The domain adaptation issues in semantic communications become critical when transmitter and receiver operate across different multiple domains or when input data during inference have different distributional characteristics than the data used to train semantic encoders and decoders. In this paper, we introduce the Multidomain Adaptive Deep Semantic Communication (MA-DeepSC) framework, designed to enhance semantic communications across multiple domains. Our framework consists of two core components: the Multidomain Adaptive Semantic Coding Network (MASCN), inherently designed to adapt semantic encoding and decoding across multiple domains, and the multidomain data adaptation network (MDAN), which transforms actual observable data into the data on which the system was initially trained, thus obviating the need for retraining the existing pre-trained semantic coding network. We validate our approach through experiments on digit datasets and CelebA, observing significant outperformance over existing techniques. In addition, we analyze the strategic benefits and drawbacks of both MASC and MDAN, assessing their applicability under various scenarios. The source code for MA-DeepSC is available at https://github.com/wongdongwook/JSAC_MA-DeepSC.

Federated learning (FL) is a framework for realizing distributed machine learning in an environment where training samples are distributed to each device. Recently, FL has employed over-the-air computation enabling all devices to transmit learning model updates simultaneously. This work proposes a communication-efficient sparse one-bit analog aggregation (SOBAA) method, incorporating new power control, layer-wise scaled one-bit quantization, layer-wise sparsification, and an error-feedback mechanism. We derive a tight upper bound of the expected convergence rate of the proposed SOBAA as a closed-form expression. From this expression, we explicitly identify the relationship between the convergence rate and compression and aggregation errors. Based on the theoretical convergence analysis, we formulate a joint optimization problem of the compression ratio and power control to minimize compression and aggregation errors, leading to the fastest convergence. In each communication round, the optimization problem is decomposed, and solved in a computationally efficient and feasible way. From this solution, we characterize the trade-off between learning performance and communication cost. Through extensive experiments on well-known MNIST and CIFAR-10 datasets, we confirm that the proposed method provides an enhanced trade-off performance between test accuracy and communication costs and a faster convergence rate than the other state-of-the-art methods. In addition, it is proven that the proposed method is more effective for more complex datasets and learning models.

This paper deals with a method to measure the quality of Giga Internet service. The existing method needs both remote servers and a client application when measuring Internet quality such as throughput(speed), delay and loss. On the other hand, the proposed method only needs a single-sided application which is installed in a personal computer(PC). The key technique is to use traffic loopback algorithm which enables the sent packets at PC to return to it via gateway equipment like L3 switch. Using this method, network providers could not only measure the quality of Giga Internet service in access network more accurately than the existing server-client method, but also reduce the cost of new measurement servers for Giga Internet service.

Accurate network topology information can play very important role in a network management system. Discovering network topology is a prerequisite for many critical network management tasks including reactive and proactive resource management, fault management, and root-cause analysis. Unfortunately, the equipment managed in an NMS includes switches as well as routers, so a network management system must discover physical links/connections between a router and a switch and between a switch and a switch. Most previous methods of discovering network topology via SNMP use ipRouteTable MIB, which has big routing information. Many network operators repeat the saying, ldquoDo not burn your house to rid it of the mouse.rdquo reffering to the concern of overloading the capacity of the router. Most previous methods were not concerned with this simple maxim, and in fact, there is no need to worry about managing overhead in a small test-bed. In this paper, the problems of previous works using ipRouteTable are presented. We focus on large OSPF backbone networks that have hundreds of routers and switches, and discover physical topology with minimal overhead. Our approach finds adjacencies between router and router / router and switch using ospfNbrIpAddr MIB which is much smaller than ipRouteTable. The experimental results clearly validate that our approach is an efficient algorithm for discovering network topology in time that is mathematically quadratic to the number of routers, as compared to cubic when ipRouteTable is used.