service regardless of the packet's source index. In the blocking policy, when a server is busy at the arrival instant of a packet the arriving packet is blocked and cleared. We assume that the same preemptive/blocking policy is employed in both transmitter and sink servers. Numerical results are provided to assess the results., Comment: 7 pages, 4 figures. arXiv admin note: substantial text overlap with arXiv:2009.14439, arXiv:2001.03959

We consider a status update system consisting of two independent sources and one server in which packets of each source are generated according to the Poisson process and packets are served according to an exponentially distributed service time. We derive the moment generating function (MGF) of the age of information (AoI) for each source in the system by using the stochastic hybrid systems (SHS) under two existing source-aware packet management policies which we term self-preemptive and non-preemptive policies. In the both policies, the system (i.e., the waiting queue and the server) can contain at most two packets, one packet of each source; when the server is busy and a new packet arrives, the possible packet of the same source in the waiting queue is replaced by the fresh packet. The main difference between the policies is that in the self-preemptive policy, the packet under service is replaced upon the arrival of a new packet from the same source, whereas in the non-preemptive policy, this new arriving packet is blocked and cleared. We use the derived MGF to find the first and second moments of the AoI and show the importance of higher moments., Comment: 15 pages, 3 figures. arXiv admin note: text overlap with arXiv:2001.03959

Information freshness is crucial for time-critical IoT applications, e.g., monitoring and control systems. We consider an IoT status update system with multiple users, multiple energy harvesting sensors, and a wireless edge node. The users receive time-sensitive information about physical quantities, each measured by a sensor. Users send requests to the edge node where a cache contains the most recently received measurements from each sensor. To serve a request, the edge node either commands the sensor to send a status update or retrieves the aged measurement from the cache. We aim at finding the best actions of the edge node to minimize the age of information of the served measurements. We model this problem as a Markov decision process and develop reinforcement learning (RL) algorithms: model-based value iteration and model-free Q-learning methods. We also propose a Q-learning method for the realistic case where the edge node is informed about the sensors' battery levels only via the status updates. The case under transmission limitations is also addressed. Furthermore, properties of an optimal policy are analytically characterized. Simulation results show that an optimal policy is a threshold-based policy and that the proposed RL methods significantly reduce the average cost compared to several baselines., Comment: 33 pages, 6 figures

We consider a system where multiple sensors communicate timely information about various random processes to a sink. The sensors share orthogonal sub-channels to transmit such information in the form of status update packets. A central controller can control the sampling actions of the sensors to trade-off between the transmit power consumption and information freshness which is quantified by the Age of Information (AoI). We jointly optimize the sampling action of each sensor, the transmit power allocation, and the sub-channel assignment to minimize the average total transmit power of all sensors subject to a maximum average AoI constraint for each sensor. To solve the problem, we develop a dynamic control algorithm using the Lyapunov drift-plus-penalty method and provide optimality analysis of the algorithm. According to the Lyapunov drift-plus-penalty method, to solve the main problem we need to solve an optimization problem in each time slot which is a mixed integer non-convex optimization problem. We propose a low-complexity sub-optimal solution for this per-slot optimization problem that provides near-optimal performance and we evaluate the computational complexity of the solution. Numerical results illustrate the performance of the proposed dynamic control algorithm and the performance of the sub-optimal solution for the per-slot optimization problems versus the different parameters of the system. The results show that the proposed dynamic control algorithm achieves more than $60~\%$ saving in the average total transmit power compared to a baseline policy., Comment: 31 pages, 12 figures. arXiv admin note: text overlap with arXiv:1912.02421

Sparse signals, encountered in many wireless and signal acquisition applications, can be acquired via compressed sensing (CS) to reduce computations and transmissions, crucial for resource-limited devices, e.g., wireless sensors. Since the information signals are often continuous-valued, digital communication of compressive measurements requires quantization. In such a quantized compressed sensing (QCS) context, we address remote acquisition of a sparse source through vector quantized noisy compressive measurements. We propose a deep encoder-decoder architecture, consisting of an encoder deep neural network (DNN), a quantizer, and a decoder DNN, that realizes low-complexity vector quantization aiming at minimizing the mean-square error of the signal reconstruction for a given quantization rate. We devise a supervised learning method using stochastic gradient descent and backpropagation to train the system blocks. Strategies to overcome the vanishing gradient problem are proposed. Simulation results show that the proposed non-iterative DNN-based QCS method achieves higher rate-distortion performance with lower algorithm complexity as compared to standard QCS methods, conducive to delay-sensitive applications with large-scale signals.

We consider an IoT sensing network with multiple users, multiple energy harvesting sensors, and a wireless edge node acting as a gateway between the users and sensors. The users request for updates about the value of physical processes, each of which is measured by one sensor. The edge node has a cache storage that stores the most recently received measurements from each sensor. Upon receiving a request, the edge node can either command the corresponding sensor to send a status update, or use the data in the cache. We aim to find the best action of the edge node to minimize the average long-term cost which trade-offs between the age of information and energy consumption. We propose a practical reinforcement learning approach that finds an optimal policy without knowing the exact battery levels of the sensors., Comment: 6 pages, 4 figures

We consider the problem of minimizing the time average cost of sampling and transmitting status updates by users over a wireless channel subject to average Age of Information constraints (AoI). Errors in the transmission may occur and the scheduling algorithm has to decide if the users sample a new packet or attempt for retransmission of the packet sampled previously. The cost consists of both sampling and transmission costs. The sampling of a new packet after a failure imposes an additional cost in the system. We formulate a stochastic optimization problem with time average cost in the objective under time average AoI constraints. To solve this problem, we apply tools from Lyapunov optimization theory and develop a dynamic algorithm that takes decisions in a slot-by-slot basis. The algorithm decides if a user: a) samples a new packet, b) transmits the old one, c) remains silent. We provide optimality guarantees of the algorithm and study its performance in terms of time average cost and AoI through simulation results., Comment: Accepted in IEEE INFOCOM Workshops

We study the information freshness under three different source aware packet management policies in a status update system consisting of two independent sources and one server. The packets of each source are generated according to the Poisson process and the packets are served according to an exponentially distributed service time. We derive the average age of information (AoI) of each source using the stochastic hybrid systems (SHS) technique for each packet management policy. In Policy 1, the queue can contain at most two waiting packets at the same time (in addition to the packet under service), one packet of source 1 and one packet of source 2. When the server is busy at an arrival of a packet, the possible packet of the same source waiting in the queue (hence, source-aware) is replaced by the arrived fresh packet. In Policy 2, the system (i.e., the waiting queue and the server) can contain at most two packets, one from each source. When the server is busy at an arrival of a packet, the possible packet of the same source in the system is replaced by the fresh packet. Policy 3 is similar to Policy 2 but it does not permit preemption in service, i.e., while a packet is under service all new arrivals from the same source are blocked and cleared. Numerical results are provided to assess the fairness between sources and the sum average AoI of the proposed policies., Comment: 32 pages, 10 figures. Extended version

In this paper, we consider a system where multiple low-power sensors communicate timely information about a random process to a sink. The sensors share orthogonal subchannels to transmit such information in the form of status update packets. Freshness of the sensors' information at the sink is characterized by the Age of Information (AoI), and the sensors can control the sampling policy by deciding whether to take a sample or not. We formulate an optimization problem to minimize the time average total transmit power of sensors by jointly optimizing the sampling action of each sensor, the transmit power allocation, and the subchannel assignment under the constraints on the maximum time average AoI and maximum power of each sensor. To solve the optimization problem, we use the Lyapunov drift-plus-penalty method. Numerical results show the performance of the proposed algorithm versus the different parameters of the system., Comment: 5 pages, 5 figures

Freshness of status update packets is essential for enabling services where a destination needs the most recent measurements of various sensors. In this paper, we study the information freshness of single-server multi-source queueing models under a first-come first-served (FCFS) serving policy. In the considered model, each source independently generates status update packets according to a Poisson process. The information freshness of the status updates of each source is evaluated by the average age of information (AoI). We derive an exact expression for the average AoI for the case with exponentially distributed service time, i.e., for a multi-source M/M/1 queueing model. Moreover, we derive three approximate expressions for the average AoI for a multi-source M/G/1 queueing model having a general service time distribution. Simulation results are provided to validate the derived exact average AoI expression, to assess the tightness of the proposed approximations, and to demonstrate the AoI behavior for different system parameters., Comment: 32 pages, 15 figures

We consider a communication system in which status updates arrive at a source node, and should be transmitted through a network to the intended destination node. The status updates are samples of a random process under observation, transmitted as packets, which also contain the time stamp to identify when the sample was generated. The age of the information available to the destination node is the time elapsed since the last received update was generated. In this paper, we model the source-destination link using queuing theory, and we assume that the time it takes to successfully transmit a packet to the destination is an exponentially distributed service time. We analyze the age of information in the case that the source node has the capability to manage the arriving samples, possibly discarding packets in order to avoid wasting network resources with the transmission of stale information. In addition to characterizing the average age, we propose a new metric, called peak age, which provides information about the maximum value of the age, achieved immediately before receiving an update., Comment: 20 pages, 7 figures