Neural network-based wireless receivers, also known as neural receivers , have demonstrated superior performance over traditional receivers, but come with greater computational complexity. The need to use these networks on energy-conscious edge devices is increasing, necessitating energy-efficient and adaptive neural receivers to function well under varying channel conditions. Transitioning static neural receivers to dynamic models using the concepts of Dynamic Neural Networks (DyNN) could reduce runtime computational complexity and energy consumption. This could be achieved by adapting the network architecture during run-time and exploit the varying channel conditions in a mobile communication system to reduce the computational complexity. This work introduces MEAN, a novel hard-gated Mixture-of-Experts (MOE) based neural receiver architecture. The main idea behind MEAN is to use several smaller Signal-to-Noise-Ratio (SNR) expert networks are used during run-time to create a network that selects the correct expert for the current data input to reduce complexity. The paper consists of the following key contributions. (1) MEAN architecture based on [ 25 ]: A hard-gated MoE model that dynamically selects the most suitable expert for each input dynamically based on current channel conditions. (2) Comprehensive system-level performance analysis of MEAN across different code rates and modulation schemes, including validation of expert selection during inference. (3) Loss function optimization for the gating network to promote the activation of specific experts in designated noise regions. This enhancement reduces gating network complexity while improving expert selection accuracy over previous MEAN implementations. (4) Hardware analysis of a gate-level implementation of MEAN synthesized in 22nm FD-SOI technology. The design is generated using a custom High-Level Synthesis (HLS) framework, providing detailed power and area evaluations. The proposed MEAN architecture for a Single Input Multiple Output (SIMO) wireless system achieves a 37.11% reduction in total power consumption with only an 11.04% increase in area, while maintaining accuracy comparable to that of a static neural network.

Achieving low delay and high data rate communication in 5 G requires preventing persistent queue backlogs at path bottlenecks. A key indicator of queue buildup is an increase in one-way delay (OWD) of packets. However, accurately measuring OWD is challenging due to synchronization issues between the sender and receiver. Moreover, using OWD to infer the onset of queueing can be inefficient. For instance, congestion control mechanisms that rely on latency as a congestion signal may result in cellular network underutilization, see, e.g., [1]. Thus, to develop more efficient rate control approaches, it is crucial not only to estimate OWD accurately but also to perform a more fine-grained analysis that provides detailed insights into the queue status. This paper presents a framework for modeling OWD in 5 G edge networks, where the wireless link is typically the path bottleneck. We demonstrate that OWD is dominated by queueing delay (QD), particularly at high data rates. We analyze QD by decomposing it into two main components: aggregation delay and channel access time. Aggregation delay can be accurately measured at the user equipment (UE), and channel access time can be estimated based on network conditions. Our key contribution shows that QD, while challenging to measure directly at the base station (BS), can be closely approximated by combining these components. We validate this model through NS3 simulations under varying network loads and single- and multi-UE scenarios. This framework provides a foundation for developing advanced mechanisms for congestion control and bottleneck detection in 5G networks.

Accurate reliability modeling for ultra-reliable low latency communication (URLLC) and hyper-reliable low latency communication (HRLLC) networks is challenging due to the complex interactions between network layers required to meet stringent requirements. In this paper, we propose such a model. We consider the acknowledged mode of the radio link control (RLC) layer, utilizing separate buffers for transmissions and retransmissions, along with the behavior of physical channels. Our approach leverages the effective capacity (EC) framework, which quantifies the maximum constant arrival rate a time-varying wireless channel can support while meeting statistical quality of service (QoS) constraints. We derive a reliability model that incorporates delay violations, various latency components, and multiple transmission attempts. Our method identifies optimal operating conditions that satisfy URLLCIHRLLC constraints while maintaining near-optimal EC, ensuring the system can handle peak traffic with a guaranteed QoS. Our model reveals critical trade-offs between EC and reliability across various use cases, providing guidance for URLLCIHRLLC network design for service providers and system designers.

Dynamic Neural Networks (DyNN) adapt their structure during runtime for improved performance and lower power consumption. DyNNs benefit neural network-based wireless receivers, or in short neural receivers, that adapt their performance under varying channel conditions. However, DyNN architectures are not yet explored sufficiently in the literature as this would require a framework for simultaneously evaluating system-level performance and accurate hardware Power-Performance-Area (PPA) under different dynamic scenarios. This paper presents two main contributions: (1) An automated framework that bridges a system-level performance evaluation model of a wireless system and a High-Level Synthesis (HLS) tool for agile Design-Space Exploration (DSE) of DyNNs. (2) A novel DyNN architecture for neural receivers in wireless systems that skips a variable number of layers according to varying channel conditions. Our proposed neural receiver architecture with layer-skipping for a Single Input Multi Output (SIMO) wireless system when implemented in 22 nm FD-SOI technology shows power consumption savings of up to 59.2% at the cost of 200% increase in area compared to a static network.

Dynamic Neural Networks (DyNN) adapt its network architecture at run-time compared to static neural networks. DyNNs benefit in wireless receivers based on neural networks (or neural receivers), that need to adapt their performance under varying channel conditions. Mixture-of-Experts (MoE) is one efficient way to realize a DyNN in neural receivers in which several smaller expert networks are dynamically combined in the run-time to create a dedicated network according to the channel requirements. This paper presents a novel hard-gated (also known as sparsely-gated) MoE-based neural receiver architecture called MEAN for Single Input Multiple Output (SIMO)-based wireless communication systems. Our proposed MEAN architecture for a SIMO wireless system shows a reduction of up to 50% in the number of active layers during runtime.

Recent technological advances allow modification and fine-tuning of the wireless network characteristics. By modifying wireless properties such as transmission timeslots or frequencies, the wireless links quality can be optimized in order to reach optimal communication at the network level. In this paper, we approach the wireless network optimization problem as a distributed constraint optimization problem. As an inherently distributed task, the number of constraints, variables, and their domain sizes can be very large. Therefore, incomplete and local-search solutions such as the Distributed Stochastic Algorithm (DSA) are best suited to solve this class of problems. In this work, we study the wireless network optimization procedure of such solvers considering wireless messaging cost. Furthermore, we introduce Weighted-DSA a stochastic algorithm for wireless optimization. By reducing the search-space of the variables and re-exploring periodically, results show that this algorithm is able to reach optimal solution quality under minimal messgeing costs.

Pervasiveness of wireless networks drives the heterogeneity and density of devices in a vast diversity of environments. To achieve high reliability and low energy consumption while enabling pervasiveness is inherently a resource allocation problem. In low-rate wireless personal area networks, multi-frequency time-division multiple access methods are identified as compelling solutions to resource allocation via scheduling transmissions in time and frequency. This work presents the TREE (TRaffic-aware Energy Efficient) algorithm, an adaptive and distributed scheduling algorithm, designed to provide high reliability in terms of packet reception ratio while optimizing the energy consumption of each device. This algorithm schedules communications according to the packets in the queue and short-memory performance. Decisions are made locally, and low-interference scheduling emerges at the network level. TREE is an adaptive threshold-based model that allocates more network resources (e.g., timeslots) when the communication queue size crosses a threshold and frees resources if the resource was underutilized. Implemented over IEEE-802.15.4-TSCH and extensively tested in simulation and on real deployments up to 81 devices, the algorithm is compared to MSF, Alice, and Orchestra, the state of the art in time-slotted channel hopping scheduling. Results highlight a high reliability regarding packet reception ratio and a lower energy consumption compared to the state of the art.

The 3rd Generation Partnership Project (3GPP) has recently established in Rel. 14 a network-assisted resource allocation scheme for vehicular broadcast communications. Such novel paradigm is known as vehicle-to-vehicle (V2V) \textit{mode-3} and consists in eNodeBs engaging only in the distribution of sidelink subchannels among vehicles in coverage. Thereupon, without further intervention of the former, vehicles will broadcast their respective signals directly to their counterparts. Because the allotment of subchannels takes place intermittently to reduce signaling, it must primarily be conflict-free in order not to jeopardize the reception of signals. We have identified four pivotal types of allocation requirements that must be guaranteed: one quality of service (QoS) requirement and three conflict conditions which must be precluded in order to preserve reception reliability. The underlying problem is formulated as a maximization of the system sum-capacity with four types of constraints that must be enforced. In addition, we propose a three-stage suboptimal approach that is cast as multiple independent knapsack problems (MIKPs). We compare the two approaches through simulations and show that the latter formulation can attain acceptable performance at lesser complexity.

Source localization is of pivotal importance in several areas such as wireless sensor networks and Internet of Things (IoT), where the location information can be used for a variety of purposes, e.g. surveillance, monitoring, tracking, etc. Time Difference of Arrival (TDOA) is one of the well- known localization approaches where the source broadcasts a signal and a number of receivers record the arriving time of the transmitted signal. By means of computing the time difference from various receivers, the source location can be estimated. On the other hand, in the recent few years novel optimization algorithms have appeared in the literature for (i) processing big data and for (ii) training deep neural networks. Most of these techniques are enhanced variants of the classical stochastic gradient descent (SGD) but with additional features that promote faster convergence. In this paper, we compare the performance of the classical SGD with the novel techniques mentioned above. In addition, we propose an optimization procedure called RMSProp+AF, which is based on RMSProp algorithm but with the advantage of incorporating adaptation of the decaying factor. We show through simulations that all of these techniques--which are commonly used in the machine learning domain- -can also be successfully applied to signal processing problems and are capable of attaining improved convergence and stability. Finally, it is also shown through simulations that the proposed method can outperform other competing approaches as both its convergence and stability are superior.

In Release 14, the 3rd Generation Partnership Project (3GPP) introduced Cellular Vehicle-to-Everything (C-V2X) mode-4 as a novel disruptive technology to support sidelink vehicular communications in out-of-coverage scenarios. C-V2X mode-4 has been engineered to operate in a distributed manner, wherein vehicles autonomously monitor the received power across sidelink subchannels before selecting one for utilization. By means of such an strategy, vehicles attempt to (i) discover and (ii) reserve subchannels with low interference that may have the potential to maximize the reception likelihood of their own broadcasted safety messages. However, due to dynamicity of the vehicular environment, the subchannels optimality may fluctuate rapidly over time. As a consequence, vehicles are required to make a new selection every few hundreds of milliseconds. In consonance with 3GPP, the subchannel selection phase relies on the linear average of the perceived power intensities on each of the subchannels during a monitoring window. However, in this paper we propose a nonlinear power averaging phase, where the most up-to-date measurements are assigned higher priority via exponential weighting. We show through simulations that the overall system performance can be leveraged in both urban and freeway scenarios. Furthermore, the linear averaging can be considered as a special case of the exponentially-weighted moving average, ensuring backward compatibility with the standardized method. Finally, the 3GPP mode-4 scheduling approach is described in detail.

In Release 14, 3GPP completed a first version of cellular vehicle-to-everything (C-V2X) communications wherein two modalities were introduced. One of these schemes, known as \textit{mode-3}, requires support from eNodeBs in order to realize subchannel scheduling. This paper discusses a graph theoretical approach for semi- persistent scheduling (SPS) in \textit{mode-3} harnessing a sensing mechanism whereby vehicles can monitor signal-to-interference-plus-noise ratio (SINR) levels across sidelink subchannels. eNodeBs request such measurements from vehicles and utilize them to accomplish suitable subchannel assignments. However, since SINR values--herein also referred to as side information--span a wide range, quantization is required. We conclude that 3 bits per vehicle every 100 ms can provide sufficient granularity to maintain appropriate performance without severe degradation. Furthermore, the proposed algorithm is compared against pseudo-random and greedy SPS algorithms.

Channel congestion is one of the most critical issues in IEEE 802.11p-based vehicular communications as it leads to the unreliability of safety applications. As a countermeasure, many Decentralized Congestion Control (DCC) algorithms have been proposed. One of the most prominent DCC algorithms is the message-rate based LIMERIC. Recently, algorithms have also been proposed to support higher vehicle density (better scalability). One of such algorithms is the combined message-rate and data-rate based congestion control algorithm (MD-DCC). MD-DCC can support around 2.7 times higher vehicular density than LIMERIC. However, if LIMERIC has been deployed, can MD-DCC be introduced and coexist well with LIMERIC. The objective of this paper is to investigate how MD-DCC coexists with LIMERIC. Given a scenario where vehicles may use either LIMERIC or MD-DCC, we study the impact of coexistence on channel load, fairness and reliability of vehicles at different densities. Simulation studies show that there is no significant degradation of reliability both for LIMERIC and MD-DCC at different densities. On the contrary, coexistence can improve the reliability of LIMERIC vehicles. Furthermore, MD-DCC can support vehicles at large densities even when it coexists with LIMERIC retaining its scalability. However, fair allocation of resources is not guaranteed when LIMERIC and MD-DCC coexist.

Vehicle-to-Everything (V2X) communications are promising solutions to optimize the traffic efficiency and safety. However, the communication channel may get congested under a high vehicular density and furthermore degrade the reliability of safety applications. In order to tackle this critical issue, various Decentralized Congestion Control (DCC) algorithms have been proposed. Each DCC adjusts different transmission parameters (message rate, data rate, transmission power, etc.) to control the channel loads. In this paper, we conducted a detailed study of the DCC proposed by the Society of Automotive Engineers International (SAE-DCC), then compared it with Packet count based Data-Rate DCC (PDR-DCC) and message rate based DCC LIMERIC. In our study, we verified the effectiveness of SAE- DCC and compared it to LIMERIC and PDR-DCC using selected evaluation metrics (CBR, position error, T-window reliability and awareness range). According to the simulation results in ns-3, in terms of position error, PDR-DCC demonstrates the best performance within 200-meter range, while SAE-DCC and LIMERIC outperform PDR-DCC when the distance increases. In view of the T-window reliability, PDR-DCC shows a clear advantage under dense and extreme traffic densities.

Conversely to mainstream cellular networks where uplink / downlink data traffic is centrally managed by eN-odeBs, in vehicle-to-vehicle (V2V) broadcast communications mode-3 eNodeBs engage solely in subchannel assignment but ultimately do not intervene in data traffic control. Accordingly, vehicles communicate directly with their counterparts utilizing the allotted subchannels. Due to its loosely controlled one-to-all nature, V2V mode-3 is advantageous for time-critical applications. Nevertheless, it is imperative that the assignment of subchannels is accomplished without conflicts while at the same time satisfying quality of service (QoS) requirements. To the best of our knowledge, there exists no unified framework for V2V mode-3 that contemplates both prevention of allocation conflicts and fulfillment of QoS. Thus, four types of conditions that are of forceful character for attaining QoS-aware conflict-free allocations have been identified: (i) assure differentiated QoS per vehicle, (ii) preclude intra-cluster subframe conflicts, (iii) secure minimal time dispersion of allotted subchannels and (iv) forestall one-hop inter-cluster subchannel conflicts. Such conditions have been systematized and merged in an holistic manner allowing non-complex manipulation to perform subchannel allocation optimization. In addition, we propose a surrogate relaxation of the problem that does not affect optimality provided that certain requisites are satisfied.

In this paper we present a graph-based resource allocation scheme for sidelink broadcast V2V communications. Harnessing available information on geographical position of vehicles and spectrum resources utilization, eNodeBs are capable of allotting the same set of sidelink resources to different vehicles distributed among several communications clusters. Within a communications cluster, it is crucial to prevent time-domain allocation conflicts since vehicles cannot transmit and receive simultaneously, i.e., they must transmit in orthogonal time resources. In this research, we present a solution based on a bipartite graph, where vehicles and spectrum resources are represented by vertices whereas the edges represent the achievable rate in each resource based on the SINR that each vehicle perceives. The aforementioned time orthogonality constraint can be approached by aggregating conflicting vertices into macro-vertices which, in addition, reduces the search complexity. We show mathematically and through simulations that the proposed approach yields an optimal solution. In addition, we provide simulations showing that the proposed method outperforms other competing approaches, specially in scenarios with high vehicular density.

In this paper we present a graph-based resource allocation scheme for sidelink broadcast vehicle-to-vehicle (V2V) communications. Harnessing available information on the geographical position of vehicles and spectrum resources utilization, eNodeBs are capable of allotting the same set of sidelink resources to several different vehicles in order for them to broadcast their signals. Hence, vehicles sharing the same resources would ideally be in different communications clusters for the interference level - generated due to resource repurposing - to be maintained under control. Within a communications cluster, it is crucial that vehicles transmit in orthogonal time resources to prevent conflicts as vehicles - with half-duplex radio interfaces - cannot transmit and receive simultaneously. In this research, we have envisaged a solution based on a bipartite graph, where vehicles and spectrum resources are represented by vertices whereas the edges represent the achievable rate in each resource based on the signal-to-interference-plus-noise ratio (SINR) that vehicles perceive. The aforementioned constraint on time orthogonality of allocated resources can be approached by aggregating conflicting vertices into macro-vertices which, in addition, narrows the search space yielding a solution with computational complexity equivalent to the conventional graph matching problem. We show mathematically and through simulations that the proposed approach yields an optimal solution. In addition, we provide simulations showing that the proposed method outperforms other competing approaches, specially in scenarios with high vehicular density.

In this poster we present a graph-based hierarchical subchannel allocation scheme for V2V sidelink communications in Mode-3. Under this scheme, the eNodeB allocates subchannels for in-coverage vehicles. Then, vehicles will broadcast directly without the eNodeB intervening in the process. Therefore, in each communications cluster, it will become crucial to prevent allocation conflicts in time domain since vehicles will not be able to transmit and receive simultaneously. We present a solution where the time-domain requirement can be enforced through vertex aggregation. Additionally, allocation of subchannels is performed sequentially from the most to the least allocation-constrained cluster. We show through simulations that the proposed approach attains near-optimality.

Vehicle-to-others (V2X) communication systems intend to increase safety and efficiency of our transportation networks. However, wireless communication imperfections such as missed messages due to collisions and fading in the wireless channel, may affect safety application reliability and lead to risky situations. Thus metrics are required to evaluate the impact of communication inadequacies on the safety applications. In this paper we perform analyses of various existing safety application reliability metrics and conclude that they do not reflect safety application risk and vulnerability of individual nodes effectively. We propose a new metric called Effective Risk Factor (ERF), which quantifies the risk at a node for each link, to identify dangers due to poor awareness of their neighbors. The ERF evaluation considers links of its neighbors, thus detecting risky situations over existing neighbor links on runtime making the ERF assessment realistic. The ERF metric is evaluated and compared with other reliability metrics for a stationary vehicle warning application in a simulated highway scenario. The results show that the ERF evaluation performed at each node on runtime is able to capture a fine time scale fluctuations in the risk experienced by an application precisely. The ERF also enables prediction of higher risk situations. The results also demonstrate that the ERF captures application risk experienced by nodes effectively compared to other reliability metrics.

Vehicle platooning using reliable wireless communication between the member vehicles is a promising method to increase road capacity, lower fuel consumption, and improve safety and driver comfort. IEEE 802.11p is a key communication technology in Vehicular Ad-hoc Networks (VANETs) for Intelligent Transport System (ITS) applications. The broadcasted awareness messages in the wireless communication channel may be used for platooning control, but there are reliability concerns especially in a highly congested network in rush hours. Keeping short inter-vehicle distance requires timeliness and reliability of the underlying exchange of control data in the communication channel. In this paper, we present a novel analytical model of communication reliability between platoon members in a channel shared with other vehicles using ITS. We use a discrete time M/G/1 queue model for occurrence of messages in Poisson distribution, and hence for the estimation of the Packet Reception Probability between the platoon members. We have evaluated the model for different vehicular densities, message rate and data rate with network simulations. Our results show that the Packet Reception Rate (PRR) from the model closely matches the simulation results. Based on PRR we estimate probability of missing packet in consecutive period of transmission. This model opens new opportunities to improve and evaluate the platooning control algorithms.

In a home environment, a Smart Grid Home Area Network (SG-HAN) platform facilitates collection and delivery of power consumption information for load profiling and informed decisions on energy management. However, one of the main challenges in HAN is the overcrowded unlicensed 2.4-GHz ISM frequency band, occupied by several types of radio technologies such as ZigBee, Bluetooth, and WiFi. It is crucial that those technologies coexist peacefully to allow each user of the radio technology to fulfill their communication goals. In this paper, we present a potential coexistence scenario in SG-HAN for homogeneous and heterogeneous wireless technologies. The coexistence impact on SG-HAN performance is then modeled and analyzed. The numerical results show significant performance degradation due to the interference for devices in close proximity with the interfering sources in a spectrum sharing environment where in the worst case scenario, SG-HAN communication is almost impossible.

Indoor downlink communication contributes to a large part of the data traffic generated in today's world. Enormous high data rate supporting devices are entering todays market. They demand for high rate wireless indoor coverage for their uninterrupted service. The main challenge lies in working with the existing wireless technologies while providing a future proof centralized optical fiber indoor backhaul for efficient indoor coverage. Fiber to the room paradigm is gaining a lot of attention in this regard. For supporting high data rates indoor, multiuser MIMO (MU-MIMO) is definitely a prominent choice. While quality of service serves as the most attractive feature that should be ensured among the mobile terminals (MTs). In this work we propose a throughput fair successive signal to leakage and noise ratio (SLNR) precoding algorithm for such a fiber-wireless (Fi-Wi) MU-MIMO indoor. The network capacity and individual MT data rate for our proposed scheme are compared against the greedy SLNR scheme and a random selection based SLNR precoding scheme. The Jains fairness index value for our scheme is shown to achieve maximal fairness.

A personal network (PN) is a person-centric network, which creates a distributed environment of a person and provides access to personal resources and services regardless the location of the person. An extra functionality added to PNs can enable different persons to cooperate and to form a group-oriented network called a Federation of Personal Networks (Fednet). Fednet is a temporal, ad hoc, opportunity/purpose driven, secure group-oriented network where the users may be the producers and consumers of the services/content/resources. In this paper, we introduce Fednets and their architecture, explain the lifecycle and the access control process.

A Personal Network (PN) is a new type of an overlay network consisting of all personal devices belonging to a user, be they remote or local. Such continuous and seamless connectivity of all personal devices belonging to a user enables the development of new applications and improved services. A Personal Network is an intelligent and user-centric network that assists the user in an unobtrusive way. In this paper we investigate means of securing communication between geographically distributed Personal Network clusters. Clusters are ad-hoc networks of co-located personal devices. Using Virtual Private Network (VPN) technology enables the creation of secure tunnels between gateways of different clusters, making it possible to transfer all types of intra-cluster traffic securely, over the insecure public network. The ad-hoc nature of clusters when coupled with the mobile and resource constrained nature of many personal devices makes enforcing security a challenging task. We investigate the suitability of existing VPN technologies to secure inter-cluster communication. Our contribution is in identifying aspects of existing solutions that render them unsuitable in their current form for constrained personal devices. We propose a modified framework based on IPSec and KINK that better satisfies the requirements of personal devices by reducing the cost of security.

The provision of one-to-many multimedia services to mobile users over scarce wireless resources has to be performed in a cost-efficient way. In addition to various systems optimized for this type of service, such as DVB-H, UMTS was recently enhanced for support of multimedia broadcast and multicast services (MBMS) over the existing infrastructure. Since MBMS provision will reduce the UMTS capacity for its essential point-to-point services, the integration of UMTS and DVB-H may be beneficial. In this paper, we address the integration of DVB-H at the UMTS radio access level. We present an interworking architecture and discuss necessary modifications to UMTS.

In this paper we evaluate our previously proposed security architecture for personal networks (PNs). Personal network is a new concept utilizing pervasive and distributed computing to meet the needs of the user. We aim to secure personal networks with lightweight security mechanisms that are suitable for resource constrained devices yet robust enough for self organization and secure communication. In order to study the behavior of our proposed security mechanisms we developed a simulation environment in NS-2. The simulations are used to evaluate the overhead of our mechanisms and to understand the effects of key parameters. The results show that our mechanisms have low delay and energy requirements and are feasible for the heterogeneous devices we envision in our PN.

Personal networks (PNs) are future communication systems that combine wireless and infrastructure based networks to provide users a variety of services anywhere and anytime. PNs introduce new design challenges due to the heterogeneity of the involved technologies, the need for self-organization, the dynamics of the PN composition, the application-driven nature, the co-operation with infrastructure-based networks, and the security hazards. This paper discusses the challenges of security, service discovery and QoS provisioning in designing self-organized PNs and combines them all into an integrated architectural framework

Personal network (PN) is a new concept utilizing pervasive computing to meet the needs of the user. As PNs edge closer towards reality, security becomes an important concern since any vulnerability in the system will limit its practical use. In this paper we introduce a security architecture designed for PNs. Our aim is to use secure but lightweight mechanisms suitable for resource constrained devices and wireless communication. We support pair-wise keys for secure cluster formation and use group keys for securing intra-cluster communication. In order to analyze the performance of our proposed mechanisms, we carry out simulations using ns-2. The results show that our mechanisms have a low overhead in terms of delay and energy consumption

Personal networks (PN) is a new concept related to pervasive computing with a strong user-focus view. The key to a successful PN realization is a general network architecture that is capable of bridging different current and future technologies and offers a homogeneous and clear view to the end-user. In this paper, we focus on forming a PN by connecting remote personal devices using infrastructure-based IP networks, including 3G networks and WLAN hotspots. One way is to upgrade the current access networks with new functionality to support PNs. Since many devices in PNs are mobile and battery powered, this may help them to achieve a faster service and to save energy. However, to deploy such functionality is not easy and may hamper the adoption of PNs altogether. Therefore, in this paper we study three possible inter-cluster communication architectures that can use current IP networks. To discern the above proposal we also give a detailed picture of PN network architecture supported by infrastructure. We believe that this detailed discussion will help the success of PNs

The formal analysis of a data-transfer protocol based on the novel idea of periodic state exchange has been presented by Gouda et al. (see IEEE Trans. Commun., vol.43, no.9, p.2475-84, 1995). In certain environments, such as the Internet, the requirements for the correctness of the protocol may impose impractical constraints. We present an alternative specification of the periodic state-exchange protocol based on the discovery that the protocol can be treated as a special sliding-window protocol. Although our protocol behaves better in an Internet-like environment it has other shortcomings with respect to the original. The comparison of the alternatives reveals the trade-offs in their design.

The authors comment on a class of minimum-latency transport protocols that have been analyzed by Shankar and Lee (see ibid., vol.3, no.3, p.255, 1995). The protocols use unique incarnation numbers and caching schemes to reduce the latency of connection setup whenever possible. They discuss three modifications to the protocol. (1) A modification to the opening procedure which eliminates some constraints for the correctness of the protocol. (2) A modification which allows data messages in the opening state of the client to be sent. This reduces the latency in some situations for the price of stricter constraints for correctness. (3) An alternate way of closing connections. Apart from these modifications, they also show that the proofs can be refined to get somewhat less restrictive constraints for the correctness of the protocol.